Follow the instruction due in 2 days

-^ THE TRANSFER OF INTERNAL

KNOWLEDGE AND BEST PRACTICE

IF ONLY WE

WHAT WE

KNOW CARLA O’DELL

C. JACKSON GRAYSON, JR. WITH NILLY ESSAIDES

A___J

U.S. $30.00 Can. $44.50

While companies search the world over to

benchmark best practices, vast treasure

troves of knowledge and know-how remain

hidden right under their noses: in the minds of their own

employees, in the often unique structure of their opera¬

tions, and in the written history of their organizations.

Now, acclaimed productivity and quality experts Carla

O’Dell and Jack Grayson explain for the first time how

applying the ideas of Knowledge Management can help

employers identify their own internal best practices and

share this intellectual capital throughout their organiza¬

tions.

Knowledge Management (KM) is a conscious strat¬

egy of getting the right information to the right people at

the right time so they can take action and create value.

Basing KM on three major studies of best practices at

one hundred companies, the authors demonstrate how

managers can utilize a visual process model to actually

transfer best practices from one business unit of the

organization to another. Rich with case studies, concrete

examples, and revealing anecdotes from companies

including Texas Instruments, Amoco, Buckman, Chevron,

Sequent Computer, the World Bank, and USAA, this

valuable guide reveals how knowledge treasure chests

can be unlocked to reduce product development cycle

time, implement more cost-efficient operations, or create

a loyal customer base. Finally, O’Dell and Grayson pre¬

sent three “value propositions” built around customers,

products, and operations that could result in staggering

payoffs as they did at the companies cited above.

No amount of knowledge or insight can keep a com¬

pany ahead if it is not properly distributed where it’s

needed. Entirely accessible and immensely readable, If

Only We Knew What We Know is a much-needed com¬

panion for business leaders everywhere.

e y <?

LIBRARY

O.

ilROU,

IF ONLY WE

KNEWwhat WE KNOW

The Transfer of Internal Knowledge

and Best Practice

CARLA O’DELL

C JACKSON GRAYSON, JR.

with Nilly Essaides

THE FREE PRESS

New York

/p THE FREE PRESS

A Division of Simon & Schuster Inc.

1230 Avenue of the Americas

New York, NY 10020

Copyright © 1998 by Carla O’Dell and C. Jackson Grayson, Jr.

All rights reserved,

including the right of reproduction

in whole or in part in any form.

THE FREE PRESS and colophon are trademarks

of Simon & Schuster Inc.

Designed by Carla Bolte

Manufactured in the United States of America

10 987654321

Library of Congress Cataloging-in-Publication Data

O’Dell, Carla S., date.

If only we knew what we know : the transfer of internal knowlege and best practice /

Carla O’ Dell, C. Jackson Grayson, Jr., with Nilly Essaides.

p. cm.

Includes bibliographical references and index.

1. Organizational learning. 2. Knowledge management.

3. Communication in organizations. 4. Benchmarking (Management).

I. Grayson, C. Jackson (Charles Jackson), date. II. Essaides, Nilly. III. Title.

HD58.82.03 1998 98-18441 CIP

658.3T24—dc21

ISBN 0-684-84474-5 C = /O . f?

0^T 2 2 1999 3 0-°°

To APQC’s International Benchmarking Clearinghouse Members who al¬

ways gave so generously of their time and best practices to us and to each

other; to the pathbreaking companies and people who shared with us their

knowledge, experiences, and insights; to the superb staff of APQC who

share a dedication to excellence and the mission; and finally, to our families

for their love and understanding.

i

CONTENTS

Preface ix

Acknowledgments xvii

PART ONE: A FRAMEWORK FOR

INTERNAL KNOWLEDGE TRANSFER

1. Definitions of Knowledge and Knowledge Management .... 3

2. KM in Action—The Transfer of Best Practices.11

3. The Barriers to Internal Transfer.16

4. A Model for Best Practice Transfer.21

PART TWO: THE THREE VALUE PROPOSITIONS

5. Find Your Value Proposition.31

6. Customer Intimacy.38

7. Product-tO'Market Excellence.47

8. Achieving Operational Excellence.59

PART THREE: THE FOUR ENABLERS OF TRANSFER

9. Culture, the Unseen Hand.71

10. Using Information Technology to Support

Knowledge Transfer.85

11. Creating the Knowledge Infrastructure.107

12. Measuring the Impact of Transfer.126

PART FOUR: REPORTS FROM THE FRONT LINES:

PIONEER CASE STUDIES

13. The View from the Top.141

14. Buckman Laboratories: Empowered by K’Netix®.144

15. TI’s Best Practice Sharing Engine.152

16. Becoming a “Knowledge Bank”.160

17. Sequent Computer’s Knowledge “Slingshot”.170

PART FIVE: THE FOUR-PHASE PROCESS:

OR “WHAT DO I DO ON MONDAY MORNING?”

18. Plan, Assess, and Prepare: Phase 1.183

19. Designing the Transfer Project: Phase 2.191

20. Implementation: Phase 3.199

21. Transition and Scale-Up: Phase 4.208

PART SIX: CONCLUSION

22. Enduring Principles.223

Appendix

The Knowledge Management Assessment Tool (KMAT)®. . 227

References 231

Index 233

PREFACE

IfTI only knevu what TI knows.

—Jerry J unkins, ex-CEO of Texas Instruments

I wish we knew what we know at HP.

—Lew Platt, Hewlett-Packard

Arthur Clarke once observed that cave dwellers froze to death on

beds of coal. Coal was right under them, but they couldn’t see it,

mine it, or use it. This is one clear case of what you don’t know can and

will hurt you. And it’s happening all over again in the 1990s. Except

that this time around, it’s not beds of coal but beds of “knowledge”—

hidden reservoirs of intelligence that exist in almost every organiza¬

tion, relatively untapped and unmined.

A few organizations, however, are not making this mistake. They’re

learning how to mine knowledge with machinery called “knowledge

management” (KM). They are tapping into this hidden asset, captur¬

ing it, organizing it, transferring it, and using it to create customer

value, operational excellence, and product innovation—-all the while

increasing profits and effectiveness.

From Amoco to Xerox, Buckman to Sequent, companies are rally¬

ing their workers around the “sharing what we know” battle cry. There

already are clear signs that effective transfer of knowledge pays off—big

time.

IX

X • Preface

HOW WE CAME TO KNOW WHAT WE KNOW

The American Productivity & Quality Center (APQC) has been ob¬

serving and studying the evolution of the transfer of best practices and

knowledge management since they first appeared on the radar screen

of American industry. The APQC is a nonprofit source for perfor¬

mance improvement and decision support—information and knowl¬

edge, networking, research, training, and advisory services—located

in Houston, Texas. Through APQC’s International Benchmarking

Clearinghouse and its Institute for Education Best Practices, organiza¬

tions of all sizes and industries—business, government, education, and

health care—partner with APQC to discover global best practices and

grow into learning organizations. We bring consertia of organizations

together to find best practices, including four consortia we have

formed on knowledge management and the transfer of best practices

(see APQC references). It is through working with those organizations

that we have discovered much of what we share in this book.

At first, the signals were but a faint flicker: Shortly after the APQC

founded the International Benchmarking Clearinghouse in 1992, we

began to notice that as we searched for best practices for our members,

we were turning up many cases of unknown and unshared knowledge

in the very firms doing the benchmarking. The grass was greener in

their own back yard. And they did not even know it. We were puzzled.

These were not your average, run-of-the-mill companies. These were

winners. Pursuers of best practices. Seekers of new ideas. Some of the

most intellectually curious, performance-oriented organizations in the

world. Yet they did not even know about practices hidden, untouched

and undocumented, inside the walls of their own organizations.

Perplexed, we asked our member companies to participate in path¬

breaking research on best practices transfer. Led by Dr. Gabriel Szulanski

(formerly with INSEAD and now assistant professor of management

with the Wharton School of Business), this 1994 study was the first

practical study of the phases and barriers to effective transfer of knowl¬

edge in organizations. (More details about Szulanski’s findings can be

found in Chapter 3.)

Why didn’t knowledge and practices transfer?

It wasn’t because people are inherently turf-protecting, knowledge-

Preface • xi

hoarding beings. Not at all! Szulanski found that the number one,

biggest barrier to the transfer was ignorance. And ignorance on both

ends of the transfer. At most companies, particularly large ones, neb

ther the “source” nor the “recipient” knew someone else had knowb

edge they required or would be interested in knowledge they had. The

most common response from employees was either “I didn’t know that

you needed this” or “I didn’t know that you had it.”

Once people recognized that a better practice existed, the second

biggest barrier to transfer was the absorptive capacity of the recipient:

Even if a manager knew about the better practice, he or she might

have neither the resources (time or money) nor enough practical de-

tail to implement it.

The third barrier to transfer was the lack of a relationship between the

source and the recipient of knowledge; that is, the absence of a per-

sonal tie, credible and strong enough to justify listening to or helping

each other, stood in the way of transfer.

Finally—and here’s the real shocker—Szulanski found that even in

the best of firms, in-house best practices took an average of twenty-

seven months to wind their way from one part of the organization to an¬

other. Over two years’ lag time, in an era when new companies are

launched every nanosecond and information rushes through network

cyberveins at lightning speed. Nothing we have learned since then has

affected our world view as much as this one bit of information.

We have shared Szulanski’s insights widely with our five-hundred-plus

APQC International Benchmarking Clearinghouse member organiza¬

tions. They have led many to freshly examine the role and methodol¬

ogy of knowledge transfer in creating value in their own organizations.

We began to see more and more little flickers of light on our radar

screen.

These organizations didn’t know what they knew. We began to study

and work with organizations who did know, and many others determined

to emulate them.

This book is about those who seek excellence in their own back

yards. It is the product of three years of listening, questioning, observ¬

ing, cooperating, facilitating, and synthesizing the experiences of over

seventy companies that have embraced knowledge transfer as a strate¬

gic thrust for the twenty-first century. It is the latest effort in a long

xii • Preface

quest to understand how organizations can derive value from the

knowledge that lies throughout their operations.

WHAT YOU DON’T KNOW WILL COST YQU—OR RUIN YOU

How can you succeed in the knowledge era without knowing what you

know?

You can’t.

Only those organizations that methodically, passionately, and pro-

actively find out and transfer what they know, and use it to increase

efficiency, sharpen their product-development edge, and get closer to

their customers, will not only survive, but excel.

The book is organized around the following main messages:

1. Knowing how to transfer and leverage knowledge and best prac¬

tices will make you money.

2. To turn knowledge into profit, you must focus your transfer efforts

on one (or more) of three value propositions: (a) improving cus¬

tomer intimacy and customer-related processes and practices, (b)

honing product-to-market excellence, and/or (c) achieving oper¬

ational excellence.

3. Finally, to ensure grand designs turn into real-life improvements,

you need a process model and road map for making KM and best-

practice transfer work. Change without a recipe is a recipe for

chaos. The transfer model must describe not only the steps in the

process, but also the enabling context that is critical to its success:

organizational infrastructure, culture, information technology,

and measurement. These “enablers” will either help or hinder

your progress.

WHY READ THIS BOOK?

This is not the first book about managing of transferring knowledge

and it is certainly not the last book about knowledge management

(KM). There are many excellent books about knowledge management

per se (Davenport and Prusak, 1998; Edvinsson and Malone, 1997;

Stewart, 1997; Svieby, 1997), but we think our book is unique.

Preface • xiii

Here’s why:

First, this book is not only about why knowledge matters. We all

know it does. It’s a book about how to improve the performance of your

organization; it’s about how to generate profits using existing methodolo-

gies and in'house know-how. We don’t advocate that you go out and buy

expensive new systems. We don’t necessarily think you must hire more

consultants. What we do strongly advocate is the use of a specific KM

vehicle that we know, for a fact, works: the identification and transfer

of best practices. More than 80 percent of KM practitioners in our

studies rely on the internal transfer of best practices to grow their collec-

tive IQ. And with stunning results.

Second, this book is not based on theories or speculation. It is am

chored in successes, mistakes, and real-life case studies. It is not a spir¬

itual guide or a technology manual. The experiences, thoughts,

insights, and conclusions herein are based on surveys, site visits, and

design work with over seventy organizations of all shapes and sizes.

The APQC has conducted four major Consortium Benchmarking

Studies on knowledge management with results of unprecedented

scope and depth (APQC, 1996, 1997, 1998). The behind-the-scenes

look they offer, at the why and how of KM and best practice transfer, is

a veritable treasure chest of knowledge. And we want to share it with

you.

This book is primarily about internal transfer of best practices in orga¬

nizations. That is, the transfer of best practices from one part of an or¬

ganization to another part—or parts—in order to increase profitability

or effectiveness. We know it works. We have evidence that it works.

And we will share this with you throughout the book with numerous

case study examples from leading organizations, both profit-making

and nonprofit.

Finally, it is also a book about the transfer of knowledge, specifically,

the effective management of knowledge inside an organization. Trans¬

ferring best practices within an organization is much more effective

when it is part of an overall environment that values the sharing of

knowledge.

This book will focus largely on “internal benchmarking”—looking

inside your own organization—and transferring best practices. (Of

course, there is also great value in looking outside your organization—

xiv • Preface

“external benchmarking.”) We find that benchmarking and knowl¬

edge management benefit from each other the way a desktop PC be¬

comes more powerful once equipped with a browser and connected to

an intranet. Each is a useful tool itself. Together, what you create lo¬

cally on your PC can be shared widely, and what others have learned is

accessible to you through the intranet.

Benchmarking is the process of finding and adapting best practices.

Once you have done so, a knowledge management system helps to

spread the useful practices around the organization. Otherwise, even

the best practices will only have local benefit, or spread leisurely or by

luck. (Europe certainly would have benefited from a better knowledge

management system after Marco Polo benchmarked the Chinese.)

To benchmark as a verb means to systematically identify and learn

from best practices, internal or external, in order to improve your own

performance. The noun benchmark is a measure of performance. Bench¬

marks tell you how good others are. Benchmarking tells you how to get

there. You need both. Done right, benchmarking is less a study than a

contact sport; a method of learning how to learn, and a key component

of knowledge creation, adaptation, and implementation.

Since the early 1990s we’ve seen dramatic acceptance of bench¬

marking as a legitimate way to speed improvement and change. Xerox,

Chevron, Texas Instruments, Kodak, IBM, Citibank, GE, Amgen,

GTE, AT&T, and the U.S. Postal Service are emphatic about the

need to overcome the Not-Invented-Here syndrome and adopt best

practices.

As evidence of the exploding activity in benchmarking, APQC’s

International Benchmarking Clearinghouse has over five hundred

member organizations—a blue-ribbon group from business, govern¬

ment, health care, and education—that are constantly sharing knowl¬

edge with one another and seeking best practices in other sectors and

countries. Many of our members are from Canada, Asia, Australia,

South America, and Europe. Other benchmarking centers are being

created, such as the Asian Benchmarking Center, the Commonwealth

Benchmarking Club, and an Indonesian Benchmarking Clearing¬

house. APQC has taught benchmarking in over thirty countries.

APQC is partnering with the European Foundation for Quality Man¬

agement on a number of benchmarking studies.

Successful benchmarking has actually led managers to see how

Preface • XV

powerful and profitable managing knowledge can be. A word of cau¬

tion: sharing only internal knowledge and practices can lead to myopia

and the self-delusion that you are best. External benchmarking scours

across sectors and industries for excellence, causing goals to be set

much higher. Gains of as much as 30 to 50 percent are achieved.

Benchmarking can also help feed a knowledge management sys¬

tem’s voracious appetite for useful content. People want to know:

“Who has looked at this issue before?” “Has anyone benchmarked this

process?” “What did they learn?” Texas Instruments and Chevron have

extremely active repositories on-line that track past, current, and

planned benchmarking studies, provide a quick summary of the status

or results, and say whom to contact for more information.

Finally, when their forces are combined, benchmarking and knowl¬

edge management accelerate change: As Bob George of DuPont said,

“Benchmarking is a change management process and the one we use at

DuPont.” One of the reasons it works so well is that skeptics get proof

that best practices can work—seeing is believing. One manager of

Xerox said: “The only way to convince most managers of a new way

was to allow them to witness it with their own eyes. Talking to them

didn’t do the trick.”

Benchmarking and knowledge management are like love and mar¬

riage. You can have one without the other. But it’s far better if you

have both.

Finally, as you read this book, there will be times you will be skeptical.

That’s okay. So were we initially. We too have seen every fad in the last

twenty years. We do not think knowledge management and best prac¬

tices transfer are a fad. They do need heightened emphasis now, but in

a few years they will be standard operating procedures for successful

firms, like just-in-time, cycle time reduction, and total quality.

Be critical, but don’t be cynical. “It’s early days for KM,” says our

friend Tom Davenport, professor of management information systems

at Boston University. We know that identifying, managing, and trans¬

ferring knowledge and best practices has worked for some companies,

sometimes saving or earning them literally billions. What we’ve seen is

impressive. It’s important. If we wait to know all there is to know, we

may well be too late.

^ '

ACKNOWLEDGMENTS

Over a year ago, we decided to write this book to find out what we

know about the transfer of internal knowledge and best prac¬

tices, for as someone said, “I don’t know what I think until I write.” But

we also said, “If only we knew what they knew,” meaning the leading

organizations who are experimenting, inventing, learning, and chang¬

ing in this rapidly expanding work of transfer of internal knowledge

and best practices. To create this book, we have learned from literally

hundreds of organizations, whether they shared experiences or ques¬

tions. Over twenty companies you will read about let us visit them,

spend time at their sites, and publish their stories.

For their help on the detailed case studies in Part Four, we would es¬

pecially like to thank Robert Buckman from Buckman Laboratories,

Melissie Rumizen, formerly with the National Security Agency and

now with Buckman; Bill Spencer with NS A, Dave Ledet, Bill Lowrie,

and Don Tomberg of Amoco; John Davis of AMP; Bob Hiebeler of

Arthur Andersen; Ken Derr, Mike Callaghan, Greta Lydecker, Jim

O’Brient, and Jim Tighe of Chevron; Marc Demarest and Roger Swan¬

son of Sequent; Tom Engibous, Cindy Johnson, and Bill Baker of Texas

Instruments; Steve Denning of the World Bank; and Mary Halaszyn

and Susan Sellier of APQC for their attention to infinite detail and co¬

ordination, good judgment, and good humor. The APQC teams who

conduct our KM and transfer projects are the real creators of knowl¬

edge we share here. Many thanks to Peggy Odem, Stephanie Carlin,

Linda Muchisky, and Cindy Hubert for leading the way.

XVII

Part One

A FRAMEWORK FOR INTERNAL KNOWLEDGE TRANSFER

Part One lays the foundation for the rest of the book. In Chapter 1 we

provide our working definitions of knowledge and knowledge manage-

ment (KM). Is it a fad, soon to fade, or something real? Answer: It’s for

real.

In Chapter 2, we discuss “KM in Action: Transfer of Best Practices,"

and we give a few illustrations of transfer. More extensive examples will

come later in the book after we have laid the foundation.

In Chapter 3, we go deeper into the “Barriers to Internal Transfer, ” for

“transfer” isn’t as easy as it may sound at first.

In Chapter 4, we provide an overview of the components of creating a

successful transfer system: discovering your value proposition; creating the

environmental enablers; and embarking on a structured process for designing

the entire initiative.

Chapter 1

DEFINITIONS OF KNOWLEDGE AND KNOWFEDGE MANAGEMENT

Knowledge management is really about recognizing that re~

gardless of what business you are in, you are competing based

on the knowledge of your employees.

—Cindy Johnson, Director of Collaboration and

Knowledge Sharing at Texas Instruments

WHAT DO WE MEAN BY “KNOWLEDGE”?

The recorded study of learning and knowledge dates back at least to

Plato and Aristotle; however, its modern-day exploration is credited to

thinkers like Daniel Bell (1973), Peter Drucker (1993), Alvin Toffler

(1970, 1980), and the philosopher Michael Polanyi (1958, 1967).

Polanyi’s work served as the basis for the much-acclaimed knowledge

management theories and books by the Japanese organizational learn¬

ing guru, Ikujiro Nonaka (1991, 1995)—as of September 1997 ap¬

pointed Xerox Chair of Knowledge at his alma mater, The Haas

School of Business, the University of California at Berkeley.

Polyani and Nonaka both point out that knowledge comes in two

basic varieties: tacit and explicit, also known as informal/uncodified and

formal/codified. Explicit knowledge comes in the form of books and

documents, white papers, databases, and policy manuals. The tacit/

uncodified variety, in contrast, can be found in the heads of employees,

3

4 • If Only We Knew What We Know

the experience of customers, the memories of past vendors. Tacit

knowledge is hard to catalogue, highly experiential, difficult to docu¬

ment in any detail, ephemeral and transitory. Both types of knowledge

are important.

Some may argue that, in a commercial context, tacit knowledge

does not qualify as “knowledge” at all. Just as value is defined by the

“transfer price” in the context of seller/buyer interaction, thoughts in

our heads are not “knowledge” until they enter the marketplace of

ideas via discussion and interaction. “It is the intersection between

tacit knowledge and explicit knowledge that creates learning,” Non-

aka wrote in the February 1994 issue of Organizational Science.

For example, a manager who has just tried out a new sales technique

has “tacit” knowledge of it. If he writes it down and posts it on his com¬

pany’s intranet site, some of that knowledge has become captured and

“explicit.” Next, another sales manager reads the description and uses

the technique on her next sales trip (hence turns it into “tacit” once

more). Knowledge has been captured, exchanged, and created (see

Steps in the Knowledge Transfer Process, below). The learning process

hence involves the continuous “intersection” of these two knowledge

types and a never-ending, closed-loop transformation process.

Other organizational experts, such as Leif Edvinsson of Skandia,

further divide commercial knowledge into individual, organizational,

and structural knowledge. Individual knowledge is solely in the minds

of employees. Organizational knowledge is the learning that occurs on

a group or division level. Structural knowledge is embedded in the

“bricks” of the corporation though processes, manuals, and codes of

ethics. At any one of these three “states,” the knowledge can be either

tacit or explicit.

Knowledge is broader than intellectual capital (1C). Whereas some

writers have chosen to expand IC to include practices and processes, in

its purest form, IC refers to the commercial value of trademarks,

licenses, brand names, formulations, and patents. In this view, knowl-

edge-as-intellectual-capital is an asset, almost tangible. Our use of

knowledge is broader: we view knowledge as dynamic—a consequence

of action and interaction of people in an organization with informa¬

tion and with each other.

Knowledge is bigger than information. Our organizations are awash in

information, but until people use it, it isn’t knowledge. While you

Definitions of Knowledge and Knowledge Management • 5

can’t have too much knowledge, you can certainly have too much in-

formation. Indeed, many organizations have already discovered that

information, carried faster and in greater volumes by electronic media,

leaves employees overwhelmed, not overconfident. Fumbling rather

than focused. Paralyzed rather than proactive.

Hence, our simple working definition: Knowledge is information in ac¬

tion. In the organizational and commercial context of this book,

knowledge is what people in an organization know about their cus-

tomers, products, processes, mistakes, and successes, whether that

knowledge is tacit or explicit.

Data (facts and figures, without context and interpretation), and

information (patterns in the data), are not in themselves knowledge

(actionable information). For example, when a British supermarket

chain implemented a high-end customer datamining application, it

began to accumulate data on buying behavior. It then took the data

and ran correlation analyses among the seemingly unrelated points to

reveal buying behavior patterns. For instance, the chain quickly found

a clear correlation between the purchases of diapers and beer on Friday

afternoon. It took this curious piece of information, and hypothesized

that men, on a Friday afternoon shopping expedition, are likely to buy

beer (for themselves) and a pack of diapers (as per their wives’ shop¬

ping list). Armed with this knowledge about its customers’ behavior,

the store took action and reconfigured the locations of diapers and

beer on its shelves.

This leads us to the next fundamental question.

WHAT IS KNOWLEDGE MANAGEMENT?

Let’s start with what managing and sharing knowledge is not:

• Knowledge management (KM) is not a new religion or a spiritual

calling.

• It is not an attempt to rally disgruntled employees around an appeal¬

ing philosophical concept.

• It is not an existentialist search for the Truth. (Actually, it’s about

the entirely worldly task of making money.)

• It is not a science or a “discipline”—yet.

• It is not the latest management fad.

6 • If Only We Knew What We Know

When explicitly managed, organizational knowledge is used to accom¬

plish the organization’s mission. Knowledge management is therefore a

conscious strategy of getting the right knowledge to the right people at

the right time and helping people share and put information into ac¬

tion in ways that strive to improve organizational performance.

Fortunately, most companies have some experience already in man¬

aging knowledge. Indeed, KM is not a radical departure or a method¬

ology in and of itself. Rather, it is a framework, a management

mind-set that includes building on past experiences (libraries, data¬

banks, smart people) and creating new vehicles for exchanging knowl¬

edge (knowledge-enabled intranet sites, communities of practice,

networks).

For most organizations, KM represents a continuation of efforts

begun in other times with other names (or acronyms), all of which

have likely yielded valuable learnings.

• Radical reengineering may have not delivered sustainable success,

but it has “delivered” the mind-set of the process-oriented organiza¬

tion. Processes can be made explicit, and knowledge about how to

make them work can be transferred.

• Total quality management (TQM) may have not always yielded big-

time change, but it laid the foundation for a corporate-wide, system¬

atic initiative for measurement and change and cross-functional

teaming, all of which, we will argue, are critical to the successful

management of knowledge

Not surprising, newly trim and lean, reengineered companies like

Amoco, Chevron, and Texas Instruments have been at the forefront of

the KM “[devolution.” Companies like these have been among the

first to realize that to accelerate growth—again—they must adopt new

approaches that leverage their internal expertise.

This leads us to another building block in the foundation for the re¬

mainder of this book.

STEPS IN THE KNOWLEDGE TRANSFER PROCESS

Managing knowledge and transferring best practices is simple in concept,

but difficult in execution. It is important to lay out the key components

Definitions of Knowledge and Knowledge Management • 7

of this seemingly simple process (see Figure 1.1). Every knowledge

management and transfer initiative we will describe had to design ap-

proaches to address all of the steps in the knowledge transfer process

shown in Figure 1.1.

Most companies start their organized efforts by focusing on creating,

identifying, collecting, and organizing best practices and internal knowl¬

edge, in order to understand what they know and where it is. Later on,

we’ll cover the difficulties in these early stages, especially when dealing

with tacit knowledge and know-how.

As we stated earlier, just knowing that the practices or knowledge

exists is not enough to ensure transfer or use. The process must explic¬

itly address sharing and understanding of those practices by motivated

recipients. Finally, the process involves helping the recipients adapt

and apply those practices to new situations, to create new “knowledge”

and put it in action. This is where the payoff really comes.

DOES MANAGING AND TRANSFERRING

KNOWLEDGE REALLY WORK?

It does—big time—and for a growing number of companies, large and

small, private and public, in services and in manufacturing, in high

tech and in chemicals. From the National Security Agency (NSA), to

Use

Adapt

Share

Organize

Create

Identify

Collect

FIGURE 1.1

Steps in the Knowledge Transfer Process

8 • If Only We Knew What We Know

Texas Instruments, from Chevron to Verifone. Megasize multination¬

als and small, niche players.

• At Buckman Laboratories, their transfer of knowledge and best prac¬

tices system helped push new product-related revenues up 10 per¬

centage points and sales of new products up about 50 percent (from

22 percent to almost 35 percent in 1996). Responding to customer

inquiries about products now takes hours instead of weeks. (See case

study in Chapter 14.)

• Texas Instruments generated $1.5 billion in annual increased fabrica¬

tion capacity (in effect, a “free” plant) by comparing and transfer¬

ring best practices among its existing thirteen fabrication plants.

Plant managers and teams from Texas Instruments’ Semiconductor

Group, led by that group’s president (now TI president and CEO)

Tom Engibous, created the equivalent capacity of an additional

semiconductor wafer fabrication plant, thereby avoiding a $500

million investment and providing needed capacity to customers.

They called it a “free fab,” and have repeated this triumph two more

times, for a total of more than $1.5 billion in cost avoidance, in ad¬

dition to going from last (1992) to first (1994) in on-time delivery

satisfaction in customer rankings. (See case study in Part Four,

Chapter 15.)

• At Dow Chemical, early efforts to manage intellectual capital

brought an immediate kick-back in the form of $40 million in sav¬

ings. Analysis of existing patents to determine which technology

streams were the strongest and which were weakest allowed more

effective negotiations with joint venture partners.

• At Kaiser Permanente, benchmarking of their internal best practices

helped drastically cut the time it took to open a new Woman’s

Health Clinic. And it opened smoothly, with no costly start-up

problems.

• At CIGNA Property & Casualty, knowledge-sharing efforts, com¬

bined with a reengineering campaign, lifted profits back into the

black. In 1993, CIGNA lost more than a quarter of a billion dollars.

By 1995, it sported a $90 million profit and has continued with

healthy profits since then.

• Skandia has leveraged internal know-how to dramatically reduce

Definitions of Knowledge and Knowledge Management • 9

start-up time for new ventures to seven months, compared to an in¬

dustry average of seven years.

• By comparing practices on the operation of gas compressors in fields

in California, the Rockies, and offshore Louisiana, a Chevron team

learned that they could save at least $20 million a year just by adopt¬

ing practices already being used in their best-managed fields.

• Chevron’s network of 100 people who share ideas on energy-use man¬

agement has generated an initial $150 million savings in Chevron’s

annual power and fuel expense by sharing and implementing ideas to

reduce company-wide energy costs. By 1996, Chevron could credit

this best-practice transfer team with generating over $650 million

in savings. And they’re still going.

• At Arthur Andersen, a Global Best Practice Knowledge base has

improved the quality of services, helped lower research costs, and

shortened delivery time in business consulting. (See more detail on

page 42.)

• At USAA, knowledge management increased the proportion of

business conducted over the phone with members from 30 percent

to 70 percent and helped establish ten new strategic alliances.

We will get back to these examples later in the book and in greater

detail. But you get the point. We certainly did. And for us, it was a

point of no return. There was no way we could ignore this outpouring

of experience, interest, and practice. We knew just how much there was

to know. And we set about to find out as much as we could.

10 • If Only We Knew What We Know

MORE THAN JUST A FAD

Is KM another management fad? We don’t think so, for four reasons:

1. It is built on the never-obsolete power of learning. As David

Garvin of Harvard Business School asks, “How can an organiza¬

tion improve without learning something new?”

2. While there are plenty of people who treat knowledge management

as a religion, real knowledge management is practical and action

oriented, not ideological and theoretical. If done right, it produces

bottom-line results—always a sure way to guarantee sustainability.

3. Unlike other process-improvement methods, KM does not rely on

technology to make processes more efficient. It relies on recogniz¬

ing the knowledge resident in people’s minds, using technology to

facilitate its sharing, not replace its human origins.

4. Finally, KM is consistent with emerging models of organizations.

Most modern business models involve people in teams coming to¬

gether on a project basis, then moving on to new relationships. All

these models are process oriented, not bound by functions, indus¬

tries, structures. Knowledge underpins their continuous existence.

Chapter 2

KM IN ACTION—THE TRANSFER OF BEST PRACTICES

Every day that a better idea goes unused is a lost opportunity.

We have to share more, and we have to share faster. I tell

employees that sharing and using best practices is the single

most important thing they can do.

—Ken Derr, chairman and CEO, Chevron Corporation

It would be naive to believe there is one answer to any organizational

malaise. It’s a consultant’s dream and every manager’s secret hope.

But in real life—and that’s what this book is about—such a simplistic

approach often fails.

As every reengineering guru will tell you, the biggest problem with

many mega-transformation efforts has been the attempt to prescribe a

single remedy to a host of organizational problems. No two companies

are alike.

But while there is no single answer, there appears to be a growing

consensus that the fastest, most effective and powerful way companies

can manage knowledge assets is through the systematic transfer of best

practices.

This evolving consensus is not a theoretical notion. Nor is it the

product of brainstorming sessions in the halls of academia or the corn-

dors of consulting practices. Rather, it is a consensus emerging from

11

12 • If Only We Knew What We Know

the day-to-day, nitty-gritty, practical experiences of companies that

have dared to put themselves at the forefront of this new management

practice.

Sharing best practices inside organizations is not confined to the

private sector. The U.S. National Security Agency’s Office of Plans

places “lessons learned” from past crises into an on-line database avail¬

able to everyone in the NSA system worldwide, 24 hours per day, and

kept updated within 48 hours.

WHY “BEST PRACTICE TRANSFER”?

Because it produces results.

Best practices take information/data and put them in the context of

real people and real experiences within the company. We learn by

doing and by watching others do. The transfer of best practices helps

others in the firm learn better, faster, and more effectively.

Recall the KM success stories we listed in Chapter 1? In virtually

every case the knowledge-transfer strategy used for promoting effective

organizational sharing has been the systematic transfer of best prac¬

tices.

The graph in Figure 2.1 details the results of our Emerging Practices

in KM Consortium Benchmarking Study (APQC, 1996). The survey

results corroborate the anecdotal evidence: Systematic best practice

transfer is the one strategy pursued by 100 percent of the organizations

pursuing value-through-knowledge.

WHAT DOES “BEST PRACTICES” MEAN ANYWAY?

We know the term “best practices” is fraught with peril; it can lead to

arguments about the validity of the term “best.”

Labeling any practice as “best” immediately raises a hue and cry of

dissenting voices in the organization. Not only is “best” a moving tar¬

get in today’s world, but “best” is also situation-specific. Opponents of

benchmarking have long argued that no one knows what’s “best,” and

what’s optimum in one place may not be even good for another. We

agree.

That is why we prefer the terms “better” or “exemplary” or “success-

KM in Action—The Transfer of Best Practices • 13

Number of Companies with Objective Number of Respondents =19

FIGURE 2.1

Objectives of Knowledge Management

fully demonstrated” to “best.” But we will continue to use the term

“best practices” because it has such common usage and because we do

always want to strive for the best.

So our definition of best practices is “those practices that have pro-

duced outstanding results in another situation and that could be

adapted for our situation.”

Nonetheless, the term “best” causes problems: Internal competition

and rivalry rear their heads as some units start to quibble about whether

they have been measured correctly. Others rightfully raise questions

about spending significant resources to transfer a practice if the poten¬

tial payoff has not been demonstrated in multiple locations.

“BEST” AT CHEVRON

One way to cope with these problems is to define levels of best practice

as Chevron has done. The corporation has adopted a simple definition

of best practices: Any practice, knowledge, know-how, or experience

that has proven to be valuable or effective within one organization

that may have applicability to other organizations.

14 • 1/ Only We Knew What We Know

Chevron recognizes four levels of best practices in both its corporate

databases and best practice teams:

1. Good idea—unproved, not yet substantiated by data but makes a

lot of sense intuitively; could have a positive impact on business

performance. Requires further review/analysis. If substantiated by

data, this could be a candidate for implementation in one or

more Chevron locations.

2. Good practice—a technique, methodology, procedure, or process

that has been implemented and has improved business results for

an organization (satisfying some element of customers’ and key

stakeholders’ needs). This is substantiated by data collected at

the location. A limited amount of comparative data from other

organizations exists. It is a candidate for application in one or

more locations within an operating company or department and

possibly at other locations at Chevron.

3. Local best practice—a good practice that has been determined to

be the best approach for all or a large part of an organization (op-

erating company or department level), based on an analysis of

process performance data. The analysis includes some review of

similar practices outside of Chevron (competitive intelligence

data). This practice is applicable at most or all locations within

the operating company or department and may be applicable to

other Chevron locations.

4- Industry best practice—a practice that has been determined to be

the best approach for all or large parts of an organization. This is

based on both internal and external benchmarking work, includ¬

ing the analysis of performance data. External benchmarking is

not confined to the organization’s industry. This process may be

applicable to other Chevron locations.

For Chevron’s on-line best practices sharing databases, Chevron

holds contributors responsible for deciding if the practice is worth

sharing with others and into which category of “best” it fits.

BEST VS. BETTER AT AMP

AMP, the world’s largest manufacturer of electrical and electronic in¬

terconnection devices ($5 billion in annual revenues, and over 40,000

KM in Action—The Transfer of Best Practices • 15

employees in forty countries) has taken a different tack to “labeling”

practices.

For years, AMP had promoted the transfer of “best demonstrated

practices.” But it found that the label proved counterproductive: It

gave people the impression that there was only one best way to do

something, provoking resentment among the laggards, and fear of being

accused of unjustified arrogance among the top performing units.

AMP switched to “successfully demonstrated practices” (SDPs) and

added credibility to its new labeling system by establishing clear crite¬

ria for successful demonstration.

• Has the SDP resulted in measurable improvement for the organiza¬

tion?

• Has the SDP been recognized by internal and external experts or

sources?

• Has the SDP been recognized through business assessments and au¬

dits?

As John Davis, benchmarking manager at AMP, points out, resis¬

tance to learning from others in the firm has not disappeared, but it has

dramatically decreased.

These very practical definitions of “best” address one of the hurdles

to transfer. In the next chapter, we address deeper issues to be over¬

come.

Chapter 3

THE BARRIERS TO INTERNAL TRANSFER

We are asked all the time, “If the potential payoff for internal

transfer of best practices is so great, why don’t all organizations

do it?”

You would think better practices would spread like wildfire to the

entire organization, but they don’t. One Baldrige-award winner has

told us: “We can have two plants right across the street from one an¬

other, and it’s the damnedest thing to get them to transfer best prac¬

tices.”

Indeed, executives have long been frustrated by their inability to

identify or transfer outstanding practices from one location or function

to another. They know some facilities have superior practices and

processes, and the results to prove it, yet executives continue to see op¬

erating units reinventing or ignoring solutions and repeating mistakes.

It happens in business, in health care, in government, in education.

So why doesn’t transfer occur?

We believe most people have a natural desire to learn, to share what

they know, and to make things better. So it’s not something inherently

wrong with human nature that’s stopping the transfer of internal best

practices. Rather, this natural desire is thwarted by a variety of logisti¬

cal, structural, and cultural hurdles and deterrents present in our orga¬

nizations. As a result, the actual process of identifying and transferring

practices is trickier and more time-consuming than most people imag¬

ine. It must involve a conscious dismantling of these organizational

barriers.

16

The Barriers to Internal Transfer • 17

HISTORICAL HURDLES TO TRANSFER

In 1994, APQC participated in research to understand what prevents

the transfer of practices across a company. The study was headed by

Gabriel Szulanski, assistant professor of management at Wharton. The

results were startling: It revealed that a practice would linger in a com'

pany for years unrecognized and unshared. Even when it was recog¬

nized, it still took over two years on average before other sites began to

actively try to adopt the practice, if at all (see Szulanski, 1995).

What was taking so long? Szulanski’s research pinpointed four key

barriers:

Reason #1: Ignorance. Those who have the “knowledge” don’t real¬

ize others may find it useful. At the same time, those who could

benefit from that “knowledge” have no idea someone in the com¬

pany already has it.

Reason #2: No absorptive capacity. Even when employees were not

ignorant of the knowledge or best practice, they lacked the money,

time, and management resources to pursue and study it in enough

detail to make it useful.

Reason #3: The lack of preexisting relationships. People absorb knowl¬

edge and practice from other people they know, respect, and—

often—like. If two managers have no personal bond, no tie or

link which preestablishes trust, they’re less likely to incorporate

each other’s experiences into their own work.

Reason #4: Lack of motivation. People may not perceive a clear busi¬

ness reason for pursuing the transfer of knowledge and best prac¬

tices.

CONFRONTING SYSTEMIC BARRIERS

Whereas some of these hurdles are personal, they are by-and-large the

result of a set of organizational structures, management practices, and

measurement systems that discourage—rather than encourage—shar¬

ing. Companies that ignore these ingrained Systems (yes, systems with

a big S) are naive. To tell people to share without first addressing sys¬

temic obstacles will only lead to disappointment and failure.

Organizational personalities come in all sorts of shapes and sizes, but

18 • If Only We Knew What We Know

more often then not, they’ll fall into five basic categories when it

comes to knowledge transfer:

The Silo Company Inc. This multinational company has structures that

clearly promote “silo” thinking and behavior on the part of managers

and employees. Silos can be as small as a function or department and as

large as a division or business unit. Each, however, is focused on maxi¬

mizing its own accomplishments and rewards. Groups tend to hoard

information to prevent others from excelling while improving their

own relative performance. As a result, they substantially suboptimize

the performance of the organization as a whole. As one manager said,

“When it comes to bonus time, we play a zero-sum game around here.

To get my share of the bonus pool, I have to take it away from someone

else. Why should I share my best ideas?” Without leadership, an incen¬

tive scheme, and culture supportive of transfer, people have little in¬

centive to overcome sharing obstacles created by departmental lines

and geographic barriers.

The NIH Company Inc. At the world-famous NIH Company, the

prevailing culture values personal technical expertise and knowledge

creation over knowledge sharing, on either the micro (interdepart¬

mental) or macro (intercompany) level. Otherwise known as the Not

Invented Here (NIH) syndrome, this sort of attitude is rampant in

engineering-based organizations as well as knowledge-based organiza¬

tions, such as consulting and research firms. The reason is simple

enough: For these companies, the ability to provide unique products

and services is the key value proposition and business driver. The em¬

phasis is therefore on invention rather than adaptation of existing

ideas/technologies. Whether on a corporate (“we don’t use what oth¬

ers have invented”) or intracompany basis (an employee’s value is

determined by original thought, not necessarily clever commercializa¬

tion), the resulting culture makes individuals feel “bad” if they “bor¬

row” someone else’s idea. Instead of being unwilling to share, as the

Silo Company Inc.’s employees are likely to be, workers at NIH are un¬

willing to absorb.

The Babel Company Inc. This company suffers from an acute case of

“Babelitis.” Far-flung employees and sometimes even co-workers lack a

The Barriers to Internal Transfer • 19

set of common perspectives and terms that can serve as the basis for ef¬

fective communications and transfer of knowledge. In many cases, the

left hand does not fully know or understand what the right hand is

doing; it may not even know there is a right hand. Different departments

use different words to describe, catalogue, and record their processes

and practices. Such departmental dialect impedes cross-functional dis¬

course. There is no single, coherent vocabulary to express processes

and performance. Employees may be willing to share and absorb, but

without a common “book of reference” this company cannot begin to

record its collective know-how and best practices. No one would un¬

derstand them.

The By-The-Book Company Inc. The By-The-Book Company is not

averse to sharing. In fact, however, it considers documented knowledge

the end-all and be-all of knowledge transfer. By-The-Book exhibits a

serious tendency to rely solely on transmitting “explicit” rather than

“tacit” knowledge. It builds elaborate databases and implements dis¬

tributed computing platforms that allow one and all to record and ac¬

cess documented knowledge. That’s valuable stuff. The problem is that

most of the important information people need in order to implement

a practice cannot be codified or written down. Practices have to be

demonstrated and “recipients” engaged in interactive problem solving

before the knowledge “sinks in.” Just creating databases will not cause

change to happen. It will not make people share. Polanyi (1966) and

Nonaka and Takeuchi (1995) have both pointed out the importance

and value of recognizing and trying to transfer tacit knowledge—the

know-how, the judgment, the intuition, the little tricks that constitute

the noncodifiable knowledge that may make the difference between

failure and success in the transfer.

The Bolt-It'On Company. This enterprising firm believes you can add

transfer and sharing responsibilities on top of everyone’s regular work

process and expect to get results. With lots of fanfare and high hopes,

Bolt-It-On starts “sharing” programs, on-line chat groups, and the like.

Then waits... and waits. .. and not much happens. Through our

work with Bolt-It-Ons, we have found that embedding knowledge

practices and the information technology to support them in everyday

processes and work is key. Knowledge management works to the extent

20 • If Only We Knew What We Know

that it helps people achieve their work objectives in support of the

organization’s mission. Overlaying additional work on top of the old

way of working will not produce new results.

All of us recognize some aspects of our own organization in one or

more of these exaggerated near-caricatures. Larger companies may

have a N1H Company in the United States, with a Bolt-It-On outpost

in Europe.

Recognizing these structural impediments, understanding them,

and changing them are perhaps the tallest hurdles faced by managers

today. The hurdles are large, but they are not insurmountable.

If you want knowledge sharing to yield the significant benefits that

it can yield, you must first adopt a strong intention and a methodical

approach to the management and transfer of knowledge. To get this

process rolling, we have found that you need:

1. A compelling need to change—something that is important to the

firm

2. A clear-eyed assessment of the current state of knowledge and

transfer relative to that problem or opportunity

3. A detailed project design and ongoing management involvement

4- A good implementation plan to provide and align supportive orga¬

nizational resources and structures (no surprise here)

Specifically, our studies indicate that a growing number of compa¬

nies as diverse as Chevron, Texas Instruments, Xerox, AMP, and many

others are taking action. To overcome hurdles and create a successful

initiative requires (1) a clear focus on the desired results, (2) an en¬

abling environment, and (3) an explicit change process. The next

chapter is an overview of how to do that.

Chapter 4

A MODEL FOR BEST PRACTICE TRANSFER

The key to making best practice transfer work is to approach this

challenge like the change initiative that it is.

First, change without purpose is change without direction or results.

To extract value out of knowledge, companies must clearly define their

value propositions—what they hope to achieve through the more effec¬

tive management and transfer of knowledge.

Second, to ensure knowledge is created, captured, shared, and lever¬

aged, companies must create an environment where the four enablers of

information technologies, culture, measurement systems, and organi¬

zational infrastructure are supportive of sharing.

And third, companies need to approach the planning, design, and

implementation of a transfer system with the same structured process

required by a change effort of any potentially quantum-leap scale. We

lay out a four-phase process.

A MODEL FOR ACTION

In Figure 4.1, we lay out a model that will guide the rest of this book. It

has three major components:

1. The three value propositions

2. The four enablers

3. The four-step change process

This model applies to knowledge and practices about customers,

products, processes, mistakes, and successes. It includes not just explicit

knowledge. It also applies to tacit knowledge: intuitions, judgment,

21

22 • If Only We Knew What We Know

and know-how. As Jerry Baker of National Semiconductor put it,

“Somebody held their tongue just right as they pulled the wafers out of

the oven and that’s what made things work.”

1. START WITH THE VALUE PROPOSITION

A map is only good if you know where you’re headed. Thus, at the core

of the best-practice transfer model are the organization’s value proposi¬

tions.

The value proposition provides its unique business rationale for em¬

barking on a knowledge-enabled change journey. Each company will

have a slightly different set of reasons for wanting to transfer knowl¬

edge and best practices. Yet, we have found that value propositions

tend to fall into three basic categories:

• Customer intimacy

• Product'tO'market excellence

• Operational excellence

If these look familiar, that’s because they are. They first showed up in

Treacy and Wiersema’s The Discipline of Market Leaders (1995). We will

Infrastructure Culture

2. Design

Measures

FIGURE 4.1

A Model for Best Practice Transfer

A Model for Best Practice Transfer • 23

devote Part Two to an exploration of the value propositions and how

transfer programs differ in focus and results depending on which one

an organization selects. A short description of each follows here.

Customer Intimacy

This strategy focuses on capturing and using knowledge across the

company about how to market, sell, and service customers more effi¬

ciently and effectively. By sharing knowledge about customer needs

and behavior, these companies hope to sharpen their sales and market¬

ing edge, retain more valuable clients, and provide higher-value prod¬

ucts and services with better profit margins. Good examples in this

category are consulting firms like Ernst & Young, McKinsey, Price Wa¬

terhouse, and Arthur Andersen, as well as Buckman Laboratories,

USAA, and Sequent Computers. All have focused their KM efforts on

arming their front-line employees—their sales force, and consul¬

tants—with the collective intelligence of the organization.

(A related, but not identical, area of KM revolves around cus¬

tomer datamining and help-desk technologies. But these information-

technology-intense applications are examples of more statistical

aspects of turning information into market knowledge, and fall outside

the primary scope of this book.)

Product-to-Market Excellence

Finally, many companies are beginning to apply the principals of

shared learning and knowledge management to the product develop¬

ment process. The idea is to use best practices in product development

to accelerate time-to-market. Good examples here are Hughes Air¬

craft, Hoffmann-La Roche, IBM, and Skandia. Some are working hard

to leverage their existing assets and patents through new product and

service development.

Operational Excellence

Organizations in this category want to boost operational excellence

by transferring best practices from one plant or location to receptive

sites throughout the global organization. Classic examples are Texas

24 • If Only We Knew What We Know

Instruments and Chevron. Both companies have used knowledge

management principles and transfer practices to eliminate “islands” of

mediocrity, create savings, and achieve process improvements and new

production capacity.

2. THE FOUR ENABLERS

We strongly believe a key reason KM efforts falter is that the enablers

of the KM process remain poorly understood and even more poorly

managed. We will spend significant time on the enablers in Part

Three, but here is an overview.

Culture

Fortunate organizations start with a culture supportive of knowledge

management, one with a strong professional ethic and pride supported

by well-honed skills in teaming, including cross-functional teams. It

also helps to have a common improvement approach (like the Mal¬

colm Baldrige Quality Award criteria) as a basis for thinking about

work, improvement, and process. But if an organization is not so fortu¬

nate to start with a supportive culture, it must expend efforts to create

such a culture, or risk failure.

How? One way is by providing true leadership. Leadership (read “se¬

nior management”) support is usually gained not by proposals, argu¬

ments, or reports, but by financial and compelling competitive needs,

such as “If we don’t do it, we’ll lose market share.” Only a minority of

firms we know of use formal financial rewards as incentives for sharing

behavior. Some do embed knowledge development and transfer in

their professional and career development systems. Price Waterhouse

is a good example. In the past, promotions were based on seniority and

tenure, which did not encourage knowledge sharing. Now, Price Wa¬

terhouse has included knowledge sharing in its performance appraisal

system. Consultants must be able to produce “evidence” of actual shar¬

ing such as tutoring/training, development of methodology, publishing

and presenting on topics, coaching and mentoring.

A Model for Best Practice Transfer • 25

Technology

The explosive growth and ready adoption of Internet and intranet

technologies has been an enormous catalyst for knowledge sharing.

The key is to understand the limitations as well as the power of tech'

nology. It makes connection possible, but does not make it happen. As

one manager put it, “If you wire it, they won’t necessarily come.”

Infrastructure

Leadership, a healthy culture, and basic information technology are

necessary but not sufficient. To work, KM must be institutionalized

into the organization through the creation of new support systems,

other than information technology with new job responsibilities, new

teams, and new formalized networking.

Robert Buckman, chairman of Buckman Laboratories Interna'

tional, a specialty chemicals company headquartered in Memphis,

Tennessee, spent years preaching to his employees the value of sharing

information about customers. They bought the message, but the payoff

didn’t happen until Buckman created a systematic approach for sharing

know'how in 1992. The company’s knowledge-sharing infrastructure

uses “section leaders” as KM champions throughout the enterprise, en-

couraging sales people in eighty countries to transfer knowledge to the

company’s R&D labs about customer requirements, and to help each

solve customer problems and make sales. In just five years, sales of new

products have risen from about 22 percent to almost 40 percent.

Measurement

While this is the least developed aspect of knowledge management, we

believe it is important to measure the projects and business processes

that are being improved through knowledge management tools, and

let the users evaluate the contribution. Probably the organization that

has done the most in measuring work is Skandia Insurance of Sweden,

where Leif Edvinsson has been able to create knowledge measures that

appear on Skandia’s balance sheet.

Understanding how each enabler affects the process of best practice

transfer is the first challenge. Next, you will have to ensure all four are

26 • If Only We Knew What We Know

managed in harmony. If the technology allows sharing, but the culture

says “keep what you know to yourself,” transfer won’t happen. If there

are no designated knowledge champions and facilitators, even a com¬

pany with a pro-sharing culture may not succeed. If there is no process

for designing and managing change, good intentions will flounder.

3. THE FOUR-PHASE CHANGE PROCESS

A change initiative requires a map to guide the company’s transition

from the current state where knowledge is managed haphazardly, un-

deliberately—if at all. There may be some attempts at the grass-roots

level, or none at all. The desired state is an organization that has em¬

braced internal transfer of knowledge as a core process designed to de¬

liver dramatic and sustainable improvement in performance.

But how do you get from “as-is” to the new and improved version?

What do you do first? What do you need to consider? We will talk more

about specific choices and courses of action companies might follow in

Part Five, and offer tools to help you make your own decisions about

what’s likely to work best for you. But no matter what your specific cir¬

cumstances, your change process would likely follow the following four

phases. (See Figure 4.1.)

1. Plan

2. Design

3. Implement

4. Scale up

Planning involves self-assessment (Where are we today?) and a list

of clearly defined value propositions (What do we want to become?).

A comprehensive design phase involves outlining the roles and func¬

tions of people and technologies, as well as any necessary overlay to

the organizational structure and performance measures. Implementation

normally involves a pilot program (that is, a proof-of'Concept) that will

test new ideas and yield lessons in what works and what doesn’t. Just as

critical, the implementation stage is likely to provide much-needed

success stories to drum up organization-wide enthusiasm. The next

phase is scaling up the pilot to an enterprise-wide process to capture the

full benefits of effective transfer.

A Model for Best Practice Transfer • LI

SUMMARY TO PART ONE

The internal transfer of knowledge is about finding out what you know,

and using it to improve performance. It is about leveraging the value of

knowledge you’ve already got. Whereas different companies adopt dif¬

ferent approaches to finding and sharing internal know-how, they all

seem to pursue one single strategy with great vigor: the transfer of in¬

ternal best practices.

Why? The reason is simple: Best practices encapsulate knowledge

inside a caplet of action/experiences. Knowledge in action is a lot eas¬

ier to digest and a lot easier to implement. We can clearly see it works.

We can even talk to the people who made it work. Indeed, increasing

numbers of companies have come to believe that the transfer of inter¬

nal best practices is the fastest and most effective way to achieve im¬

provements.

Sure, companies have embarked on change efforts before. This one

is different. It’s different because the improvement work is anchored in

real-life practical knowledge—the know-how and intelligence other

people within your own company have developed and used. It has

worked for others. It can work for you.

The key for making this work is threefold:

1. You’ve got to have a clearly defined purpose: the value proposition.

2. You’ve got to understand and leverage various organizational en¬

ablers, from infrastructure to technology, from measurement to

culture.

3. You’ve got to have an organized way for achieving it: the four'

phase process.

The remainder of this book is about these three components and

the companies that illustrate their effective use.

*

Part Two

THE THREE VALUE PROPOSITIONS

Companies must transfer knowledge and best practices to create value,

and value is created by translating knowledge into action. But exactly

what “value” are we talking about? Higher stock value? Higher dollar sales?

Fatter profit margins?

The following section presents examples of how companies harness their

know Tow and best practice transfer efforts to achieve one of three specific

value propositions: customer intimacy, product'tO'market excellence, or op'

erational excellence.

Infrastructure Culture

1. Plan

4. Scale-up

s~\ a Value V

■ Proposition

3- Implement

2. Design

Technology r r, , Measures Enabling Environment

A Model for Best Practice Transfer

Chapter 5

FIND YOUR VALUE PROPOSITION

Knowledge creates value when it is put into action. But not just

any knowledge or any action. An enthusiastic exhortation of

“Let’s all share!” is likely to go unheeded—if you are lucky. If you

aren’t, dozens of erratic—albeit well-intentioned—attempts at trans¬

fer will be launched, only to quickly fall back to earth with little to

show for the effort.

To translate knowledge into value, companies need focus. They

need to take a hard look at the set of challenges and opportunities that

drive their particular business. Next, they must determine the areas in

which sharing of knowledge and best practices can help improve per¬

formance.

The first step in a firm’s pursuit of knowledge and best practice

transfer is identifying its own value proposition: What is the strategic

problem or process that could be improved if knowledge about best

practices were better shared and transferred in your organization? Ex¬

hortations to “share” are just so much hype without a business problem

that people really want to solve. The key to effective management of

knowledge is to create processes that will put that knowledge into ac¬

tion.

DIFFERENT STROKES FOR DIFFERENT FOLKS

A value proposition is the logical link between action and payoff: If we

do A, then B will happen, and the return on B is higher than either the cost

of A or the potential risk of not making B happen. The value proposition is

the business case for action (and investment).

31

32 • If Only We Knew What We Know

Each company has a different set of factors that drive or impede its

performance. For banks and utilities, caught in the throes of deregula¬

tion, reinvigorating customer intimacy is a key driver of value.

Some high-tech firms, in contrast, compete more on the speed of

product innovation than on whether or not they remember their cus¬

tomers’ birthdays. For them, best practice in product development is an

important driver of success.

Finally, capital-intensive organizations such as chemical firms often

find that operational excellence is a “make or break” proposition. If their

plants churn out less product at higher cost, they have no hope of com¬

peting effectively.

For any firm, however, getting value from transfer of knowledge

and best practices starts with focusing efforts on that firm’s value proposi¬

tion. Ask yourself this: Do you want to compete on the basis of customer

knowledge and service, product development and time-to-market; or

low-cost, high-quality operational efficiency? All? Two of three?

FOCUS, FOCUS, FOCUS ...

Why is it so critical that companies define their value proposition and

focus their efforts accordingly?

Focus ensures valuable resources are applied to high-payoff areas. There are

only so many resources to spend on managing knowledge. By figuring

out what counts—first—companies have a better chance at reaping

big-time results, boosting not only the bottom line but also the credi¬

bility of the transfer effort.

For example, before Sequent Computer Systems, Inc. launched a

company-wide best practice and knowledge sharing initiative, it iden¬

tified three business features that affect profitability:

• knowledge-intensive pressure points in its value chain

• a detailed model of cycle time and flow in its channel operations

• a good handle on project costs

Sequent then focused its transfer initiatives on points in the value

chain that (a) reduce project cycle time, (b) decrease costs, (c) increase

Find Your Value Proposition • 33

volume (of customers, projects, and therefore revenue), or (d) manage

risks associated with operations.

Focus helps guarantee transfer of the “right” knowledge. It helps people de¬

termine what knowledge should be captured and transferred and al¬

lows people to see a clear reason for this KM and transfer activity.

Finally, focus gets management’s attention—and funding. Senior manage¬

ment is bombarded daily with requests to support some program that is

going to “transform the company.” Most have seen enough fads come

and go; they are no longer swayed by glossy proposals or snazzy reports.

What does catch management’s attention is demonstrated results

(a.k.a. financial); successes within their organization or at other firms

which they respect; or a compelling competitive need (“if we don’t do

it, we’ll lose market share”).

One certain way to get executives involved is to ask them to define

what knowledge they think is valuable and where they think the orga¬

nization is missing the boat. At Sequent (Sequent Corporate Elec¬

tronic Library—see case study in Chapter 17), top executives were

asked to draw up a map of the revenue “value chain,” the key processes

that made Sequent money, and then identify gaps where more and bet¬

ter knowledge could make a real difference. The company ended up fo¬

cusing on its sales and marketing function.

The principles of KM and transfer may be similar across the board,

but their application differs for each of the three value propositions. It

differs both in the “types” of knowledge that will be captured (know-

what, know-why, know-how, know-who), and in the way transferred

practices will be used at the receiving end. (More on the particulars of

each value proposition in the following chapters in this section.)

WHICH VALUE PROPOSITION IS BEST FOR ME?

The simplest way to choose an area of focus is to look for the highest

pain or the highest gain. This is the classic methodology for targeting

improvement work and investment: Where are the highest costs?

Greatest revenue potential? Biggest competitive threat? Find those,

and focus your efforts there for the biggest payoff.

34 • If Only We Knew What We Know

If you are a senior manager (i.e., you control the purse strings), then

we suggest you take a hard line and insist on a business case for prof¬

itable results (even if it costs to get one) before you invest in any trans¬

fer initiative. Perform internal and external benchmarking to arrive at

an estimate of what the new initiatives may yield in terms of measur¬

able results and what it would cost to close the gap between your re¬

sults and the benchmark level. (See Part Five.)

Ultimately, companies should align their transfer work with their

strategy for competitive advantage. One should feed the other. When

Sequent figured out that its clients value its sales force’s knowledge, it

made sure knowledge and best practices could flow quickly to the front

lines. At Texas Instruments, competition and an expensive production

base forced an initial focus on operational excellence.

In their study of over forty organizations (Treacy and Wiersema,

1993), Michael Treacy and Wiersema advanced a three-pronged value

discipline theory that has since become the basis for many a book and

reengineering methodology.

Leading firms, Treacy and Wiersema said, have become so “by nar¬

rowing their business focus, not broadening it. They have focused on

delivering superior customer value in line with one of three value dis¬

ciplines—operational excellence, customer intimacy, or product lead¬

ership.”

Behind the push for a more focused approach to excellence has been

a shift in customers’ perception of “what counts.” In the past, cus¬

tomers judged value on the basis of price, quality, or a combination

thereof. These days, value is a much broader term, encompassing ease

of purchase, postsale support, dependability, and more.

For example, firms with customers that care more about support

than cost can choose customer intimacy as their route to greatness (an

example is Home Depot). Companies whose customers care more

about convenience then hand-holding often pursue operational excel-

lence as their road to success (for example, Dell Computers). Finally,

where novelty and “first-to-market” are essential qualities, product ex¬

cellence is the key to profitability. “Companies that push the boundaries

of one value discipline while meeting industry standards in the other

two gain such a lead that competitors find it hard to catch up,” wrote

Treacy and colleagues.

Find Your Value Proposition • 35

THE FOCUS CAN CHANGE OVER TIME

To say you’ve got to have focus for your KM efforts does not mean you

pick one value proposition and stick to it no matter what. Business

changes. The environment changes. Companies grow and shrink, shed

operations and acquire new ones. Markets rise and fall. Flexibility is

key to success in KM as in other areas of business. You certainly need a

focus, but that focus may well change with time and events.

Consider the case of Texas Instruments. When TI first began to tar-

get its best practice transfer efforts, it chose to focus on improving op¬

erational excellence. But as its operations began to near world-class

levels and with its market increasingly focused on new technologies,

the giant multinational nimbly shifted its KM gears to a product-tech¬

nology focus. Cindy Johnson, of TI’s Office of Best Practices, has iden¬

tified how this transition has occurred. (See TI’s Value Proposition.)

KNOWLEDGE, THE NEW FRONTIER

“Know thy customer, then know thyself’ was the central message in

Treacy and Wiersema’s article. Even in 1993 (the pre-KM era, that is),

they recognized that those companies that manage to excel not only re¬

design processes but also place an emphasis on the institutionalized and

active sharing of knowledge across functional silos.

In all of their examples, from GE’s operational savvy to Dell’s deliv¬

ery wits, the basis for leapfrogging competition had been the sharing of

customer, operational, and product information.

• When Kraft decided to make customers it primary focus, it empow¬

ered its sales people by giving them detailed customer and product

knowledge and allowing them to make executive decisions at the

local-store level.

• To launch its successful disposable contact-lens business, Johnson &

Johnson relied on a call from a J&J executive in Denmark who

shared knowledge of a new practice in that market. J&J captured a

big chunk of the U.S. market before any of its competitors could get

their act together.

But while well-managed organizations have realized the need to share

“know-how” before, the mere exchange of raw data and information

36 • If Only We Knew What We Know

TVS VALUE PROPOSITION

On Identifying “What Counts.” According to Cindy Johnson, “The key

message is the focus message. You’ve got to look at your business. You

can’t take KM and just say ‘We are to become a knowledge-creating com¬

pany.’ Before you do anything else, you have got to find the key leverage

points: Operational expense? Customer intimacy? Product excellence?”

On Changing with the Times. “For TI,” continues Johnson, “coming off

some difficult years in the early nineties, reducing operational expenses

was the big-bang win. But it is already changing, to a very product-tech¬

nology focus. We are now applying transfer concepts to innovation and

how to launch new business and new product lines more quickly. In

1997 and 1998, the focus is on the product. Prior to this, it was opera¬

tions. We could not have not have addressed all of those issues at the

same time with any success.”

On Qetting Started. “If you are looking at a knowledge system, look at

where it’s going to bring you the returns,” says Johnson. She has used the

following checklist:

• Identify the value proposition.

• Understand your processes.

• Empower your people through teams and standardized technologies.

“Once that’s done, move forward with a people-based solution that’s sup¬

ported from the outset with technology to promote dialogue and cata¬

loguing of know-how.”

are no longer sufficient. To compete today and into the future, compa¬

nies must up the ante of their collective intelligence; hence databases

must be interpreted and the interpretation (sense making) shared

broadly and effectively enterprise-wide; successful practices and

processes have to transfer quickly, to help launch new businesses and

new products faster—and more effectively—than the competition.

High-value know-how is fast becoming the fuel that feeds the search

for excellence.

Find Your Value Proposition • 37

In the past, inefficient processes and uninformed people made it im-

possible to be good at more than one area. There were not enough re¬

sources to leverage existing knowledge across departments and

business units. With new technologies and new understanding of how

best practices transfer, these barriers to excellence can be removed to

allow companies to respond to their customer demands in ways that

would not have been imaginable only five years ago.

WHO FOCUSES ON WHAT?

A snapshot of some KM pioneers’ early choices of value propositions.

PF = Primary Focus; SF= Secondary Focus

ORQANIZATION Customer

Intimacy

Product-

to-Market

Operational

Excellence

Amoco PF

Arthur Andersen PF SF

BP PF

Buckman Labs PF SF

Chevron PF

Cigna PF

Dow Chemical PF

Hewlett-Packard SF PF

Hoffmann-La Roche PF

Hughes S&C PF SF

IBM PF

Kaiser Permanente PF

NS A PF

Price Waterhouse PF

Sequent PF SF

Skandia PF

TI SF PF

US A A PF SF

World Bank PF

Chapter 6

CUSTOMER INTIMACY

This [KM] system is dedicated to the front line; it is dedicated

to addressing opportunities with customers.

-—Bob Buckman, president, chairman, and

CEO ofBulab Holdings, Inc.

VALUE PROPOSITION #1

Increase revenue, reduce cost of selling, and increase customer

satisfaction and retention.

This value proposition focuses on capturing knowledge about

customers, developing and transferring knowledge and under-

standing of customers’ needs, preferences, and business to in¬

crease sales, as well as bringing the knowledge of the organization

to bear on customer problems.

If you don’t have a customer, you don’t have a reason to exist. Amaz¬

ingly enough, in our fervor to improve processes, some people actu¬

ally lose sight of this fundamental. As Peter Drucker pointed out, the

basic purpose of an organization lies outside itself. The customer may

be a consumer, a business, a taxpayer, a soldier, a manufacturer, a buyer,

or a parent. Forget that and you’re toast.

So value proposition number one is: Customer, customer, customer.

Why would you want to transfer best practices around customers?

38

Customer Intimacy • 39

The obvious answers are to enhance revenue, margins, and customer

satisfaction and retention. Specifically, transfer of knowledge and best

practices can help:

• Present a single face and corporate image to the customer, no matter

how many divisions or products you have.

• Provide “one-stop-shopping” for the convenience of—in particu¬

lar—business customers.

• Give service representatives the information they require to treat

customers as individuals and to answer questions and solve prob¬

lems more quickly, and on the first call.

• Become so intimate with your clients—and so easy to do business

with—that the “cost of exit” becomes too high for your customers to

consider.

• Help sales people grow more efficient and effective at selling, cross¬

selling, and repeat selling.

THE REENGINEERING PARADOX

At first glance, it is surprising that the arenas of sales and marketing

and customer service have frequently eluded the reengineering knife of

recent years, while transaction-based “back office” processes such as

accounts receivable and payable have undergone major surgery. Was it

because we were all so great at customer-focused processes? Not likely.

Or were marketing, sales, and service “stealth” processes that flew

under the reengineering radar? We don’t think so.

We think it was because most senior managers feared tampering

with the revenue stream, or upsetting sales people, when they could

not yet offer the tools or methodologies to truly help. Sales and service

are fundamentally people-based processes, hence perfect candidates

for knowledge-based principles and tools. And those tools are only

now beginning to emerge.

Of course, companies have been using technologies such as decision

support systems to pull customer information from massive data ware¬

houses; some have been linking computers and telephone applications

at call centers. But the “information” contained in such databases is in

no way “knowledge.” It’s raw. It’s basic. It may give service reps an edge

by letting them know when a customer last called and what his or her

40 • If Only We Knew What We Know

five recent purchases were. But in most cases, only top management

gets to see the results of more advanced studies of trends and patterns

(real knowledge), and most systems are incapable of capturing and

sharing “practices,” descriptions of actions, in ways that are usable by

others. That sort of sharing is almost always better handled in person,

in teams, and via competency groups.

A TWO-PRONGED APPROACH

Companies can take two approaches to knowledge and best practice

transfer to strengthen their bond with customers. First, they can “em¬

power” their front-line service employees with the knowledge they

need to effectively handle customer inquiries, complaints, and needs.

Knowledge is power. By giving it to the customer service organization,

the vague term “empowerment” finally gets its meaning.

Second, they can ensure everyone throughout the organization

“knows” the customer, and keeps intimacy as a top priority.

Some of the practical strategies such knowledge seekers use include:

1. Capturing and providing the most effective sales tools

2. Giving call-center and service reps rapid access to all product and

price information

3. Giving all contact persons information about the customer’s pref¬

erences, prior conversations, purchasing history, and the like

(Note: Getting to know your customers also gives you the chance to know

what product and service needs are emerging before your competitors do, and

hence the chance to launch new products and services; more about this in the

next chapter.)

SEQUENT’S “KNOWLEDGE-INTENSIVE” SALES FORCE

At this UNIX systems integrator (see case study in Chapter 17), KM is

focused on providing design, sales, and marketing teams with the prac¬

tices and tools they need to create high-end customer solutions. The Se¬

quent Corporate Electronic Library (SCEL, appropriately pronounced

sell) allows Sequent associates to capture best practices, proposals,

Customer Intimacy • 41

designs, frequently asked questions, and customer and competitor in-

formation and make them immediately accessible anywhere in the

world to any other Sequent team. The vision for SCEL has been: “The

one place to go to find anything you need to help you do your job more

effectively.”

With multimillion-dollar projects and a sophisticated customer

base, Sequent knew that the know-how it brings to each integration

job is its key selling point. It simply cannot afford to put forth an “un¬

informed” sales person.

Each and every individual representing the firm must make cus¬

tomers feel certain that Sequent is on top of the issues and ahead of the

competition. “Our customers cannot tolerate excessive, long-cycle, or

unrecoverable failures, so we have to rule out bad design options, inap¬

propriate implementation practices, unworkable technologies, and so

forth—and keep them ruled out,” says Marc Demarest, the former

chief knowledge officer at Sequent.

BUCKMAN’S DISPERSED PRESENCE

In 1992, $270 million Buckman Laboratories International (for case

study see Chapter 14) launched a system to help sales people in eighty

countries instantaneously share knowledge with everyone in the orga¬

nization, including Buckman’s R&D labs. The result is a better success

rate for new products, and overnight advice to help far-flung sales folks

solve customer problems and make new sales.

Just as important, new ideas and customer solutions get shared

widely, overnight. Bob Buckman recalls a case of a Buckman associate

in Singapore who posted a request for help at 12:05 P.M. Ohio time.

The first response came within three hours from Memphis. Within

forty-eight hours, associates from fifty-two countries responded with

knowledge. Buckman got the contract, not only because of the quality

of the proposal and the quick turnaround time, but because the cus¬

tomer was impressed. By showing the customer the alacrity of re¬

sponse, the associate demonstrated the power of the total organization.

The customer realized that the organization was not reliant on only

one person on the ground in Singapore, but that it had the full support

of the global organization, literally at its finger tips.

42 • If Only We Knew What We Know

ARTHUR ANDERSEN EMBEDS KNOWLEDGE INTO SERVICE

At Arthur Andersen, one of the world’s largest multidisciplinary pro¬

fessional services firms, multiple knowledge bases and information

sources have been integrated into a “one-stop shop” called Knowledge-

SpaceSM for the front-line professionals. The goal is to bring the full

knowledge of the firm to bear on every engagement. “Our mission is to

transform the capability of Arthur Andersen, and our clients, to cre¬

ate, share, apply, and value knowledge,” says Bob Hiebeler, Managing

Director of KnowledgeSpace at AA. “The word ‘capability’ refers to

human beings. And this is what knowledge management is all about,”

he says.

Why manage knowledge? “For AA,” says Hiebeler, “it’s really for

the purpose of increasing customer intimacy; 99 percent of what we

sell is knowledge. Getting close to our clients means transferring

knowledge very effectively and efficiently. As consultants, our value

proposition to our clients equals the combination of personal and or¬

ganizational knowledge.” The more of the latter that is brought to bear

on the customer, the greater the value. “The whole purpose of the

transfer and sharing effort is hence to dramatically increase the knowl¬

edge our people can ‘tap into.’ When you are standing in front of the

client, you have the resources of the entire firm at your fingertips. And

that has impact!”

Just what’s at the consultants’ fingertips? Everything from best prac¬

tices, relevant engagement experience, performance measures, diagnos¬

tic tools, methodologies, tools and techniques, and to presentations, a

hotline list of experts to call for more information and knowledge.

THE RENAISSANCE OF RELATIONSHIP SELLING

The examples of Buckman, Arthur Andersen, and Sequent (plus the

larger USAA case study on page 43) all share a basic business assump¬

tion: The better a firm knows its customer, the more it is likely to help

the customer become successful by shaping products and services to fit

and anticipate his or her needs. The more you can anticipate your cus¬

tomers’ needs, the more likely they are to come back for more and

more expensive offerings.

Customer Intimacy • 43

Indeed, every study that’s ever been done shows that the cost of at'

taining new customers is many times the cost of retaining good cus¬

tomers. In fields as divergent as high technology and chemicals,

relationship selling is making a renaissance. Not because companies

have suddenly turned all warm and fuzzy, but because fierce competi¬

tion and rising “customer acquisition” costs have made companies re¬

alize that keeping their best customers is their shot at decent margins

and sustainable profitability.

In the past, customer intimacy typically meant a high level of seg¬

mentation in the firm’s marketing: knowing which customers bought

what, when, and then targeting marketing efforts accordingly. Such

“know thy customer” efforts are critical, but no longer sufficient.

In order for customers to achieve excellence in their operations,

today’s market leaders must know more about what works, and what

doesn’t—fast and first. They have to apply intelligent solutions quickly;

they have to effectively share their best know-how and practices.

USAA BRIDGES THE GAP TO CUSTOMER INTIMACY

The driving force behind knowledge management at USAA is the

need to satisfy members’ requests in one phone call and continue to

provide them with excellent service. Every contact is a business oppor¬

tunity. To support this notion, USAA has implemented a comprehen¬

sive customer feedback system that quantifies this feedback and

improves overall knowledge of the customer.

USAA was founded in 1922 by twenty-five army officers to provide

auto insurance for fellow officers and their families. Today, the San

Antonio, Texas-based consortium offers property and casualty insur¬

ance, life insurance, banking, investment, and real estate development

services to more than three million customers, including more than 95

percent of active-duty military officers.

USAA’s members are located around the world and conduct their

business primarily by toll-free telephone calls, fax, and mail. USAA

handles approximately 400,000 voice calls per day, and 2.5 million

pieces of mail monthly. The company services approximately 90 per-

44 • If Only We Knew What We Know

cent of its business by telephone, from multiple call centers. The re-

maining 10 percent is serviced by mail.

Distance Leads to Intimacy

This long-distance arrangement creates a natural emphasis on cus¬

tomer service. That’s especially critical for USAA, which is chartered

as a reciprocal inter-insurance exchange—that is, the company’s mem¬

bers are also its owners.

To ensure members/owners are satisfied and telephone contact does

not breed anonymity, USAA has deployed a state-of-the-art knowl¬

edge and transfer system that performs two primary functions:

• It lets every service rep know all there is to know about a customer or

a problem, and provides them with the current “best practices” for

addressing issues.

• It tracks customer satisfaction on an ongoing basis, reveals regional

and other trends, and shares performance measures with each and

every employee so that everyone knows “how they’re doing” all the

time, every time.

Keeping in Touch

Every contact is an opportunity to learn more about customer needs

and expectations. A comprehensive customer feedback system called

ECHO (Every Contact Has Opportunity) helps service reps store and

quantify feedback and improve overall knowledge of the customer.

This feedback system lets USAA maintain and increase an already

high level of customer satisfaction, track market factors and organiza¬

tional considerations on a daily basis, and update each customer-con¬

tact employee on any relevant items.

ECHO’s purpose is to monitor, study, and respond to member com¬

ments. This real-time customer database and problem-support system

helps customer-service reps bring the knowledge of USAA to every

customer’s call.

ECHO also captures types of inquiries and looks for trends. Reps are

provided with cases showing how to diagnose and solve a customer’s

problem, based on prior successful examples and best practices. The

technology automates some aspects of best-practice knowledge sharing.

Customer Intimacy • 45

About 1,500 customer comments are entered per week. The com-

ments are collected and forwarded to the proper agents for action. If

the action requires a change by management, a supervisor can forward

the comments to a senior manager.

The powerful system also allows anyone with access to the USAA

Information Management System to do a basic root-cause analysis on a

transaction recorded there. “We can look at patterns to see if, for in¬

stance, members in a certain area have financial difficulties,” says Tim

Timmerman, executive director of member relations and feedback.

A “Hot Topics Program” asks reps to listen for—and record in a spe¬

cial area—comments on new products or processes. Another part of

ECHO, called “Employee Feedback,” allows reps to enter—anony¬

mously if they wish—any ideas for removing barriers to productivity

inside or outside the rep’s area. (At last count, the average number of

“can-do-better” ideas was seventy-five per week.)

“Although ECHO doesn’t collect numerical data, the anecdotal in¬

formation it contains has tracked uncannily close to what we know

statistically about our strengths and weaknesses,” says Timmerman.

About 6,000 of USAA’s employees currently have on-line access to

ECHO; it will be available to all employees in the future.

Just making ECHO available would not have accomplished

USAA’s goal of creating customer intimacy were it not for the organi¬

zation’s unique culture. “We have some advantages other companies

do not,” says Ross Miles, a business analyst in strategic planning.

“There’s a sense of motivation and of team.” Timmerman agrees: “The

excitement [and] the opportunity this culture provides is that it con¬

stantly questions itself, trying to improve the way it works. The sense

of dynamism that exists here is truly remarkable.”

SO, HOW IS USAA DOING?

So far, very well, thank you. USAA boasts a 98 percent policy reten¬

tion rate—one of the highest in its industry—and attributes a lot of it

to its culture of attention to customer needs.

“USAA is one of the few companies I have ever encountered that

is totally committed to customer service,” says Tom W. Blackburn,

national call center manager at USAA. “With great confidence, I can

46 • If Only We Knew What We Know

send you out into the hall and you can ask anybody what the most im¬

portant thing they do is. They will reply, “customer service,’” says

Blackburn.

The key, says Timmerman, is realizing that knowledge is the lifeline

of today’s organizations. “I use a biological analogy,” he says. “Organi¬

zations are moving from a hierarchical, rigid environment to one in

which they are almost adaptive—the way the human body is. The

body of a person sitting in a meeting is making a multitude of correc¬

tions in circulation, heart rate, body temperature, and so on, to react

appropriately to the messages it is getting from the environment and

from its own systems. For instance, is the room hot? Is it cold? What

are the others in the room saying? An organization ought to be able to

process information instantly and pass it on to the areas that need it in

a similar way.”

Chapter 7

PRODUCT'TO'MARKET EXCELLENCE

The only legal monopoly there is, is to get to market first.

—Rosabeth Moss Kanter

VALUE PROPOSITION #2

By reducing time-to-market, and designing and commercializing

new products more quickly and successfully, we will increase rev¬

enue, retain market lead, and grow our profit margins. This value

proposition is focused on two transfer strategies: (1) ensuring new

ideas and new design from inside and outside the organization are

incorporated into product and service offering; and (2) accelerat¬

ing the product development process by reusing lessons learned

from earlier attempts.

In our earlier sections we talked about the growing importance of

speed as a driver of business change. Speed means there is less time

to procrastinate on new product ideas, less time to rerun experiments

or try out new products before a full launch. If you don’t do it, someone

else will. Once they do, you may not get a second chance.

Speed also means new products become yesterday’s news more

rapidly, and companies cannot rest on their laurels for decades after

47

48 • If Only We Knew What We Know

launching a blockbuster product. If your product development

pipeline is not full of new ideas/products, your company may become

yesterday’s news as well.

WHO’S GOT TIME TO REINVENT THE WHEEL?

No one.

Hence the growing pressure to invent more, faster and better, is

driving a two-pronged knowledge management and transfer strategy.

• Companies must ensure they’re getting the “right” products and ser¬

vices out to their customer base by ensuring ideas and practices flow

throughout the organization, from customer service to R&D, and so

on.

• Companies need to use knowledge to help get products developed

the “right way” by reusing what other parts of the company, at other

times, have learned about getting their product to market—faster,

better, and with better results.

Specifically, transfer of knowledge and best practices in product de¬

velopment can help companies:

• Shorten the time-to-market process (by reusing old practices and

designs)

• Embed the right knowledge and functionality in the product or ser¬

vice (by ensuring ideas and comments flow to the right place—

R&D and/or marketing)

• Consequently, cut the costs associated with successful and unsuc¬

cessful products, allowing more profitable product launches per year

Some examples of know-how and practices that help firms achieve

excellence in product development include:

1. Understanding of the characteristics of good and not-so-good

products and designs

2. Best practices on moving through the product development

process, including how to launch a new product

3. Reusable designs and research

4. Best practices from departed (disbanded) teams

Product'tO'Market Excellence • 49

5. Understanding and use of past experience with regulatory bodies

(who can slow you down faster than a flat market can)

FINDING THE BEST KITCHEN

As we were writing this book in the fall of 1997, we were struck by an

article in Fortune about McDonald’s Corporation. It described the fast-

food giant’s fall from grace; it concluded the once-Wall Street darling

was being eclipsed by arch rivals Burger King and Wendy’s (pun in¬

tended.)

Why? Mostly because McDonald had been putting out product duds

and embarking on disastrous marketing efforts such as the ill-fated

C-55 campaign. That’s a satisfying answer to some extent. But a more

revealing reason for the trouble at the Big Mac maker is found behind

the slew of unprofitable product launches. At root, it appears, are three

separate but related factors:

• The deteriorating relationship between McDonald’s and its fran¬

chise operators

• A highly centralized management apparatus

• An entrenched culture where fifteen-year veterans are still consid¬

ered the new guys on the block

“In the good old days at McDonald’s, most new product ideas came

from the field. The Egg McMuffin, the Hot Apple Pie . .. even the Big

Mac were cooked up in operator’s kitchen,” noted Fortune. “The Cor¬

porate kitchen at Oak Brook, by contrast, has popped out mainly

losers, like McPizza, the McLean and the Arch Deluxe.”

The diagnosis: Combined, these factors have fatally clogged up the

company’s information lifelines. Knowledge, best practices, and good

ideas about new products no longer transfer from the front lines to

headquarters. The results are not pretty.

Fortunatley, McDonald’s new CEO, Jack Greenberg, has turned

things around. After spending two years working a double shift as both

CFO and regional manager of hundreds of McDonald’s stores, Greenberg

knew customers wanted a made-to-order product. He gave authority to

managers in the field, and rolled out the new cooking system, “made-for-

you.” It has worked: McDonald’s shares are up 40% in 1998. The lesson:

50 • If Only We Knew What We Know

in markets as fickle as food tastes, listen to the customers’ real product re¬

quirements and rapidly respond. The inability to transfer or act on cus¬

tomer knowledge can be as unappetizing as cold fries.

STRATEGY #1: GETTING THE RIGHT PRODUCT OUT

By transferring knowledge and know-how from the front-line cus¬

tomer service and marketing processes to the product development

processes, companies can use market and customer information to help

guide the development of product and services.

How do firms do this? One common approach to transferring

knowledge is to bring teams together to share knowledge about prod¬

ucts and customers, and in the process create new and better knowl¬

edge about how and what to bring to market. Another is channeling

information and comments from the sales and marketing force to the

people who do the design and development work. Whether by using

technology or cross-functional networks, the transfer of product expe¬

rience and know-how from one part of the company to the other can

reduce the risk of new product development—substantially.

Introducing a “bad” product or service is a no-win situation. “Our

customers cannot tolerate excessive, long-cycle, or unrecoverable fail¬

ures,” says Demarest, former CKO at Sequent Computers. “So we have

to rule out bad design options, inappropriate implementation prac¬

tices, unworkable technologies, and so forth—and keep them ruled

out.”

The same is true for any firm that counts high contents of expert

knowledge as part of its product or service. Take consulting and pub¬

lishing firms: “Knowledge is our service,” says Paul Pederson, national

director of change integration at Price Waterhouse. “To maintain

competitiveness we have no choice but to create and leverage knowl¬

edge.”

STRATEGY #2: GETTING PRODUCTS OUT THE RIGHT WAY

Companies that have experience with the development of new prod¬

ucts should know what works and what doesn’t. Yet, more often than

not, the experience and learning of past development efforts do not

Product'tO'Market Excellence • 51

make their way in an organized and deliberate fashion to current ini¬

tiatives. The result is costly waste of time and resources.

The payoff to sharing here can be tremendous. Many firms have cal¬

culated just how much money is lost for each day a product is delayed.

At pharmaceutical giant Hoffmann-La Roche (see more detail on page

56), it’s as much as $1 million per day! What’s the opportunity cost for

your organization?

To eliminate unnecessary delays, companies can bring together peo¬

ple who’ve been through the new product development process with

those who are developing the next generation, in order to (1) avoid

prior mistakes, (2) build on market knowledge, and (3) cut cycle time.

And if they methodically record past practices and experiences, they

won’t have to repeat research, trials, and errors.

HUGHES REUSES OLD DESIGNS FOR NEW PRODUCTS

El Segundo, California-based Hughes Space & Communication has

over 5,000 employees and generates approximately $1.2 billion a year

in sales of communications and satellites technology to private and

public sector customers.

In the past, Hughes was able to produce its products in a near “craft”

environment. But as the commercial market exploded and the govern¬

ment market shrank, Hughes found it must cut costs and create a more

structured approach. Technical excellence was considered the top dif¬

ferentiator in communication satellites. These days, with prices falling

20 to 30 percent per year, cost and schedule (production cycle) have

become bigger issues in customer eyes. Hughes has had to adjust its

business to eliminate unnecessary costs but maintain its ability to in¬

novate and design new products.

In the production of spacecraft, 50 percent of the cost is labor, and

50 percent of that is design. Hence the firm could reap big benefits if it

could reduce costs by reusing designs instead of starting from scratch

every time. Indeed, Hughes has calculated a potential cost reduction of

$7 to $25 million per spacecraft if it were able to reuse designs, based

on the assumption that labor costs would be about 90 percent less for

adapting an existing design than for creating new ones.

“Knowledge reuse is a major goal,” says Arian Ward at Hughes. “Our

52 • If Only We Knew What We Know

knowledge is codified mainly in the form of product designs and other

technical documents. We are working to greatly increase our reuse of

this knowledge.”

Hughes relies on a plethora of knowledge-sharing systems to ensure

nothing is invented twice. The Hughes “Knowledge Highway” com¬

bines an intranet, a database of lessons learned and best practices, and

pointers to experts and “human” networks. Editorial teams analyze

captured “knowledge” and best practices before storing them on

shared-access databases or intranets.

TEXAS INSTRUMENTS FREES UP INVENTION RESOURCES

If you spend time and money on rediscovering the “truth” you can’t

spend the same resources on inventing new products and services. At

Texas Instruments, which began its quest for excellence by pursuing

new operational efficiencies, the new goal is to ensure that product de¬

velopment can proceed without delays.

“The driving force behind KM is reduction of cycle time and freeing

up of resources that would otherwise be spent rediscovering knowledge

that others in TI already possess,” says Cindy Johnson at TI. “These re¬

sources can then be focused on achieving innovation, which TI must

have to attain greatness in the twenty-first century,” she says. “Knowl¬

edge is the asset we use to create the next major innovation that will

provide value to TI’s customers.”

NS A REDUCES UNCERTAINTY IN HIGH-RISK R&D

With an enormous range of customers-—and needs—the National Se¬

curity Agency, a federal agency, which monitors foreign electronic

communications to protect the security of the United States, must

demonstrate excellence in the selection of information to which it

pays attention. And turn data into knowledge and into leading-edge

R&D.

To ensure it gets the best ideas, first, NSA has set aside a multimillion-

dollar annual IDEA (Innovative Development and Enterprise Ad¬

vancement) funding pool for high-risk R&D. Its goal is to provide a

simple, fast, and streamlined process for sponsoring exploration of

Product'tO'Market Excellence • 53

technical innovations with breakthrough potential, in an environ¬

ment highly conducive to sharing of ideas and innovation.

Funds are available for expenses such as research and development

contracts, conference fees and travel, other types of travel (data col¬

lection or field testing), technologies, books, and other items. Agency

staff can be released from current duties on a full- or part-time basis to

pursue their IDEAs.

The IDEA program provides a clear and uncomplicated process to

quickly assemble resources and endorsements necessary to explore new

ideas. IDEA creates an environment where people with new ideas and

entrepreneurial inspirations can push the boundaries of technology,

processes, and methods to improve products and services. IDEA is

open to the entire NS A community, and in the United States to:

• Large and small businesses

• Academic and eligible nonprofit institutions

• Government agencies and research centers

• Collaborative ventures from mixed sources

The IDEA program sponsors awardees through technical mentoring

contracts and technical fellowships. For fiscal year 1998, NS A allo¬

cated IDEA funds to support approximately twenty individual pro¬

jects, including in-house, contracted and hybrid efforts. The program

sponsors award projects for up to twelve months and $250,000.

MONSANTO REDEFINES THE COMPETITIVE EDGE

In many industries today, from pharmaceuticals to technology to

chemicals, a firm’s competitive edge comes from its ability to launch

new products.

Says Bipin Junnarkar, director of knowledge management activities

at Monsanto: “Those companies that can deal with and make sense of

incomplete information faster will have a great advantage. Those who

procrastinate will lose significant advantage.”

For Monsanto, the driving force behind the management and trans¬

fer of knowledge has been increasing the “sense making” capabilities of

its employees, particularly in light of the “information overflow” of

past years.

54 • If Only We Knew What We Know

“By managing knowledge, we were able to increase the exchange of

qualitative information and to help reduce the time to bring a new

product to market,” says Junnarkar. “The ultimate value of the KM ar¬

chitecture effort is that it has allowed Monsanto to bring innovations

to market quicker, improve upon the operational efficiency of Mon¬

santo’s businesses, and serve its customers better.”

THE COST OF LOST INVENTIONS

Whereas most firms have neither the time nor the resources to rein¬

vent the wheel, is there a danger that companies’ growing reliance on

reuse and transfer will ultimately “kill off’ innovation even where they

need it most? Just how many potential “innovations” could be “lost”

because people take ready-made solutions based on preexisting de¬

signs?

There is no way to answer this question with anything close to sta¬

tistical accuracy, but we can venture a few thoughts.

First, there’s a difference between innovation and invention. Innova¬

tion, in fact, has to do with adapting existing designs in new ways. In¬

vention means creating something from nothing. In most firms, the

latter is rare and not always possible—or desirable.

Second, is it is truly efficient for every single employee to be engaged in

the process of innovation? There is no question that by relying on best-

practice transfer, companies ensure not everyone in the organization

will be engaged in the process of dreaming up brand-new ideas. But is

it really desirable for everyone everywhere to be innovative? Perhaps

there are areas where innovation is not likely to yield significant ad¬

vantage, and it may take up valuable time better spent otherwise.

Some things really do not have to reinvented, like the wheel (yes, that

tired example).

Does KM sacrifice innovation in the quest for greater reuse? In some

situations, such as the development of automated problem diagnosis

scripts for callcenter reps, it probably does. But if 80 percent of cases

have the same answer, innovation and creative solutions are a waste of

time.

In other cases, companies have to be more cautious about reusing

without rethinking. The creative process does not happen in a vacuum;

Product'tO'Market Excellence • 55

rather, it is the process of taking various existing inputs, and creating

something out of them that did not exist before. To get their creative

juice flowing, people need to interact, learn, talk, read, examine what’s

been done and reexamine it. The “creation” part cannot be auto¬

mated, and it certainly cannot be replicated.

IBM RENEWS THE SPIRIT OF PRODUCT INNOVATION

Holding the largest number of patents in the world, IBM is perhaps the

quintessential intellectual capital enterprise. But Big Blue did not

begin to manage knowledge in an organized, methodical fashion until

the early 1990s, when it began a major overhaul of its operations.

“Knowledge is core to our business,” asserts Kuan-Tsae Huang, a

knowledge management expert with IBM Consulting. And indeed,

when Big Blue decided to manage its intellectual capital assets, it

quickly realized it had been doing so for years, simply in a less explicit

and organized manner.

“We’ve had tools for conferencing and structured discussion since

the eighties,” says Huang. But the creation of the Intellectual Capital

Management (ICM) group was an attempt to institutionalize KM and

make it more formal.

WHAT HAS IBM DONE?

To get its initiative going, IBM launched a variety of efforts, including

a massive Lotus Notes deployment and an ICM intranet. Lotus Notes

is a windows-based collaborative application designed to facilitate

group work by making e-mail, schedule-sharing, database access, and

document collaboration effortless. IBM’s system is organized along

competencies: logical groupings of people and resources that relate to

particular business areas. The competencies cut across IBM’s organiza¬

tional silos. Some are processes—like supply chain management.

Other are topical, like network design architecture. ICM has set a

process in place to identify a competency, its members and leaders, and

set up structured discussion area. In mid-1997, IBM had about 6,000

employees accessing the various competency discussion groups. By

56 • If Only We Knew What We Know

1998 the company expects to have eighty different competencies up

and running.

WHAT HAS IBM LEARNED?

First, you have to show the value of your efforts. Quoting Mr. Huang:

“KM has to be useful. It requires investment. There is no free lunch.

And once you do [invest], you have to show the value, or in one to two

years you are out of a job. It is key to show the benefit.”

Second, you need to have a framework to manage it (that is, knowl¬

edge). Without a management framework, knowledge sharing will not

be sustainable. You need to be able to energize the employees, creating

a vision and a value system that will provide a framework. “The core is

the management system,” says Mr. Huang. And from the outset, that

system should provide incentives (carrots) and sticks. If you make the

work part of the management or certification report, they will do it.

Measuring participation is pretty easy. IBM can measure activity,

submissions of documents, hits, number of IDs issued. “But activity

does not satisfy management for long,” Huang says. “The next ques¬

tion is what is the business impact? How does intellectual capital affect

performance? That cannot be automated. There has to be a separate

tracking mechanism for reporting how the information was used.

What did you use the knowledge for? Competition? Time saving? Rev¬

enue purposes? Making the customer happy? We measure time saving,

revenue increase, and customers’ satisfaction quarterly.”

Third, watch out for deployment of knowledge management initia¬

tives. Deployment of any enterprise-wide KM effort is a big issue for

IBM, at both the software level (having to install a homogenous plat¬

form globally) as well as a language issue (sharing knowledge across

borders, in multiple languages).

HOFFMANN-LA ROCHE GETS IT RIGHT THE FIRST TIME

USING TRANSFER TO IMPROVE PRODUCT DEVELOPMENT

When somebody starts investing mega millions in knowledge manage¬

ment, we know they’re serious. In the case of Swiss pharmaceutical

Product'tO'Market Excellence • 57

giant Hoffmann-La Roche, the commitment to managing knowledge

dates back to 1992 and the start of a program La Roche labeled Right

the First Time.

Right the First Time centers on speeding up the drug approval process

by ensuring that the documentation required by regulatory agencies cov¬

ers all the right areas and leaves no room for doubt and time-consuming

follow-up questions (Amidon and Skyrme, 1997). The company wanted

product development teams in different parts of the organization to

share experiences and best practices from previous development efforts

and documentation processes in order to cut down the time it takes to

prepare documents, gather relevant data, and resolve the inevitable

follow-up regulatory queries.

For La Roche, the incentive to improve sharing was sizable: In its

industry, the opportunity cost of delaying a drug for one day can be as

high as $1 million!

Hence reducing time-to-market is a critical driver of success. Plus,

the competitor that succeeds in gaining approval first is able to secure

patents for new drugs that lock out its rivals from years of fat revenue

streams. Time is of the essence.

Focusing on Documentation

To get its program going, La Roche invested in both technology sys¬

tems and the creation of a full-time project management team, aided

by several hundred part-time contributors. Much of the team’s initial

effort focused on speeding up the regulatory approval process by re¬

viewing past applications, and ensuring that new ones incorporate the

best and brightest practices.

Putting together the documents required for approval of a new drug

is a daunting task, to say the least. Typical documentation can run to

200,000 printed pages. Getting the right information into the applica¬

tion on the first shot, and reusing existing data and research can dra¬

matically reduce the time it takes to prepare these documents. In

addition, by figuring out what research has been done already by them¬

selves or others, the company could cut down on costly and lengthy

clinical studies.

To figure out what La Roche “already knew,” the Right the First Time

team examined 60,000 documents from four recent drug approval

58 • If Only We Knew What We Know

processes. In particular, it looked for follow-up questions from regula¬

tors, to see what sort of information gaps these questions indicated.

Analysis of existing applications showed, for example, that some regu¬

lators ask the same questions every time. Simply making sure the doc¬

uments included answers to those questions on the first shot prevented

follow-up queries.

• As a result of its analysis, the team was able to design logical hierar¬

chies—decision trees—to help put the right information into the

application by following simple logic: What does the customer (reg¬

ulator) need to know? Does the product work? What is its safety

profile?

• The team also developed knowledge maps that help users access

previously written applications and research results, as well as com¬

piling a corporate “yellow pages” which lists experts by area of spe¬

cialty, and includes their CVs and a photograph.

• Finally, by encouraging teams to use prototype documents, La Roche

has given its workers a tool with which to produce, early on, key

data, information, and issues relating to the approval process.

Not surprisingly, management has noted a significant improvement

in the quality of documentation and the speed of development. It now

takes La Roche less time to write and file new-drug approval docu¬

ments. In some cases, project managers report they see a direct link be¬

tween Right the First Time and their ability to come in ahead of

schedule by as much as one to two months! (Based on a conservative

estimate of twenty work days per month, that’s some $40 million right

there.)

Chapter 8

ACHIEVING

OPERATIONAL EXCELLENCE

We cannot tolerate having world-class performance right

next to mediocre performance simply because we don’t have a

method to implement best practices.

—-Jerry Junkins, ex-CEO of Texas Instruments

VALUE PROPOSITION #3

Boost revenue by reducing the cost of production and increasing pro¬

ductivity, and raise performance to new highs.

This value proposition focuses on the transfer of operational

processes and know-how from top-performing business units and

processes to less-well-performing businesses, ultimately improv¬

ing the organization’s overall performance, reducing expenses,

and increasing revenues.

Our third value proposition lies in process and operational improve¬

ments through the transfer of best practices.

What do we mean by business processes and operations? Everything

from the back office to management practices, from production to ser¬

vice delivery. Any firm with many sites performing similar operations

has a big opportunity here. As Jerry J unkins, the late chairman of

59

60 • If Only We Knew What We Know

Texas Instruments, said: “We have world-class operations side-by-side

with others who just don’t get it.”

We already know that just because one part of the organization is

performing at a better level does not mean its “best practices” will

spontaneously diffuse throughout the organization. We have never

seen a best practice that could be transplanted like a begonia. More

like in the case of an organ transplant, the recipient rejects the dona¬

tion if it isn’t recognized as “self.” The implications are that people

have to adapt a practice to their situation; they have to transform it

and add a bit of themselves to it.

DEVELOPING A STRATEGY

Companies that succeed in raising their entire operations to the next

level—and reduce variation in performance levels—know that to

close performance gaps they need a conscious strategy that encourages,

supports, and rewards best practices and knowledge transfer.

When can transfer of best practices bring operations up to a higher

level of performance, whether it be in cost, quality, or cycle time?

1. When an organization has dozens or hundreds of similar opera¬

tions, such a plants, offices, and retail outlets. Think of the bene¬

fit if performing operations even came up to the median!

2. During mergers, when two organizations want to create a true

synthesis by combining best practices: not just getting economies

of scale, but economies of knowledge as well.

3. During strategic alliances and outsourcing partnerships, when

both parties need to share best practices to collectively achieve a

desired result.

These are big-ticket items. In reality, every company can benefit

when one part of its operations shares what works best with another. In

some cases, simply bringing everyone up to a median level yields mil¬

lions in savings and benefits. And as TI’s Johnson has noted, by fine-

tuning operations using available practices and know-how, firms can free

up resources to focus on other areas, such as new product development.

We draw no line between manufacturers and service firms here. For

consulting and other knowledge-intensive service providers, the ques¬

tion of excellence in operations may be expressed a bit differently:

Achieving Operational Excellence • 61

“How can we bring the knowledge of the firm to bear on any client’s

problem, and look really smart to them, even if we’re sending out a

twenty-three-year-old to do the project?” asks Bob Hiebeler, managing

director of KnowledgeSpace at Arthur Andersen. By capturing the

learning from prior engagements, the practices and experiences of

other teams (facilities/units/plants), the best methodologies and

process redesign theories, firms like Andersen attempt to bring each

and every one of their “engagement teams” to world-class levels.

SKANDIA GETS NEW BUSINESS OFF THE GROUND—FASTER

When Swedish Insurance giant Skandia expanded its “points of sale”

from 5,000 to 50,000 in under five years, senior management began

looking for a more effective and efficient manner of transferring

knowledge and increasing its use throughout its global operations.

Core to Skandia’s approach has been the notion of recycling ideas

and experiences on a global scale, in order to create a worldwide base

of “structural” intellectual capital, while retaining local “human” capi¬

tal. That means the firm, as a whole, has worked to develop a set of

common practices and experiences that are homogeneous across re¬

gions and countries.

The result is a “repeatable” process that can be transferred with ease

on a need basis. For instance, when Skandia decides to launch a new

business, the start-up can “borrow” administrative manpower from an

already established business unit in another country to get things going.

By reusing existing expertise and teaching newcomers “best tricks,”

Skandia reduces the start-up costs and lead time while leapfrogging

productivity and quality measures compared to its rivals.

As a result of these concerted efforts to share operational best prac¬

tices across the company’s global business units, the lead time of starting

a new business has shrunk to seven months (vs. an industry standard of

seven years!). “Now,” says Leif Edvinsson, vice president/director, in¬

tellectual capital at Skandia, “it’s possible to complete two start-ups

per year instead of only one.”

TEXAS INSTRUMENTS BUILDS NEW CAPACITY

In 1994, Tom Engibous, now president and CEO of TI and then presi¬

dent of TI’s Semiconductor Group, identified a wide disparity in yields

62 • If Only We Knew What We Know

and productivity among the company’s wafer fabrication sites, each of

which cost between $500 million and $1 billion to build. The wafer

fabs were operating with a high level of variability in cycle time and

yield—some were world class while others lagged far behind. He pre¬

sented his wafer fab managers with a challenge: create a “free fab”—

that is, increase capacity at existing sites equivalent to a free wafer fab

site by identifying and transferring best practices from the firm’s thir¬

teen factories. “We had pockets of mediocrity next door to world class

simply because [one site] did not know what was happening at a wafer

fab in another part of the world,” says Cindy Johnson at TI.

By using geographically dispersed teams, connecting different ex¬

perts together via technology and face-to-face meetings, and systemat¬

ically seeking out what works best in other parts of TI, the managers

lived up to the challenge. A year later, TI boasted increased capacity

equal to a new fabrication site simply by squeezing out inefficiencies

and sharing best practices around the world. They have now repeated

the process two more times, for total savings of $1.5 billion!

WHAT WORKS BEST?

The experiences of TI, Skandia, and Chevron (see Chevron example

just below) illustrate a key point: Even after years of reengineering,

process redesign, and quality improvement, most companies can still

achieve significantly higher levels of productivity from their existing

operations. Not by firing more people. Not by buying more machines.

Not by forcing people to stay later and work harder. The significant

improvements come from allowing people to learn what works best in

other areas and try it out in their own back yard. And by ensuring they

have all the knowledge and experience they require to do their work at

their best level.

DOING CHEVRON’S BEST

USING TRANSFER TO IMPROVE OPERATIONAL PERFORMANCE

Chevron used early successes in best practice sharing to build credibil¬

ity and a business case for more sharing of best practices in key pro¬

cesses.

Achieving Operational Excellence • 63

Once convinced, nothing is as compelling as a CEO like Ken Derr,

chairman of Chevron, who says: “Every day that a better idea goes un-

used is a lost opportunity. We have to share more, and we have to share

faster. I tell employees that sharing and using best practices is the sin-

gle most important thing they can do.”

Here’s a challenge: How does a $42 billion organization, with

37,000 employees in over one hundred countries and multiple, inde¬

pendent business units, each with its own objectives, management,

and financial targets share? How can you make sure an optimal level of

operational excellence across regions and business units is achieved?

Indeed, merely being big does not count as much anymore in terms

of competitive advantage. Information technologies and telecommu¬

nication have enabled small firms to “mimic” large-size ones without

expending hefty amounts of capital. The real advantage of size is in the

access it provides to knowledge. It’s access smaller firms do not usually

have.

“The driver for us was in acknowledging that in a large decentral¬

ized company there is abundant opportunity for learning in a lot of dif¬

ferent places,” says Mike Callaghan, formally a consultant with

Chevron Quality Staff, and currently with Chevron’s South America

business unit . “That is our competitive advantage in being a large

company. It is not, certainly in the oil business here in the United

States, being large in terms of economy of scale.”

Toward Operational Excellence

In the early 1990s, Chevron’s top management was looking for a way

to close its performance gap vis-a-vis the competition, while still mas¬

tering the knowledge and intelligence resources it knew it had in-

house. Chairman Ken Derr and his key advisors began talking about

best practice sharing. “They began talking about the fact that they had

seen and observed great differences throughout the company [in per¬

formance],” says Callaghan.

They also realized that a lot of the knowledge they needed to bring

Chevron up a notch was right inside their own organization. “They

knew that we had all this good stuff inside—we didn’t use the word

‘knowledge,’” says Callaghan. “All this stuff is within our control. It is

all under our roof. If we could just get it spread around.”

The idea was simple. According to Derr: “Sharing best practices is a

64 • If Only We Knew What We Know

way of leveraging the knowledge, know-how, and experience that has

been accumulated in one part of Chevron and using it elsewhere.”

The goal was ambitious. Derr wanted to create a company where

“best practice sharing, reporting units working together, and cross-

functional teamwork are the norm.” By using a combination of hard

and soft approaches, all centering on the transfer of best practices and

knowledge, Chevron intended to overlay its existing operations with a

neural network-like structure, which allows intelligence and know¬

how to flow seamlessly throughout its vast and widely dispersed corpo¬

rate body.

In 1992, Chevron began systematically to track implementation of

process improvements and performance metrics, while building a

global summary of best practices companywide. One initiative was

headed by Bruce Frolich, then vice president of refining, who quickly

set up thirteen teams made up of refinery managers, operating person¬

nel, technical experts, and volunteers. Together, they set out to exam¬

ine “hard processes” like crude distillation and “soft processes” such as

energy conservation with the goal of finding the best ones and helping

others implement them.

Chevron defines best practice as any practice that has proved to be

valuable or effective within one organization and may have applicabil¬

ity to another. Their initial focus was on areas where improvement

could yield short-term and big-time results. For energy conservation,

for example, the team first looked at Chevron’s biggest energy use

processes, such as crude distillation.

Chevron’s value proposition was simply articulated: “By transferring

what works best in the various decentralized units to other units

we can avoid ‘reinventing the wheel’ and duplicating the efforts of

others. ... If we can share best practices—well—Chevron will have a

competitive advantage,” says Derr.

To ground this value proposition in a business objective, Chevron

set clear targets for reducing costs corporate-wide and achieving supe¬

rior total shareholder return to their competitors. Chevron achieved

this goal with an 18.1 percent return, the highest among their com¬

petitors.

Achieving Operational Excellence • 65

Combining Teamwork with Databases

The results of the original thirteen teams’ efforts were placed in Lotus

Notes databases easily accessible to all, and categorized by issues and

expertise. “The database, however,” says Derr, “is merely an enabling

mechanism. The key to progress will be willingness to learn from

others.”

The actual learning process takes place via two main “soft” mecha¬

nisms: networks and best practice teams. The networks are more gen¬

eral in focus, and involve ongoing topical discussions. There are

networks surrounding issues such as safety, planning, refinery opera¬

tions, oil-field management, and training. Some use e-mail to commu¬

nicate, others use collaborative software tools, and still others rely on

“old faithfuls” such as meetings and conferences.

Best practice teams are smaller in size and targeted specifically at un¬

covering and transferring best practices in particular areas. Chevron’s

U.S. refining business currently boasts thirteen such teams. Their work

is credited with improving everything from catalytic cracking to plant

maintenance. Teams meet three to four times per year, usually for two

days, and set aside time to share ideas and lessons learned, as well as

categorize knowledge into various levels of “applicability” from good

ideas to best practices.

Best Practices Yield Best Performance

The impact of this comprehensive effort of ensuring that best practices

permeate the entire organization has been nothing short of spectacu¬

lar. One team, in just four months, was able to identify twelve steps to

monitor and thus conserve the use of expensive catalytic chemicals.

“Their efforts are saving the company several millions of dollars per

year,” says Derr.

Another team in U.S. production compared data on the operation

of gas compressors in oil and gas fields in the United States and

achieved savings of at least $20 million a year by simply using practices

from best-managed fields elsewhere. The energy-use network, which

evolved from a best-practice team on energy conservation, saved

Chevron $648 million in power and fuel expenses by the end of 1996.

And that’s just for starters. “In our three pilot projects in the North

American production company, the variance in performance for those

66 • If Only We Knew What We Know

three processes was significant enough that we thought that by getting

everyone who volunteered to participate in the pilot phase to the me¬

dian level of performance would save or create value worth about $10

million per process,” says Callaghan. “That’s $30 million from just

those three pilots. So the actual level of achievement that is possible is

significant.”

Making It All Work

Deciding to share and actually reaping the benefits of sharing are two

different things. According to Chevron, the key to translating words

into action is in a combination of factors. Top on the list is highly visible

top management support. At Chevron, managers act as sponsors of

change, set clear expectations, and lead by doing. Of course, CEO Ken

Derr’s total support and leadership has been instrumental in instigating

a cultural shift.

Next, Chevron emphasizes that it takes no more than six to twelve

months to achieve success and gain credibility. Such front-end loading,

achieved by picking process improvements that can yield visible

change within a reasonable time, is critical for building credibility.

When new transfer teams are charted, an expectation is set for early

results.

The use of experts in providing continuity and keeping the process

moving, as well as ongoing involvement from process managers in de¬

veloping and implementing the best practice solutions, ensures long¬

term goals that are in line with each business unit’s goals. Team leaders

are usually process experts.

The bottom line is that sharing requires behavior changes at every

level. Operating supervisors have to make the time to meet and talk

about change and share ideas; technical managers must be able to cede

control over R&D by discussing product innovation with marketing

and front-line employees. Chevron has instituted sophisticated mea¬

surement and coaching tools to help change and reward collaborative

behavior, including 360° feedback systems for appraising managers and

supervisors.

The result is a new way of doing work. “If they [employees] have an

initial problem, their first reaction should not be to just sit down in a

closed room and try and solve it themselves. The reaction should be to

Achieving Operational Excellence • 67

go and ask questions, like ‘Who has done this before?’ ‘Who has infor¬

mation on this?”’ says Callaghan. “And unless these people have the

desire or the commission to go ask somewhere—and that is another

issue—how do you ask, where do you ask? But unless they go look for

it, it is very difficult to get them the knowledge or information.”

SUMMARY TO PART TWO

Knowledge in action equals value. But value is a subjective concept.

What is of value to one firm may not be to another. The first step in de¬

signing an effective transfer and KM system is to identify what is of

value to your customers, and hence what is the driver of your own com¬

petitive advantage:

Is it customer intimacy ? If so, you would want to center your efforts on

identifying, capturing, and sharing knowledge and best practices about

customers, developing and transferring knowledge and understanding

of their needs, preferences, and businesses.

Is it leadership in product development? If so, you would want to reuse old

designs, gain access to past experience in development, and make sure

every bit of intelligence that could be useful to R&D flows across de¬

partmental silos and informs new designs.

Is it operational excellence? If so, you would encourage your employees to

find, record, and share best practices in plants, business units, and

other parts of your operations, across business units and geographic

barriers, closing performance gaps, reducing producing costs, and im¬

proving performance overall.

Deciding where to focus your efforts is key. Why?

1. Although the principles of KM and transfer will not vary from

one value proposition to the other, the types of knowledge and

practices captures, and how they would be transferred most effec¬

tively, do.

68 • If Only We Knew What We Know

2. A sharp focus helps target valuable resources on high-payoff

areas.

3. It gets management attention.

4. It gets employees’ attention—and heaven knows they’ve heard

lots of improvement tales before.

5. Competition has never been more fierce nor response time

shorter. We must respond and change fast and leverage every

inch of existing assets.

6. The old-time emphasis on cost cutting is giving way to revenue

growth.

7. Customers are becoming more sophisticated, and if you know

your customers, you can figure out what they value the most.

8. You only have so much energy, resources, and time. Organiza¬

tions only have so much tolerance for change. Spend it wisely.

Just which one of the three value propositions is best for your orga¬

nization will depend on (1) where the greatest potential for improve¬

ment may be and (2) your firm’s competitive strategy. But no matter

what action you ultimately pursue, keep in mind that markets, business

environment, customer preferences, and even your own operations are

dynamic. Adapting your focus when the time comes is as critical as

choosing the first course of action.

We can guarantee that exhortation to “Share more!” will not work.

It takes systems and systematic approaches like internal benchmark¬

ing, mapping the knowledge terrain, capturing and summarizing of

lessons learned, creating new practices. This costs money and time.

And that is why having a clear business case and a value proposition is so

important.

Remember that the best change and improvement techniques in

the world are of little use without a clear idea of the value proposition

and how knowledge and best practices can enhance it!

Part Three

THE FOUR ENABLERS OF TRANSFER

In the preceding sections, we laid the foundation for profitably leveraging

the knowledge in your organization. The most effective way to share

knowledge and best practices is through systematic transfer. The first step to-

ward profitable management of your company’s knowledge assets is choosing

the right value proposition.

Part Three now focuses on creating the most supportive environment for

transfer, by designing and aligning the enablers of transfer: culture, technol-

ogy, infrastructure, and measurement. The following four chapters describe

how each works, and how they all work together.

Infrastructure Culture

1. Plan

4. Scale-up

/ \ a Value M

T Proposition ▼

3. Implement

2. Design

Technology Enabling Environment

A Model for Best Practice Transfer

Measures

.

Chapter 9

CULTURE, THE UNSEEN HAND

Question: What are the three critical factors in knowledge

management1

Answer: Culture, culture, culture.

—Bob Buckman, president, chairman, and

CEO ofBulab Holdings, Inc.

Infrastructure, culture, technology, and measurement are all neces¬

sary enablers; none alone is sufficient. Rather, they must all work in

concert to achieve sustainable success.

But whereas all four are important—even critical—one is perhaps

more potent and more difficult to alter. It yields less quickly or easily to

innovation, because it is so much a function of your past. We are talk¬

ing, of course, about an organization’s culture. Each company has one.

In fact, each division, even each department, has one.

Culture is the combination of shared history, expectations, unwrit¬

ten rules, and social mores that affects the behavior of everyone, from

managers to mailroom clerks. It’s the set of underlying beliefs that,

while never exactly articulated, are always there to color the percep¬

tion of actions and communications. Whereas culture is the more

ephemeral of enablers, it’s arguably the most potent.

If your organization’s natural tendency is to share and collaborate,

all you have to do is eliminate structural barriers and provide enablers

(like technology and facilitators) to allow practices and ideas to flow

seamlessly across time and space. But if your company’s nature is to

71

72 • If Only We Knew What We Know

hoard knowledge, then the best and greatest KM application may not

be enough to alter your employees’ behavior.

PEOPLE/CULTURE ARE THE KEY TO TRANSFER

Why?

First, because learning and sharing knowledge are social activities.

They take place among people.

Second, practices embedded in people, culture, and context are complex

and rich. Descriptions are thin and pale. (Think of the difference be¬

tween a map and the journey itself.) Dialogue and demonstration can

help enrich the learning.

A CULTURAL SELF-ASSESSMENT TEST

A PRO-SHARING CULTURE AN ANTI-SHARING CULTURE

Learning through teaching

and sharing

There are no incentives or sanctions to

promote sharing of information and insights.

Many incentives and support systems work

against this concept.

Communal understanding

through story telling

Staff feel they are constantly fighting the

clock. Little time or attention is given to

identifying lessons learned from projects.

Continuous exchange and

creation of new knowledge—

as experimentation occurs,

people share and learn

Assumptions about projects or activities

are not challenged.

Common areas of interest and

expertise

Individuals are hired and promoted based

on technical expertise.

Common issues and problems;

strong professional ethics

Management and staff are reluctant to talk

about projects that did not work well

(“sharing your failure”).

Personal relationships The different missions and visions of divi¬

sions or departments produce different cul¬

tures that inhibit the transfer of knowledge

and lessons learned.

Culture, the Unseen Hand • 73

Third, to ensure practices and knowledge not only transfer, but

transfer effectively and make a difference, you have to connect people

who can and are willing to share the deep, rich, tacit knowledge they

have.

Across all cultures, mutual obligation is one of the most powerful

social forces there is. Inside an organization, once people start helping

and sharing with one another, the effort becomes a self-perpetuating

cycle. We find that communities of practice and project teams are pop¬

ping up in knowledge-managing organizations like so many mush¬

rooms after the rain. They are the vehicles by which the rich, tacit

knowledge gets shared among people who feel an obligation to help

each other.

In fact, in his study of barriers to transfer, Gabriel Szulanski (1995)

found that one of the strongest predictors of best practice transfer was

the strength of the relationship between the source and recipient. The

potential adopter of a best practice (the recipient) has to believe that

the source is credible and knows what he or she is talking about. The

source has to be generous enough to devote the counseling and com¬

municating time the recipient needs in order to really understand,

adapt, and implement practice.

Effective knowledge management requires creating a supportive,

collaborative culture and eliminating traditional rivalries. Of course,

all this assumes a basic level of modem organizational skills, such as

collaborative work, using problem-solving paradigms, and effective lis¬

tening. Organizations that haven’t mastered these basic skills aren’t

going to be able to do business at all, probably. So knowledge manage¬

ment works best in healthy, sophisticated firms, or in small firms that

have the luxury of starting out with such positive skills and norms.

LEVERAGING THE PEOPLE FACTOR

“Are you information-focused or people-focused?” asks KM expert

Karl-Erik Sveiby of Queensland, Australia. “The choice of actions and

the investment decisions that you will be comfortable with will be very

different.”

Both focuses can be called “knowledge management,” but ulti¬

mately, says this expert, only one of them will be successful. “I believe

74 • If Only We Knew What We Know

that only a people-focused approach to KM will be competitive long¬

term, because at the end of the day, IT is readily available to everyone

and any IT solution can easily be copied by the competition. Compet¬

itiveness rests in the tacit, not in the explicit.”

Knowledge management practitioners are in agreement with

Sveiby’s premise. “In managing companies,” says Roger Swanson,

manager of corporate research services at Sequent Computers, “you’ve

got to focus on values from the get-go. It’s not unlike raising children.

If you don’t pay attention to the values in the early years, you’re prob¬

ably not going to like them very much when they become adults.”

Even at companies that installed state-of-the-art IT infrastructure,

the human aspect remains paramount. “We always run into the ques¬

tion, ‘Is this work about databases?”’ says Dave Ledet, director of

shared learning at Amoco. “To some degree it is, but it is also very

much about establishing networks of people across plants who share

knowledge.”

At the APQC, we’ve known for a very long time that a company’s

success is a function of the people that make it work. Call it organiza¬

tional know-how, intellectual capital, or structural knowledge. Now,

there is more recognition and empirical evidence to support this per¬

spective.

A 1997 global study of the world’s most admired companies, by For¬

tune magazine and the Hay Group, revealed a critical determinant of

organization success. The study found that the world’s elite organiza¬

tions share one thing in common: They don’t claim their people are

their best asset. They act on it.

The world’s most admired companies (among them Boeing, Mi¬

crosoft, GE, FedEx, Pfizer, British Airways) are companies that take

mission statements seriously. They attract successful people, use in¬

tense testing to pick the best of them, and then provide intense train¬

ing to make sure they stay the best. These organizations view career

development as an investment, not an expense; they promote inter¬

nally, reward top performers, and make sure that their employees—

these key stakeholders—are satisfied with their work. By so doing, they

inspire a culture that recognizes the importance of people. And by rec¬

ognizing the importance of people, they lay the groundwork for the

critical task of leveraging their employees’ experiences and practices.

Culture, the Unseen Hand • 75

THE KM MANAGER’S CULTURAL OVERHAUL “TO-DO” LIST

One could expect firms that have gone through numbers of manage¬

ment improvement initiatives to be skeptical of another new change

effort; our research indicates that the opposite is true. And it’s pretty

easy to see why. Companies that have done a good deal of process im¬

provement and quality work often already have in place the cross¬

functional communication and collaborative settings that are critical

to sharing knowledge and know-how. (They also have other aspects of

the infrastructure in place, as we will discuss in Chapter 11.)

ORGANIZATIONAL CAPABILITIES SUPPORTING TRANSFER

Through the 1980s and 1990s, competitive organizations spent much

of their time developing capabilities that set the stage for a successful cul¬

ture of sharing, transfer, and change. These capabilities include:

1. a process improvement orientation

2. a common methodology for improvement and change

3. the ability to work effectively in teams

4. ability to capture learnings

5. technology to support cataloguing and collaboration

Still, if you find yourself on the anti-sharing side of the table, or you

believe your company’s culture is one that’s unlikely to nurture shar¬

ing, do not stop reading here! Cultural changes are possible, albeit diffi¬

cult. To make them happen, there are at least six things you can do:

1. Believe People Want to Share

Call us eternal optimists, but from our experiences with a multitude of

companies we know this to be true:

• People like to see their knowledge and expertise used, if we haven’t

created negative consequences

• People want to help their colleagues

• People want to learn from others they trust and respect

76 • If Only We Knew What We Know

2. Prepare to Lead by Doing

Actions speak louder than words. And nowhere do they speak more

loudly than in the arena of changing basic perceptions. The only way

to make employees “believe” your vision is to act it—day in and day

out. The impetus for success comes from the top.

Buckman and other KM and transfer pioneers strongly believe se¬

nior management must participate and lead by sharing to nurture the

right cultural tone. For example, Buckman management avoids pro¬

moting anyone who is not recognized for sharing knowledge.

At Chevron, chairman and CEO Ken Derr believes it is senior man¬

agement’s role to see that people do not confuse power building with

knowledge hoarding. Says Derr: “I think that a CEO should lead by ex¬

ample. This means participating and showing that you are personally

committed to learning and the process of change. Whenever I visit

with employee groups, I tell them sharing and using best practices is

the single most important thing they can do.”

“Knowledge management implies knowledge sharing and democ¬

racy, and if knowledge is a source of power it will be hoarded rather

than shared,” points out Onno van Ewyk, of HCI, a consulting firm in

Sydney. “When executives create ‘knowledge fiefdoms’ this works

against the interests of the company as a whole.”

“Probably the most important thing that has allowed us to build a

sharing strategy is leadership support,” says Amoco’s Dave Ledet. “Our

president, Bill Lowrie, is adamant about Shared Learning. He talks

about it constantly in his speeches and presentations. We are very for¬

tunate in that we did not have to get leadership support. It was already

there. That is what has allowed us to move forward at the speed we

have.” (See The view from the corner office, below.)

For many executives, the tallest challenge in instituting cultural

changes may be giving up the notion that accumulating knowledge

and keeping it from others is a source of power. In modem organiza¬

tions, the most powerful individuals are those who inspire and lead by

sharing. Instead of “command and control,” executives have to “men¬

tor and inspire.”

Sometimes top executives truly want to share, but simply don’t

know how. At Texas Instruments, Cindy Johnson has been spending a

Culture, the Unseen Hand • 77

lot of her time talking to top managers about how to share the lessons

of their daily experiences with “the troops.”

3. Rely on the Twin Forces of Capitalism and Democracy

As a nation, we swear by them, both capitalism and democracy. We

live by them. We even fight for them. As companies, we rarely do.

“The biggest problem in most organizations is that they are centrally

planned economies,” Professor Hallal of George Washington Univer¬

sity noted in a personal interview.

Creating a free-enterprise system which accurately reflects economic

reality, on an intrafirm basis, breeds accountability. Accountability dri¬

ves the need for knowledge. When people need the knowledge, they

tend to trade it and share it.

THE VIEW FROM THE CORNER OFFICE

Many executives from companies profiled here have been quite vocal

about the power of shared learning and transfer:

Our behavior is driven by a fundamental core belief: the desire and the

ability of an organization to continuously learn from any source, and to

rapidly convert this learning into action is its ultimate competitive

advantage.

—Jack Welch, CEO, General Electric

Shared Learning is an amazing concept. It allows the corporation to cash

in on what we already have invested in: the knowledge and experience of

our talented people. It gives us the ability to continually harvest a crop of

innovative solutions and creative applications. It’s essential that we accept

the challenge of figuring out how to make the necessary investments so

that Shared Learning will secure a place at the core of Amoco culture.

—William Lowrie, president, Amoco Corporation

The most powerful individuals will be those who do the best job of

transferring knowledge to others. .. . Frankly, I do not think you can

have a successful knowledge transfer effort without that proactive

entrepreneurial support from the top.

—Bob Buckman, president, chairman, and CEO, Bulab Holdings, Inc.

78 . If Only We Knew What We Know

SEQUENT’S KNOWLEDGE ECONOMY

Nowhere is the notion of a free-enterprise knowledge system more

clearly illustrated than in the “knowledge economy’’ transfer system es¬

tablished by Sequent Computer Systems, Inc. “It always amazes me that

we want capitalism everywhere but inside our own firms,” notes Marc

emarest, ex-CKO of Sequent. Sequent used two overriding metaphors m designing its transfer initiatives:

1. “We are supporting and nurturing a knowledge economy, and we want it to be a capitalist one.”

2. “We are managing a ‘city' of knowledge, and as city managers we

are not in the business of telling merchants what they ought to be selling.”

Sequent does not offer specific rewards and incentives to transfer sys-

tem participants- “In our knowledge economy, we encourage and reward

pu hshers, by using their materials and most importantly by attributing

expertise' to those publishers when use events produce desirable eco¬ nomic results for the user,” explains Demarest.

The value and relevancy of knowledge is always in the eye of the be¬

holder. Is information relevant? Relevant to whom? Under what cir-

cumstances? For what purposes? The whole point of KM is that no

one most definitely not some wonk with the title CKO in a headquar-

ers office somewhere-is in a position to understand the hundreds of

different, specific contexts in which knowledge is required to satisfy a

customer or business partner,” says Demarest. “We teach producers to

think about their consumers: their needs, their roles, the processes in

which they participate; we teach consumers to reassemble componen-

.zed knowledge into precisely the tools they need at precisely the time

they need them, and we teach them to discipline publishers who produce poorly formed, unusable, or toxic components.”

“We see ourselves as a knowledge economy within a city of knowl-

e ge producers and consumers meeting and exchanging value in a (largely

romc) marketplace,” explains Demarest. Whereas such laissez-faire

approach may not work in an organization that is accustomed to planned

economics, this model of distributed responsibility has proved so far to be the only one that works within Sequent.

Culture, the Unseen Hand • 79

4. Develop Collaborative Relationships

Collaborative relationships enable tacit knowledge and high-value

practices to transfer. That’s the stuff you cannot put on the informa¬

tion systems. From teams to communities of practice, these basic units

of innovation and collaboration bring together like-minded people

and people with a shared goal. Projects to transfer best practices, such

as the Chevron Best Practices teams, are one vehicle to create these

relationships.

5. Instill Personal Responsibility for Knowledge Creation and Sharing

If people are the engine of knowledge, they should be responsible for

identifying, maintaining, expanding, and sharing their knowledge

base. The profitability and viability of enterprises are directly related to

the degree to which people are able to act intelligently in all situa¬

tions. A successful future depends upon the extent to which each indi¬

vidual builds, shares, and applies his or her knowledge.

ON SPACE AND SHARING

Sometimes, just making sure office space and architectural design match

the desired collaborative atmosphere can go along way. One sure way to

tell the cultural bent of an organization is to count the number of shut of¬

fice doors and the available “communal space.” If there is no “coffee

room,” the chances that people will informally interconnect are slim. If

most doors are closed, you may rightly wonder what people may be hid¬

ing, or hiding from.

Soon after forming, the World Bank’s Education Knowledge Manage¬

ment System (EKMS) (see case study in Chapter 16) had the opportu¬

nity to move to an open office space designed to stimulate team building.

The staff agreed to move in for a six-to-eight-week trial period to see how

the alternative environment affected their work. The new area had no

walls, it organized the staff in teams, and put them close to each other; it

also included several common areas. “The open office has made a real dif¬

ference,” says Martha Pattillo-Siv, coordinator for the Education Advi¬

sory Service. “There is a lot of spontaneity in this atmosphere.” The

upshot: The staff is not asking to go back to the old office arrangement.

80 • If Only We Knew What We Know

This philosophy is in line with a new type of social contract be¬

tween employees and the firm: Employees are responsible for their

growth, and the firm is responsible for providing them with an oppor¬

tunity and the tools to grow. This contract has to be two-sided. If you re¬

quire personal responsibility but provide nothing in return, employees

are going to balk at cooperating, as well they should.

For example, at USAA—a firm believer in personal responsibility

for knowledge—displaced employees are retrained, not fired. Turnover

is less than 6 percent for the whole organization. In return, USAA ex¬

pects employees to make themselves “continuously employable.” And

it works!

This “pull” philosophy puts the responsibility with employees to

TI’S THREE-TIER APPROACH TO CULTURE

Some companies know they have to start from scratch. Others are lucky

enough to have the basic cultural ingredients. When Texas Instruments

began its best-practice transfer initiative in 1994, it was certainly among

the lucky ones.

“My job has been really relatively easy,” confesses Cindy Johnson of

TI. “We have had strong leadership in place, as well as a strong process

of teaming and a true customer focus. We had the culture to begin with.”

Even so, it took work, and a combination of senior management and

transfer team efforts to ensure culture and knowledge management fit

snugly together.

According to Johnson, there are three layers to the development of a

sharing culture: “You have to provide purpose and motivation; you have

to provide the tools and finally, you have to complement both with a re¬

ward and incentive structure.” (These tools are detailed extensively in

the TI case study in Part Four.)

TI s transfer efforts took off in earnest in 1994, after top management

formulated a new vision and initiated a focused movement toward build¬

ing a companywide sense of purpose. That vision, for leadership in digital

solutions to drive the network society, resonated throughout TI’s global

operations. “It’s been our battle cry,” says Johnson. “And it has driven

the effort to create new business and adopt new technologies more

quickly.”

In addition to formulating the grand vision, “the role of top manage¬

ment is to model the behavior,” says Johnson. “CEOs spend the bulk of

Culture, the Unseen Hand • 81

seek out knowledge to improve the performance of themselves and

their processes. According to all our research at the APQC, this “pull”

approach to knowledge sharing is essential for long-term success.

While some knowledge and practices can be “pushed,” the active quest

inspired by “pulling” indicates a desire for learning which is a prerequi¬

site for successful transfer (more on these “prerequisites” in Part Five of

this book).

6. Create a Collective Sense of Purpose

Communities of practice and project teams share on a “micro” basis to

help meet common needs and goals. Similarly, company-wide sharing

their time learning and connecting with other people.” “They talk to cus¬

tomers, financial organizations, government/regulatory bodies, other

business leaders to learn about what works and what doesn’t and under¬

stand the business environment. Basically, they should want all of their

people to do the same for their respective areas.”

Easier said than done. Whereas most senior managers practice knowl¬

edge management in “private,” they rarely see the benefit of publicizing

their daily activities. “They should make it very obvious that what

they’re doing is really about learning and applying the learning to the

business,” says Johnson. “Most of them think this should be very appar¬

ent to everyone. But it isn’t!”

Often, the failure to share is not intentional. “People who are very

good at something are frequently the last ones to show up when they’re

most needed,” notes Johnson. For example, a thin person is unlikely to

attend a weight loss class; however, he or she may have best practices in

weight loss. And sometimes people don’t know what they did to become

good. They need help recognizing and understanding how they’ve be¬

come successful before they can share their best practices.

While TI’s top management was busy creating a vision and leading the

transfer initiatives, Johnson’s group was charged with providing the sup¬

portive systems and creating reward schemes that encourage transfer.

“It’s a fine balance,” admits the veteran transfer expert. Rewarding trans¬

fer activities specifically may work initially. But ultimately, the organiza¬

tion must change its overall promotional compensation scheme to align it

with new values.

82 • If Only We Knew What We Know

must be inspired by a common goal. A vision. A sense of social and or¬

ganizational purpose that defines the raison d’etre of the entity. Knowl¬

edge and practices are meaningless if not put in context of “purpose”

and surrounded by a set of overriding values. Without them, it be¬

comes nearly impossible to enlist people’s cooperation and enthusiasm.

“We’re here to make money” is not good enough. The common goal

must be both more specific than that (in terms of industry/product

leadership), and more inspiring than that (including some social rele¬

vancy). Working together to achieve a common objective, whether on

a division basis or on a global, enterprise-wide basis, for a common goal

is the safest way of ensuring a culture that’s bent on sharing.

TO REWARD OR NOT TO REWARD, THAT IS THE QUESTION

In all three of APQC’s large-scale studies on knowledge management,

we reached similar conclusions about the role of rewards: if the process

of sharing and transfer is not inherently rewarding, celebrated, and

supported by the culture, then artificial rewards won’t have much ef¬

fect, and can make people cynical.

Reflecting the importance of embedding knowledge management

tools and transfer processes into the work itself, we continue to find a

greater use of intangible versus tangible rewards for engaging in transfer

activities.

A good transfer system should provide intrinsic rewards to the pro¬

fessionals who use it. For example, does such a system let its users bet¬

ter, more easily, and more efficiently achieve their project and work

objectives; do they receive more peer recognition as key contributors

and experts; and is their work faster, richer, and more rewarding?

Explicit rewards and incentives go only so far. After years of study¬

ing corporate incentives schemes, we’ve reached the following firm

conclusion: If the practice helps people do their work, they will share.

* The World Banh, the Africa Live Database provides incentive to

update data because analysts become more effective.

• At Sequent Computer Systems, Inc., sales and marketing teams know

they can quickly find the best advice, sales presentations, system so¬

lutions, and customer and competitive information by accessing the

KM system.

Culture, the Unseen Hand • 83

• At Arthur Andersen LLP, consultants can rapidly get the training

and content they need to be successful in client engagements.

Think about e-mail. It has exploded onto the scene and caught like

fire both at work and at home. Does anyone offer workers big rewards

for using e-mail? No. They know intuitively that it helps them do their

work better. Telling everybody that between 5 and 6 P.M. they should

be contributing knowledge, and that if they do, they will get a mouse

pad, won’t change the way people work.

Not surprisingly, we find that only a minority of firms use formal fi¬

nancial rewards to promote sharing behaviors. Instead, successful firms

place a big focus on a personal responsibility strategy, and embedding

knowledge development and transfer into their employees’ profes¬

sional and career development systems.

Price Waterhouse is a good example of how a consulting firm can

encourage knowledge creation and transfer. In the past, the blue-chip

accounting firm’s promotions were based on seniority and tenure, not

excellence and mentoring. In the mid-1990s, Price Waterhouse added

knowledge sharing to its performance appraisal system to ensure that

employees’ efforts to share with others are recognized in their career

path and compensation. Employees must be able to produce “evi¬

dence” of actual knowledge sharing such as tutoring/training, develop¬

ment of methodology, publishing and presenting on topics, coaching

and mentoring, and so on.

RECOGNITION AS THE CURRENCY OF CHOICE

People want their expertise and knowledge to be used—and acknowl¬

edged. Individual recognition is how organizations have traditionally

met this need, and that will continue. But if companies want to en¬

courage sharing, they need to recognize and celebrate that behavior as

well. Texas Instruments created the NIHBIDIA Award: Not Invented

Here But I Did It Anyway.

Begun in 1996, TI’s annual sharing day (ShareFair, where all the

best practice teams man booths to publicize and answer questions

about their practices) culminates in an award ceremony for those orga¬

nizations that have most successfully shared best practices and knowl¬

edge—and produced great results. Both organizations (and sometimes

84 • If Only We Knew What We Know

there are more than two involved) receive an award from senior exec¬

utives for collaborating on the exchange of best practices. This is a

highly prestigious award at TI, because it reinforces both the process

and the results.

STRUCTURING A REWARD SYSTEM: A CHECKLIST

When creating your own knowledge management and transfer reward

and recognition system, keep these points in mind:

1. Recognition lies in being perceived as an expert by employees

and management. Ensure that an internal expert’s name is at¬

tached to documents, guidelines, and presentations they created.

2. Using the knowledge system has to be self-rewarding to the con¬

sumer; users have to get something out of it, be it knowledge they

need or a sense of status and recognition.

3. Time to use and create knowledge has to be recognized and re¬

warded; if participants feel that have to “steal” time from the

“real” work to do this—they won’t.

4. Formal rewards may be demeaning to professionals; don’t give

cash bonuses to people motivated by a sense of involvement and

contribution.

5. Create recognition for transferring and using best practices; you

can do that by celebrating best practice success stories, and prop¬

agating tales of big savings and important contributions.

6. Recognize both parties or units involved in the transfer; at any

given time, an employee is contributing or receiving knowledge.

If both ends are not feeling rewarded, you’ll run out of content

pretty quickly.

7. A “standardized” reward system will help institutionalize the

practice into the common culture.

To repeat, if the transfer activity helps employees achieve their

overall goals, they will pursue it. Hence rewards and recognition may

be healthy and useful in the early stages of building enthusiasm for

transfer. However, in the long run and for a sustainable effort, employ¬

ees have to find the work itself rewarding.

Chapter 10

USING INFORMATION

TECHNOLOGY TO SUPPORT

KNOWLEDGE TRANSFER

Technology has given us new ways to keep in touch with each

other and share information. We have to take advantage of

that technology to get better, faster.

—Bill Baker, Texas Instruments

It is no coincidence that information technology (IT) has blossomed

at the same time that knowledge is becoming recognized as the most

valuable of a firm’s assets. There is a powerful synergistic relationship

between KM and technology; that relationship drives increasing re¬

turns and increasing sophistication on both fronts. As information

technology has become our personal desktop tool and our link to each

other, we have grown to covet even more access to information and

other people’s knowledge. In turn, we demand ever better and more ef¬

fective IT tools, ones that become part of the way we work.

A SYNERGISTIC RELATIONSHIP

We use the term synergy; other experts have other ways to describe the

relationship between technology and knowledge management. Noted

85

86 • If Only We Knew What We Know

KM expert Karl Eric Sveiby has described IT systems as “hygiene fac-

tors.” “IT is for KM like a bathroom is for a house buyer,” says Sveiby.

“Essential because without it the house is not even considered by buy¬

ers. But the bathroom is generally not the vital differentiating factor

for the buyer.”

Hygienic or synergistic, take your pick. The key is that the rise of

distributed technologies like intranets and Lotus Notes has had critical

implications for knowledge managers—and not all of them are posi¬

tive. (See page 93). On the one hand, IT has reduced the cost and sped

up the process of transferring best practices and knowledge. On the

other, IT has led, in many cases, to a flood of information (not knowl¬

edge—note the distinction here!) that has seriously overloaded em¬

ployees’ capacity to make sense of their environment.

“Technology has made it realistic to globally share knowledge,” says

Cindy Johnson at TI. “Before, it was very expensive, slow, and tedious.

We could create enormous repositories, but we would never get any¬

thing out of them.” With search engines and browsers, the process of

delving into knowledge repositories has become both simplified and

faster.

At the same time, however, these technologies have exponentially

increased the amount and speed of information, often leading to orga¬

nizational paralysis—a disease that occurs when a swelling amount of

information is chasing a shrinking number of brains. When the human

capacity to absorb and make sense is exceeded, managers cannot make

decisions and employees don’t know where to get quick answers to

their customers’ questions.

“Over ten years ago, when electronic information just started to

show up from external and internal sources, the initial problem was

getting access and organizing the information,” says Bipin Junnarkar at

Monsanto. But not anymore. Now, the problem is getting the right in¬

formation to the right people, at the right time.

So, while the explosive growth and ready adoption of Internet and

intranet technologies has been an enormous catalyst for knowledge

sharing, just throwing IT solutions at the knowledge management

challenge is not going to do it. IT makes connection possible, but does not

make it happen. Buying systems and implementing state-of-the-art ar¬

chitectures does not, in itself, guarantee the sharing of best practices

Using Information Technology to Support Knowledge Transfer • 87

and know-how. Sometimes, too much technology or the wrong type of

technology can hurt KM efforts.

SHARING-ENABLING TECHNOLOGIES MAY NOT BE NEW,

BUT...

There’s nothing totally new about the use of technology to help

“spread the word.” Communications technologies have been helping

us share stories, events, and experiences since our tribal days. Even in

modem time's, we’ve been leveraging IT to manipulate and share

knowledge, going back to the telephone, faxes, and even Artificial In¬

telligence in the 1950s. So what’s changed?

The answer is in the pervasive use of groupware and Internet/in¬

tranet technologies.

Every new “technology” follows a similar path: One person buys,

say, a color television. Then the next. Then the whole neighborhood.

And before long, no one you know still uses a black-and-white TV.

Tools that work become pervasive. Once they do, their “installed base”

perpetuates and enhances their usefulness. What use is a telephone if

you’re the only one who has it? Same with Lotus Notes and intranet

technologies—but with an added benefit: The telephone, or even

e-mail, does not “record” and “detail” our practices and experiences.

The stories are told. And once they’re told, they are gone. They can¬

not be searched, consulted, remembered, shared, and expanded. The

reasons intranets and Lotus Notes work for knowledge management is

that they combine organized retrieval with ubiquitous access. Re¬

trieval and access. That’s half the solution.

The other half is not technology—it’s people. So although this

chapter focuses on IT, we have to reiterate that technology is not the

solution; it is part of the solution. Technology is necessary but not suf¬

ficient to make transfer happen. The popular slogan is “Build it so they

will come.” Our slight amendment is: “If you build it, they will come,

but only for e-mail.”

If you are going to be successful, you need the technology in place; it’s

got to be good and it’s got to be easy to use. However, as our friend Tom

Davenport says, “If you’re spending more than one third of your energy

on the technology side, you’re probably not going to be successful.”

88 • If Only We Knew What We Know

Ultimately, knowledge and best practices are in people’s heads. The

behavioral aspects of the system are therefore more important than its

architecture. “The world is littered with the remains of KM programs

that companies built and then nobody came,” says Davenport. You

don’t want to be among them.

KM/IT RULES OF THUMB

Tom Davenport uses the “stay under a one-third of resources” litmus

test to ensure IT does not become “the be all and end all” of KM. We

offer two other helpful rules of thumb:

1. The more “valuable” the knowledge, the less sophisticated the technology

that supports it. Here’s how it works: Databases and datamining

tools, for example, are high on the technological sophistication

scale. The knowledge they contain, however, is truly low-grade.

In fact, databases do not contain knowledge at all. They contain

data. In contrast, help desks, equipped with nothing more than

humans and telephones, are low-tech but offer a very high

knowledge value. Hence, the higher the grade of knowledge, the

lower-tech the solution. The two are inversely correlated.

2. Tacit knowledge is best shared through people; explicit knowledge can

be shared through machines. Or, the more tacit the knowledge, the less

high-tech the solution. If you take the continuum between tacit and

explicit knowledge (from totally tacit, poorly organized, some¬

what documented, to highly documented and organized explicit

knowledge), the more explicit the knowledge, the more it lends

itself to high-tech solutions. Tacit know-how, meanwhile, is

often best transferred via people or “help desks.” For example,

when the World Bank began to organize a sharing mechanism for

its tacit and poorly organized explicit knowledge, it set up discus¬

sion groups and help desks to help transfer best practices, instead

of trying to document them in some mega database.

Using Information Technology to Support Knowledge Transfer • 89

BUILDING THE TRANSFER PLATFORM

Let’s review what we’ve said so far:

• IT and KM have a symbiotic relationship.

• Companies must be careful not to confuse databases with knowledge

management. New technologies are certainly enabling and catalytic

(in particular Notes and intranets), but they are not a solution in and

of themselves.

• Technology may not be the most important component of KM, but

try doing without it and you will quickly discover the limits of lunch

clubs and informal get-togethers.

• Finally, IT has to be used with intelligence, matching knowledge types

and needs with the right IT applications.

You may not need the latest and greatest IT to get your transfer ef¬

forts going. But you will need a standardized company-wide architec¬

ture to ensure the sustainability and scalability of those efforts. Perhaps

unlike any other area of technological design, knowledge and best

practice sharing cannot take place if companies allow the proliferation

of separate systems and “IT archipelagoes”—i.e., department-specific

programs. If a firm does not have the architecture of an organization-

wide solution in mind when designing local KM/IT solutions, then the

organization will, over time, face problems in integration and scalabil¬

ity; subsequently, it stands to lose much of the leverage knowledge

management can create.

KNOWLEDGE-ENABLED INTRANETS

Most companies today have already figured out their basic information

technology architecture for other reasons (such as the Year 2000 prob¬

lem or enterprise-wide computing needs). Typical choices involve

corporate intranets and Web-based technologies, supported by collab¬

orative groupware and database applications. However, whereas most

of this basic IT infrastructure is either already in place or rapidly com¬

ing on-line inside organizations, the real challenge is in deciding how

the actual knowledge management applications will be used and main¬

tained.

90 • If Only We Knew What We Know

Here are a few helpful guidelines to help ensure your intranet-based

KM solution achieves its purpose:

1. Understand the business purpose for what you are trying to ac¬

complish—define the objectives. ,

2. Determine whether the current technology can be adapted or

can be purchased off the shelf.

3. Assess the ability of the physical infrastructure currently in place

to handle the kind and quantity of traffic moving around on the

intranet.

4. Identify internal support requirements for maintenance of the

system.

5. Organize your content; there may be a need for librarians.

6. Choose a central location for your Web site, from which one can

navigate to any other area. If you select a decentralized location

instead, think about issues surrounding policies and procedures,

training, and redundancies in processes and information.

7. Design for ease of use—intranet and Web-based solutions are al¬

most training-free in terms of the technology.

8. Facilitate universal access and universal use.

9. Consider initial costs, including shareware and software, as well

as secondary costs, which can include training, operations and

management, increased bandwidth installations, lost productiv¬

ity due to obsessive use of Web content creation tools, and fruit¬

less forays into Web junkyards.

Few of the decisions associated with intranet implementation are

brand new. So practice what you preach and don’t cast off learnings

from implementing on-line transaction processing systems or decision

support systems. They all apply here.

INTRANETS EMPOWER SHARING EFFORTS

At Arthur Andersen, KnowledgeSpaceSM (an intranet-based applica¬

tion) now integrates various legacy databases and information sources

into a “one-stop shop” for the information used to perform AA’s busi¬

ness, including content from a Global Best Practice database.

Using Information Technology to Support Knowledge Transfer • 91

THE BENEFITS OF K-ENABLED INTRANETS

The benefits of knowledge-enabled intranets are numerous. They include:

• Lower communication costs, driven by reducing expenses related to

printing, mailing, and processing of documents.

• Improved productivity by making information more widely and

quickly accessible.

• Higher team productivity, created through collaborative work envi¬

ronments.

• Rapid implementation as a result of open protocol standards.

• Relatively low costs for hardware and software.

At National Semiconductor, the knowledge-enabled intranet began

as a grass-roots effort with the development of its first Web pages in

August 1994. Since then, it has expanded beyond departmental home

pages to encompass enterprise-wide services such as employee direc¬

tory services, teams and communities of practice, research and devel¬

opment findings for new product development, and staffing/career

opportunities. The intranet is also used for day-to-day activities such as

scheduling conference room usage and placing orders ranging from of¬

fice supplies to assembly/fab supplies.

At Buckman Laboratories, web-based forums and the global infra¬

structure provided by CompuServe have allowed associates to access

project tracking systems, customer relationship management systems,

groupware, bulletin boards, virtual conference rooms, and databases

that capture institutional memory twenty-four hours a day.

At Sequent Computer Systems, the Sequent Corporate Electronic Li¬

brary (SCEL) is implemented as a combination of services using database

management systems, full-text retrieval engines, file system storage,

and complex clusters of programs, all of which are integrated into Se¬

quent’s worldwide intranet and all of which are accessible through

consumers’ Web browsers. The external Internet, as well as core on¬

line transaction processing and decision support systems, are inte¬

grated into this infrastructure.

92 • If Only We Knew What We Know

COLLABORATIVE TOOLS AND GROUPWARE

More and more companies are turning to the Web-centric model as a

way of encouraging cross-enterprise collaboration, but sometimes a

more advanced tool such as groupware works better. That’s mostly the

case for more advanced and structured documents sharing.

Collaborative tools, such as groupware, have been around since

1989 and focus primarily on communication—individual, group, pro¬

ject, or companywide. Groupware supports work groups and people

working together. Lotus Notes is the most prevalent example. Knowl¬

edge managers use collaborative tools for a variety of reasons, ranging

from e-mail to the sharing of factual knowledge.

Most companies with Lotus Notes combine it with an intranet to

allow both structured and free-flow sharing.

For example, at Buckman, LeamingSpace from Lotus has been

added to the KM application suite to facilitate the company’s distance

learning project. LeamingSpace is a distributive learning tool with in¬

terconnected networks of Notes databases and a structured training

model. Its purpose is to deliver asynchronous, just-in-time (yet stan¬

dardized) education and training to associates throughout the world.

National Semiconductor uses Windows 95 as a base infrastructure

with Lotus Notes as an e-mail and knowledge-sharing tool. Lotus

Notes is installed on more than 5,000 desktops. The field sales group

has been using Lotus Notes for the past few years to share customer

data. The mainframe e-mail system is being transferred to Notes,

which will provide an opportunity for a wide variety of people to use

National Semiconductor’s knowledge-sharing technology.

Texas Instruments placed its core Best Practices KnowledgeBase on

Notes in March of 1995. The original facilitators, the Best Practice

Sharing Team, and the Quality Leadership Team loaded the original

supply of best practices; to date, the database contains more than 500

practices.

The World Bank relies on Lotus Notes to help its traveling staff

members to stay in touch when on missions. Each notebook computer

will contain a Lotus Notes-based knowledge management navigation

system. This way, staff members can stay up to date no matter where

and how long their mission lasts.

Using Information Technology to Support Knowledge Transfer • 93

BOB BUCKMAN’S IDEA-SHARING SYSTEM

Bob Buckman envisioned the characteristics of an ideal electronic net¬

work, and developed the following design features for it, presaging by ten

years the characteristics of effective “knowledge management” systems.

1. Keep the number of transfer steps in the transmission between in¬

dividuals to one to alleviate the potential distortion of knowledge.

2. Allow all employees access to the system.

3. Allow all employees to make contributions.

4. Make the system available from any location, twenty-four hours a

day.

5. Have a user-friendly system able to search on every word.

6. Allow the users to contribute in their native language and where

deemed appropriate provide the required translation.

7. Provide a system that updates automatically as information is pro¬

vided.

NOTES VS. THE INTRANET

Notes and the intranet are both potent sharing tools; however, they

will not do your work for you. Don’t be lured into thinking that mere

installation of an intranet or a Lotus Notes implementation will make

everyone want to share.

The success of the KM application of your choice is dependent on the

human infrastructure or the soft side of knowledge management for its

success. The soft side is often the most difficult piece. (See Chapter 9.)

The good news however is that whereas, in the past, companies

needed to make a choice to go with Notes or an intranet architecture,

that’s no longer necessary. Although the debate still rages about which

“architecture” is the most knowledge-inclined, most users end up com¬

bining them and rely on Domino to allow access back and forth.

Some companies, like Sequent, swear by the intranet as the quintes¬

sential KM application. Its very “spirit,” they argue, is about sharing, and

no one “owns” it, hence your migration path is not tied to a particular

vendor’s plans. True enough. On the other hand, Notes provides an

environment for more sophisticated collaboration and structure ex¬

changes. The trick is choosing the best technology for the specific KM

94 • If Only We Knew What We Know

INTRANETS MEAN ROI

In 1997, The Meta Group Inc., a market research and consultancy in

Stamford, Connecticut, analyzed the return on investment (ROI) of in¬

tranet applications. According to Meta, 80 -percent of companies sur¬

veyed generated a positive ROI, with an average annualized return of 38

percent.

Meta surveyed 55 companies, ranging in size from $2 million to $35 bil¬

lion, and employing intranet technologies across a broad range of applica¬

tions, including collaboration, customer service, inventory management,

and database access. By and large, applications that were interactive and

affected the firms’ value chain directly yielded much higher ROIs. Meta

also found that the culture of the company and its IT department have a

significant impact on the success or failure of intranet technology.

effort. Neither one is necessarily better than the other. But in any event,

Notes and intranets should be used with discretion.

FROM BACKBONE TO BODY—TOOLS AND APPLICATIONS

If the greatest database in the company is housed in the individual

minds of the associates of the organization, then that is where the

power of the organization resides. These individual knowledge bases

are continually changing and adapting to the real world in front of

them. We have to connect these individual knowledge bases

together so that they can do whatever they do best in the shortest

possible time.

—Bob Buckman, president, chairman, and CEO, Bulab Holdings, Inc.

If Notes and/or the intranet are the technological backbone of the

transfer process, then this backbone holds a complex body of KM-

specific tools and applications that enable particular aspects of sharing.

Two broad classes of IT applications are being used to support trans¬

fer, and each contains a number of tools specific to the application:

1. Knowledge Transfer and Exchange

• Structured document repositories (a.k.a. databases)

• Discussion databases

Using Information Technology to Support Knowledge Transfer • 95

• Pointers to expertise

• Document exchange and video infrastructure

2. Data Analysis and Performance Support

• Performance support systems

• Problem resolution systems

• Data-to-knowledge conversion systems

—Data mining

—Decision support

—Real-time intelligent data analysis

We will focus on the first group here, primarily because it reflects the

market’s focus and practice. In our 1997 study “Using Information

Technology to Support Knowledge Management” (APQC, 1997), 89

percent of participants indicated that their goal was to capture and

transfer knowledge and best practices, not analyze data.

Furthermore, data analysis and performance support applications

involve distinctly different and highly specialized technologies and de¬

sign considerations. Such tools are applied to raw data and/or informa¬

tion, not knowledge, so they fall outside our best-practice focus.

Still, we believe there may be a vast potential for combining the two

areas into a holistic knowledge and practice sharing system. To that

end, we encourage readers to find out more about data analysis and de¬

cision support, and provide a few thoughts/examples (see page 100,

KM’s Uncharted Territory).

TRANSFERRING VIA STRUCTURED DOCUMENT REPOSITORIES

Structured document repositories typically contain databases that

have structured content consisting of regular, alphanumeric data capa¬

ble of being stored in conventional relational databases. A good exam¬

ple is best practices databases, which are usually repositories of short

descriptions of best practices and/or pointers to contacts within the or¬

ganization who have knowledge about these practices.

Databases that are organized around customers—which may include

information about customer inquiries, needs, and interactions—and

databases containing competitor intelligence may also be stored in these

repositories. These databases typically include sales presentations, re¬

ports, engagement information, competitor analysis, and external feeds.

96 • If Only We Knew What We Know

The following three tips may be helpful when implementing struc¬

tured document repositories:

1. It is important to supply “magnet content” that causes people to

use the database in the context of their work.

2. The best systems automatically capture content rather than re¬

quiring people to take an extra step to reenter it.

3. Every system needs a disciplined process for creation, evaluation,

categorization, maintenance, and renewal—that is, people have

to be assigned these responsibilities.

At the World Bank, the Africa Region Live Database (LDB) pro¬

vides data and analytical tools to country and sector economists, who

are then responsible for maintaining the quality and timeliness of data.

The analytical tools are closely adapted to the work needs of the econ¬

omists. They are offered as an inducement to economists to do their

work within the LDB and to perform the data maintenance.

One of the ways the Sequent Corporate Electronic Library is supplied

with valuable information is through the hypermail approach. This

simple technology retains e-mail exchanges and organizes them by

“threads” or subject so that others can search/find them in the future.

Instead of re-inputting solutions as new documents, the learnings are

automatically captured and catalogued.

TRANSFERRING VIA DISCUSSION DATABASES

These are discussion groups of project or work teams (whether e-mail

listsserves, news groups, or Lotus Notes—based discussion groups).

They may also support communities of practice, project work teams, and

salespeople or sales and marketing teams.

Helpful implementation tips include:

1. Make sure you have shared norms.

2. Provide some payback for participation (tangible or intangible).

3. Ensure active facilitation by a moderator.

4. Offer a “one stop” solution for e-mail and discussion.

5. Nourish the on-line community through other means as well

(e.g., face-to-face meetings).

Using Information Technology to Support Knowledge Transfer • 97

6. Remember that content has to be good, if these are to attract vis¬

itors.

7. Use logical, intuitive structure for views.

8. Use library scientists who understand the technology, and tech¬

nologists who understand library science.

9. Provide some editorial and publishing support to increase value

and accuracy.

10. Create automatic systems for “cleaning out the closet” and archiv¬

ing past discussions.

Sequent has two kinds of mechanisms for discussion: NetNews and

group distribution lists. These discussion groups may be created by any

employee, archived at the request of an employee, and either moder¬

ated or not, as the community using it chooses. Employees can add

themselves to, and remove themselves from, discussions at will when

the discussion group is not moderated, and by permission of the mod¬

erator when moderated.

Texas Instruments implemented a discussion database through Lotus

Notes, enabling employees to communicate business gaps and solu¬

tions. Externally identified best practices and benchmarking informa¬

tion also were incorporated into the Best Practices KnowledgeBase.

AA OnLine is Arthur Andersens firmwide conferencing forum

designed to help its professionals stay informed about industry and

service-line issues and communicate with other firm members. AA

OnLine is an interactive conferencing tool that provides users with

the means to talk electronically to a large group of people and the abil¬

ity to share resources and expertise. The database is organized around

“groups” or knowledge networks, which are defined as communities of

practice with a need to share valuable materials or ideas. AA OnLine

is currently deployed throughout the firm and hosts announcements

and discussions; it holds resources for about eighty-plus AA knowledge

networks. There are currently 20,000 users.

TRANSFERRING VIA POINTERS TO EXPERTISE

Pointer systems are applications such as “yellow pages” of internal ex¬

perts, project managers, or interested parties. They may also include

98 • If Only We Knew What We Know

human resource (HR) listings and profiles of employees. While these

are technologically simple, and extremely effective in enabling the

transfer of high-value knowledge, they are not as simple to implement.

When building a pointer system, use the following guidelines:

1. Do not rely on users to maintain their own HR information.

2. Information needs to interface or be fed by the corporate HR in¬

formation system.

3. Maps of experts and communities of practice are at times out of

date but are better than nothing.

4- Formal discussion groups can be catalogued.

5. Experts must be motivated to supply their profiles.

6. “Expert” may be a highly political concept.

7. Don’t create a pointer system that’s basically an electronic ver¬

sion of a phone book. Add value by interconnecting the yellow

pages with other components of the KM application suite, so that

users can use “hotlinks” to further refine their needs.

8. Make sure there’s a simple and standard format to follow for new

submissions and revisions.

Teltech Resource Network Corporation, a company based in Min¬

neapolis, has made the corporate yellow pages its core business. Tel-

tech specializes in putting together networks of technical experts and

helps its clients search for the right expert. In fact, Teltech has been so

successful it has taken the lessons it learned from building its own busi¬

ness and is now helping others design and create knowledge and trans¬

fer systems with an emphasis on the human connection. “The most

immediate payback [for KM effort] is through creating the navigation

capabilities to put people in touch with people,” says Andy Michuda,

Teltech’s president and CEO. Indeed, he says, he’s “seeing more and

more companies initially focused on computers and technology now

starting to migrate to people-based capabilities.” What these compa¬

nies realize is that it’s not all about systems. “From a process stand¬

point, the key is to find the pockets of intellectual capital and bring

them together in a timely manner.”

At Sequent, the corporate yellow pages are provided by integrating

SCEL with PeopleSoft, Sequent’s on-line HR system, which allows

dynamic access to personnel profiles maintained by the HR systems.

Using Information Technology to Support Knowledge Transfer • 99

Subject matter expertise is advertised through “professional profiles”

maintained by employees. Sequent Computer Systems planned to im¬

plement a corporate skills repository in the HR organization in 1998.

That repository would be integrated into, and accessible from, SCEL.

TRANSFERRING VIA DOCUMENT EXCHANGE AND VIDEO

E-mail is the most pervasive and effective means of electronic commu¬

nication and collaboration. It is real-time, fast, easy, and user-driven.

Video conferencing is growing, but not ubiquitous from the desktop,

like e-mail. Though e-mail is a rich source of information, there is too

much of it and it is unfiltered—thus, gems are frequently lost.

At Sequent, e-mail and news groups are provided using the same un¬

derlying technology suite (TCP/IP and Internet protocols) as SCEL.

News groups are accessible through the browser’s built-in news-reading

capability. E-mail discussions on shared aliases are treated as docu¬

ments by SCEL and archived for future reference by interested con¬

sumers.

At CIGNA Property & Casualty, when an underwriter gains insight

on a topic, he or she only needs to push a “shared observation” button

to immediately transfer the comments as an e-mail, which allows the

underwriter to communicate with other underwriters in the network.

THE ROLE OF THE IT ORGANIZATION

It’s time to stop and take a breather.

We’ve run through pages of examples, lists of tools and applications,

helpful tips and implementation guides. Certainly lots of trees. But do

you see the forest?

In addressing the technological aspect of knowledge management,

we address more than the technology platforms and applications. We

also have to address strategy and underlying paradigms that drive the

design, as well as the roles of people throughout the system. We’ve talked

about the backbone/platform, we talked about the parts of the “body.”

Now, we talk about the neural networks that make it all work in sync.

One of the most important characteristics of a knowledge man¬

agement architecture is that structures and roles are never static. In a

100 . If Only We Knew What We Know

knowledge-intensive economy and organization, it is not surprising

that managing knowledge requires that people be specifically assigned

to every stage in the process—collecting, organizing, adding value, dis¬

seminating, and supporting knowledge-in-use. Four-fifths of produc¬

tive value added by technical change will continue to result not from

KM’S UNCHARTED TERRITORY

We may know more about using technology to enable transfer than we

know about data analysis and performance support. But in our travels

throughout the KM landscape, we happened to find several examples of

such applications that certainly fall within the boundaries of knowledge

management.

Data analysis and performance support applications are usually found

within the business units supporting the actual work processes. These

applications include the following:

• performance support systems

• problem resolution systems

• data-to-knowledge conversion systems, including data mining, deci¬

sion support, and real-time intelligent data analysis

Performance Support Systems. Used in real time to support the individ¬

ual’s job performance and learning, these types of systems usually:

• support a broader set of process and strategic objectives

• require significant training and retraining to use

• relate to specific processes or functions

Sales and call center support systems are a good example of this type

of application. They typically support real-time customer interaction and

rely heavily on effective training and orientation. They may use expert

systems to support training and just-in-time learning. Scripts for cus¬

tomer interaction (tacit knowledge made explicit) are also used.

At CIQNA Property & Casualty, for instance, management realized

that a limited number of their own experts had an abundance of tacit

knowledge that needed to be exploited and captured.

Identifying the high performers and knowledge gaps was the first step

toward latent knowledge extraction. The next was to create a decision-

support infrastructure that allowed knowledge to be provided to the rest

of the organization. This knowledge needed to be translated from “many

Using Information Technology to Support Knowledge Transfer • 101

the technology itself but from the new arrangements of organization,

management, and people who can make the best use of it.

We’ll outline these emerging roles and responsibilities in the next

chapter, on Infrastructure. But we can’t leave this chapter without dis¬

cussing the changing role of the IT function and how to leverage the

to many” by: collecting insight, analyzing case summaries, segmenting

cases by intent categories, identifying case patterns, and developing a case

library with examples of best practice by category.

During the second stage, CIGNA developed an Underwriting Deci¬

sion Summary model. The model allows employees to share their insights

and create information.

Problem Resolution Systems. Problem resolution systems apply “case-

based reasoning” technology to the resolution of customer and other

types of problems faced by an organization. This type of system uses

knowledge that is structured into “cases” and probable scenarios. The

benefits here are faster resolution, fewer expert personnel, and fewer

hand-offs for problem resolution.

Typical applications are the IT customer support systems offered by

Hewlett-Packard, Compaq, Inference, and other computer product com¬

panies. These systems provide real-time information and problem-

resolution support on products, pricing, and customer questions and

problems.

Data-to-Knowledge Conversion Systems. These systems rely on neural

networks or other statistically oriented intelligent algorithms. They are

usually combined with data visualization tools. However, it requires a

smart person to know where to look and to generate hypotheses. This

system is very popular with marketing groups. Data-to-knowledge con¬

version systems can include:

• Data mining—used in the analysis of large transaction-based data¬

bases in order to obtain customer purchasing behavior and demo¬

graphics, which in turn are used to target markets with the right

promotions and products.

• Decision support for knowledge work—used sometimes for under¬

writing and risk analysis.

• Real-time intelligent data analysis—systems that recalibrate waiting

times, purchasing patterns, etc.

102 • If Only We Knew What We Know

knowledge of technologists and KM practitioners within the organiza¬

tion.

Participants in our studies report that they are drawing heavily on

information technology departments to support their transfer effort;

even better, they tell us their information technology staff is readily re¬

sponding. In fact, by all accounts, KM practitioners are partnering

with IT colleagues to adapt the appropriate information technology

solutions to fit their evolving needs.

In 59 percent of the cases APQC has studied, corporate IT (or IS—

information systems) is actively partnering with the knowledge man¬

agement initiatives. And a full 35 percent of IT professionals have

already been trained in knowledge management and best practice

transfer concepts. “Only through symbiotic interaction between peo¬

ple and information technology can we truly gain competitive advan¬

tage in today’s global economy,” says Bipin Junnarkar of Monsanto.

This reality has not escaped the notice of KM pioneers:

At Sequent, KM and information technology organizations are func¬

tionally separate but work closely together. The CKO and CIO are peer

positions in the organization. The information technology organization

provides technology and human resources to support knowledge man¬

agement programs and maintains the knowledge management infor¬

mation technology infrastructure. Whereas the CKO formulates KM

strategy, the CIO and the information systems (IS) organization are re¬

sponsible for defining information technology standards for the organi¬

zation. The CKO and CIO jointly chair a corporate-wide steering

committee to ensure that the common infrastructure is used, and used

properly, by SCEL and other projects.

At CIGNA, technology is a critical component in learning and

knowledge building. In developing an information technology structure,

the IS team worked closely with the reengineering team to determine

how information technology can best support the evolving needs.

At the World Bank, the Information and Technology Services De¬

partment is charged with supporting the knowledge management sys¬

tem with the appropriate technology, ensuring that the knowledge

management system is fully integrated with other information systems.

At National Semiconductor, knowledge management initiatives are

supported by corporate IS and engineering IT. There are fifteen to

Using Information Technology to Support Knowledge Transfer • 103

twenty-five individuals in engineering dedicated to defining standards,

architecture, and tools for the engineering community. Corporate IS

sustains a knowledge management/information access group of about

ten individuals responsible for maintaining the data warehouse for cor¬

porate, sales/marketing, and manufacturing. Desktop integration is an

additional responsibility of the corporate IS group.

Buckman reorganized to optimize knowledge sharing by creating

a Knowledge Transfer Department (KTD) in March 1992. The KTD

consolidated the IS and telecommunications departments and merged

with the technical information center. Its mission is to plan, organize,

and manage information system applications, infrastructure, and

associated resources for rapidly disseminating collective industry, tech¬

nical, and market knowledge that Buckman needs to gain a global

competitive advantage.

CONCLUSION AND DESIGN LESSONS

Technology can be a powerful enabler of transfer effort. It can link

people across time and space. It can spread knowledge and practices

rapidly, efficiently, and cheaply throughout your organization. It can

help document and capture formerly tacit knowledge. There is no

doubt that behind the current popularity of KM is the potent combi¬

nation of newly pervasive technologies and companies’ realization

that knowledge assets must be managed.

But while new technologies are making the transfer of practices and

knowledge more affordable then ever before, companies that think

that simply by developing an intranet they will make sharing happen

are dangerously wrong.

There are both IT challenges (which platform, application, and

tools to use and how to use them) and more important, supportive in¬

frastructure issues (if you build it, will they come?) that companies

must keep in mind when designing their KM/IT architecture.

Here, in a nutshell (it was a big nut, we know . . . ) are the lessons

gleaned from our experience and that of participants in our KMIT

study that can help you take the first step, the right way.

Establish standards; they are the key to sustainability. Beware of the IT

hodgepodge. If you are using technology to support your transfer effort,

104 • If Only We Knew What We Know

choose a standardized platform at the start. Be it an intranet or Lotus

Notes setup, or perhaps a combination thereof, by designing an ex¬

pandable system from day one, you ensure the future of the transfer ef¬

fort. The benefits are twofold. First, investment in IT will not become

obsolete. Second, by standardizing the core, firms can allow diverse

parts of the organization to participate actively in developing new ap¬

plications while ensuring that every new venture is compatible with its

predecessors. Such careful attention prevents the formation of archi¬

pelagoes of knowledge, islands of expertise no one can reach without a

boat.

Match the KM system with the KM objectives. There is no single right

way to design a technology architecture to support knowledge transfer.

By and large, the more tacit and high-value knowledge transfers often

involve pointer systems or help desks manned by experts. A loose in¬

terchange and brainstorming session in the conference room with

cookies may be more productive than using groupware. Hypermail can

help you track and “remember” important spontaneous discussions.

Simply throwing the latest and greatest at your employees will not

make the share. So figure out what you want to share, how often and

for what reason, and then build an IT solution that makes it happen.

Create a structure for classifying knowledge. Many companies reported

that they could not organize their knowledge nor provide sufficient,

IT-enabled access to it without an underlying structure or taxonomy.

The structure specifies the categories and terms for the knowledge in

which the firm is interested. It lets knowledge contributors classify

their contributions, and later allows those searching for knowledge to

find it easily. Other terms used to describe such a structure include

“metadata” (data about data), a “content classification scheme,” and

“knowledge models.” Some firms included the classification of knowl¬

edge contributions as a part of their design guidelines for knowledge

contributions. Several firms also found it helpful to create a thesaurus

for knowledge relevant to the organization. A thesaurus allows searchers

to find the knowledge they seek even when it is classified under terms

with which they are not familiar.

Using Information Technology to Support Knowledge Transfer • 105

Heavily market your transfer applications and ensure they meet users’

needs. “If we build it, they will come” works only with substantial at-

tention to marketing. Several firms also stated that knowledge system

designers must understand the daily behaviors of their customers; if

there isn’t an easy answer to the question “How will this improve my

day?” the system will not be used. Some companies pointed out that

knowledge managers should constantly publicize examples of success¬

ful use—another form of marketing. One noted that it was important

in marketing the knowledge capability to use the language of the

user—local terminology such as “best practices,” “reuse,” “cycle-time

reduction,” or “access to expertise,” rather than the jargon of the

knowledge management movement. “If they understand it, they will

come” is probably a very true statement.

Remain flexible; nothing is etched in stone. Any particular knowledge

management system or approach must be treated as provisional and

temporary. Both knowledge and the technology for managing it

change rapidly. Most firms started their knowledge management initia¬

tives with pilot programs to learn how best to approach the problem.

But others argued that the pilot stage is permanent: “All programs are

provisional, all technologies subject to abandonment.” One company

felt that current IT tools are particularly immature and thus require

“careful planning and trade-offs to deliver functional, well-perform¬

ing solutions.” Several firms emphasized flexibility in building and

maintaining technology platforms. The key is to understand how cus¬

tomers are using the technology and respond rapidly with appropriate

changes.

Maintain a pragmatic vs. perfectionist approach. One firm’s byword was

literally “Focus on what works, not perfection.” Another company

maintained that the right design for a knowledge management system

will only be determined over time through customer behavior. Practi¬

cality was emphasized for knowledge structures, desired behaviors, and

technology. One firm recommended that knowledge systems should be

as easy to use as possible because a switch to a knowledge orientation is

difficult enough without complex technologies.

106 • If Only We Knew What We Know

Keep people first; KM is inherently people-based. Or, as one conceptually

minded firm put it, KM is “fundamentally sociocultural. Because

knowledge can only be surrendered voluntarily, many firms empha¬

sized the importance of viewing knowledge management in human

terms. Get as many people participating as possible. View every em¬

ployee as both a content provider and a consumer. Use people, not just

technology, in support roles for knowledge management. Some sug¬

gested emphasizing natural work groups and existing “communities of

practice” rather than trying to create new ones. The strongest advo¬

cates of this lesson argued that best practice came from “focusing more

on people and less on IT.”

Measure the Impact of Knowledge Management. Because someone will

ultimately question whether benefit is being derived from spending on

knowledge management, many KMIT participants suggested that

value and impact should be monitored from the beginning. One firm

argued for close ties to the firm’s measurement systems. Others advo¬

cated more anecdotal means of value measurement: “Collect anecdo¬

tal examples and testimonials of the contribution of knowledge

management to the business result.” Measuring and managing impact

is difficult, and not all firms in the study do it. But many of those that

do not measure and manage the impacts have nagging doubts that

someday the proof of value will be required.

Chapter 11

CREATING THE KNOWLEDGE INFRASTRUCTURE

Even if you announce full senior management support for a best

practices and knowledge transfer initiative; even if you put in the

most sophisticated “anywhere-anytime-anybody” technology for shar¬

ing; and even if you provide incentives for sharing (recognition, pro¬

motion, money), you may get lousy results.

Why?

Because people need help in understanding and transferring best

practices.

Almost every successful organization we have worked with realizes

the need and importance of an explicit and institutionalized organiza¬

tional infrastructure to assist the transfer of knowledge and best prac¬

tices. (How they do it varies. Who does it varies. What they do varies.

But not whether they do it.) These early practitioners know that trans¬

fer would never happen without a process, and an infrastructure of

people dedicated to facilitating the process.

WHAT’S A KNOWLEDGE INFRASTRUCTURE ANYWAY?

Infrastructure includes the transfer-specific mechanisms put in place to

ensure best practices flow throughout the enterprise. These include

technology, work processes, and networks of people. Infrastructure also

includes the organizational structure surrounding the processes: the es¬

sential line and staff roles that must be played to support the new ini¬

tiative of knowledge transfer. That organizational structure has at least

two layers.

Layer 1 is the micro structure. In the trenches are the “unsung heroes”

107

108 • If Only We Knew What We Know

of transfer; the people in nooks and crannies of the organization. We

call them knowledge brokers, facilitators, librarians, and a hundred

other new and old names. But whatever their title, we mean the people

who are organized into roles, systems, and structures to make transfer

happen.

Layer 2, the super structure, is the link to the formal organization

structure. At some companies, the KM staff reports to a senior level of

KM executives—be they CKOs, CIO, or director of quality knowledge

networks (more on that later in this chapter).

Finally, infrastructure includes the cross-functional management

processes that incorporate KM into the fabric of the organization: bud¬

geting (who’s gonna pay for this stuff?) and implementation (who

“owns” it and is accountable for results?).

We find the most reliable test of the seriousness and effectiveness of

a company’s best practice transfer is whether or not it has put in place

an explicit, supportive infrastructure that makes knowledge sharing a

primary focus of someone. If no one is “charged” with watching out for

transfer, it will fizzle and fade.

DON’T LEAVE CHANGE TO CHANCE

Like a lot of other things in life, transfer of knowledge and best prac¬

tices doesn’t just happen because it makes good sense. Or because

management says it ought to. Building the right infrastructure to facil¬

itate change is critical to ensuring the success of your transfer projects.

Change isn’t automatic, or easy, or quick. We’ve mentioned some of

the barriers to change in earlier sections, but as a reminder, there are six

barriers that specifically hinder the transfer of know-how and practices,

and which reinforce the need for an explicit knowledge infrastructure:

1. Hidden knowledge. People don’t know what it is that they know,

or that someone else needs it. They’re sitting on a potential gold

mine but don’t realize it. They can’t transfer something they

don’t know they have.

2. Blindness. Knowledge and best practices may exist (even down

the hall), but the potential recipients don’t know about it. They

may suspect it exists but don’t know how to find it. They don’t

search or don’t know how to search. In effect, they are blind.

Creating the Knowledge Infrastructure • 109

3. Locked'Up tacit knowledge. Tacit knowledge—the know-how and

judgment that come from experience, intuition, tricks, rules of

thumb—is often 80 percent of the valuable knowledge in a

process. But because such tacit knowledge is typically very hard

to express and difficult to codify, the really valuable stuff remains

“between the ears and behind the eyes” of the source. The recip¬

ient gets only the explicit (20 percent) portion of the valuable

knowledge, and then wonders why the transplant failed.

4. “We’re different” blinders. Potentially valuable knowledge and

practices often don’t transfer across functions, companies, indus¬

tries—even entire sectors like education—because potential re¬

cipients focus on differences, not similarities of processes. They

reject valuable knowledge due to parochialism, tunnel vision,

ego, ignorance, protectionism, arrogance.

5. “Sorry—I’m too busy.” Even if the transfer would save them time,

they don’t have time to save time. The transfer never gets a hear¬

ing.

6. Implementation is hard. Even if all of the previous barriers are

overcome, the transfer of knowledge occurs, but action doesn’t

follow. It remains under construction or simply dies, for a variety

of reasons—no money, fear of change, lack of leadership, no buy-

in, turnover, no support, lack of training, change of manage¬

ment . . . you name it.

Unsuccessful organizations ignore these barriers and keep relying on

the false assumption that just because best practices exist, Economics

101 predicts they will automatically transfer. They don’t. Or if they do,

it is only after a long lag and probably a slip in competitive rank.

How long?

Research by professors at Vanderbilt University showed that inno¬

vations in education have taken twenty-six years on average to move

from the first 10 percent of early adopters to widespread acceptance.

That’s long. Unfortunate in education, but disastrous in industries

where the rate of change is measured in nanoseconds.

THE BRAVE NEW WORLD OF CHANGE AGENTS

Successful organizations don’t leave change to chance. They assign spe¬

cific roles and responsibilities, and in some cases invent a nomencla-

110 . If Only We Knew What We Know

ture to describe these new roles. Some “titles” are borrowed from useful

metaphors, such as the publishing industry or sports. Some organiza¬

tions, such as Chevron and Hewlett-Packard, designate “process mas¬

ters” or “internal consultants” who have specific process expertise in

addition to facilitation expertise. Some companies centralize the

transfer-related work under one head—the chief knowledge officer, or

director of knowledge networks. At other organizations, KM “middle

managers” report to business unit heads, the CIO, or the CEO.

Regardless of the reporting structure, these individuals, operating at

the front lines of the best practice transfer movement, are charged with

providing daily assistance, encouragement, and involvement to employ¬

ees throughout the firm. Their job is to facilitate the flow of know-how

and best practice, add value, edit, search, filter, and sort out.

THREE APPROACHES TO INFRASTRUCTURE

As we roam the world of transfer practitioners, we’ve run into all man¬

ner of enabling infrastructures. They range from low to high tech, from

a few people to several hundreds. Regardless of the model’s size and

complexity, these infrastructures share some basic processes and fall

into three general design approaches:

1. Self-directed

2. Knowledge services and networks

3. Facilitated transfer

The three are not mutually exclusive. In fact, they coexist in many

organizations. The more advanced a company’s KM practice, the greater

EMERGING KM TITLES

Knowledge management champion(s) Process masters

Knowledge management managers Progress directors

Publishers and publishing coordinators Progress consultants

Librarians and information specialists Internal consultants

Network or team leaders, and section managers Coordinators Help desk (content and process) Knowledge brokers Facilitators Cross-pollinators Shepherds Lightning rods Coaches

Creating the Knowledge Infrastructure • 111

FIGURE 11.1

Three Approaches to Infrastructure

the likelihood that its transfer infrastructure will be rich in texture and

multidimensional in design.

These three approaches can be placed on a continuum of ever-in¬

creasing likelihood of results and higher resources/commitment (see

Figure 11.1). In Table 11.1, we’ve put some descriptive words that ex¬

plain the characteristics and methods or each approach.

TABLE 11.1

How the Three Approaches Address the Barriers to Transfer

Six Barriers Self-Directed

APPROACHES

Knowledge Services

& Networks Facilitated Transfer

1. Don’t know <8 <8

2. Blindness <8 <8 <8

3. Tacit knowledge <8

4- We’re different <8 <8

5. Too busy <8

6. Implementation <8

112 • If Only We Knew What We Know

APPROACH#!: SELF-DIRECTED

Descriptive words: storage, codification, repository, database, re¬

trieval, navigation, pointers, yellow pages, dissemination, m

tranets, Internet.

The self'directed approach to designing a knowledge infrastructure es¬

sentially says, “Here it is, now go use the technology to help you find

what you want.”

The self-directed approach is often augmented with “pointer sys¬

tems” which can be anything from knowledge maps to corporate yel¬

low pages. These systems don’t contain the information or description

of the best practice; rather they direct the user to someone who might

know more about a particular topic.

The database is the key component. Companies that rely on this ap¬

proach employ databases, repositories, autonomous agents, and search

methodologies to allow employees easy access to the knowledge they

require. Some electronic repository or knowledge base is the building

block for all three design approaches, but in this case it s the corner-

stone.

The knowledge base’s main function is to capture data and informa¬

tion. It collects information from newspapers, journals, analysts re¬

ports, other databases, field reports, the Internet, presentations, and

the like, and stores them either as full text or abstracts. Think of these

databases as electronic libraries and card catalogs. You can go there to

look things up.

Texas Instruments calls this component of its KnowledgeBase the

“Document Database.” Sequent Computer employs two librarians to

constantly record and access knowledge. Some companies even capture

best practices tales in these knowledge bases. Both TI and Chevron have

a best practices database where description of practices and results are

electronically stored for later browsing.

Whatever their “name,” these databases alone typically do little to

transform the information into knowledge beyond some low-level purg¬

ing and pruning; there is often no attempt to integrate the information

into “valued oriented” packages tailored for use by the organization, or,

Creating the Knowledge Infrastructure • 113

as some firms call it, “making sense” out of information. That’s up to

each and every user to figure out.

Essentially, these databases passively disseminate; users “pull down” in¬

formation they need, when they need it. Some databases provide value-

added or somewhat-intelligent software (e.g., grapeVINE, Hoovers, or

Excalibur) that serve as “autonomous agents.” Agents kick out infor¬

mation to the user in accordance with a preset “interest profile” pro¬

vided by the user.

Locator systems are the second main feature of this design approach.

These are often called pointers or maps and tell users where to find par¬

ticular information. Paraphrasing Samuel Johnson, the writer and

philosopher, “The next best thing to knowing something is knowing

where to find it.” These maps range from simple directories supplying

names, phone numbers, faxes, and e-mail addresses to search engines

with taxonomic or full text research.

Most maps are electronic, but some companies, like Chevron, actu¬

ally started with a paper version that folds just like a highway map.

Chevron’s map includes names, location, phone numbers, and other

pertinent information for people in Chevron worldwide who are work¬

ing in specified knowledge arenas. Hughes Space calls their locator the

“Knowledge Highway.” AT&T relies on an on-line directory of exper¬

tise. Coopers & Lybrand employees can call 1-800-KNOW-HOW to

access a help desk with a human researcher who uses a database to lo¬

cate expertise.

The self-directed approach is certainly useful. It helps users find in¬

formation directly or to find a person who has the information they re¬

quire. Thus, it partially addresses the “search” problem listed as one of

the six items that frustrate the flow and implementation of knowledge.

However, if falls short on addressing some of the other “barriers.”

• A database really can’t represent the treasure chest of “tacit knowl¬

edge” which may hold the greatest potential for improvement

• Even human pointer systems (a la Teltech—see below) are limited

in that they rely on the users to initiate the quest and incorporate

the findings

• It does little to overcome the internal cultural barriers of “we’re dif¬

ferent” and “we’re too busy”

• It does nothing about moving to action

114 • If Only We Knew What We Know

Ultimately, self-directed is a useful, but not sufficiently comprehen-

sive, approach.

TELTECH’S LOCATOR SYSTEM

In 1984, Joseph Shuster leveraged his knowledge of how technical pro-

fessionals work into a one-stop knowledge shop: The Teltech Re¬

sources Network Corp. The Minneapolis-based firm is a sort of a KM

hybrid which allows its clients to access a plethora of knowledge

sources,” from human experts (over 3,000 of them) to article clip¬

pings. Teltech “gatekeepers” help knowledge seekers find experts with

knowledge on their topic of interest. This blend of human expertise

(and “human search engines”) and technological savvy allows Teltech

clients to dramatically reduce the time it takes them to locate relevant

experts.

The network has been so successful that Teltech has managed to

embed its growing base of knowledge about locator systems into a new

service; it now helps others set up their own locator systems. Says CEO

Andy Michuda: “A couple of years ago, all the leading examples of KM

revolved around Notes and intranet application; more recently, people

have began to recognize that the people/expert component is the key.”

Michuda recalls a recent client—a semiconductor maker—that has

been laboring for years to “convert” all of its existing tacit knowledge

into explicit form. The result: a very long process and a high level of

frustration, not to mention a reservoir of mostly outdated information.

Teltech helped the semiconductor company design a “low-tech” ex¬

pert locator system instead. “The system offers navigational capabili¬

ties to access current, live, relevant knowledge,” says Michuda, “in less

time, with less money, and with a higher payback.”

APPROACH #2: KNOWLEDGE SERVICES AND NETWORKS

Descriptive words: information services, help desk, networks, dis¬

cussion databases, communities of practice, knowledge managers,

knowledge integrators, knowledge packagers, want ads.

Creating the Knowledge Infrastructure • 115

The second approach to designing knowledge infrastructures, knowl¬

edge services and networks goes further. In addition to providing self-di-

rected components, this approach also provides a variety of knowledge

management services and organized networks to assist in the transfer

process. Knowledge managers and knowledge integrators add value by

scanning the flow of information, and organizing or “packaging”

knowledge into a more digestible and applicable format.

This approach also involves extensive networks of people who

come together to share and learn from one another face-to-face and

electronically (sometimes known as communities of practice—at one

consulting firm, a network was four hundred members strong).

Given that knowledge is the principal product of consulting firms, it

is not surprising firms such as Arthur Andersen, Ernst & Young, Price

Waterhouse, Coopers & Lybrand, and McKinsey have major invest¬

ments in infrastructure and employ this approach. But services and

networks are also prevalent among many nonconsulting outfits as well,

including Amoco, Buckman Labs, Chevron, the National Security

Agency, and Texas Instruments.

This second approach not only involves sophisticated databases,

but establishes the services and networks to leverage the collective ex¬

periences, skills, and intelligence of the organization.

ERNST & YOUNG

Like its Big Four consulting firm brethren (we offered examples from

Arthur Andersen earlier in this book), Ernst & Young has made a sig¬

nificant investment in this KM infrastructure.

The firm’s internal research found that up to 80 percent of its resi¬

dent knowledge is not being applied to business processes in a system¬

atic manner. Armed with the valuable knowledge of just how much it

did not know, E&Y quickly adopted an extensive “Global Knowledge

Sharing” approach; the infrastructure includes:

• Global Knowledge Steering Committee—with chief knowledge of¬

ficers (CKOs) from key countries

• Global Knowledge Council—addresses specific functional strategy

issues

116 • If Only We Knew What We Know

• Five Centers for Business Knowledge which house the support staff

for their initiatives

. One hundred Knowledge Networks—practice professionals respon-

sible for collecting, storing, updating, and advancing the knowledge

of the firm

• PowerPacks—a database for each E&Y Knowledge Network to store

its reusable information

. EY/KnowledgeWeb—collective catalogued information available

to all—200,000 accesses per month

• Knowledge Services Group, with three processes

—Quick Response: fifteen information professionals who respond

to inquiries, with a goal of responding within two hours. About

50 percent of the requests are for information for proposals; 25

percent want info for client service, and 25 percent relate to firm

strategic planning issues

—Research Group: handles in-depth inquiries where time and re¬

sources are required

—Business Analysis Group: analyses of competitors

E&Y has many other specialized knowledge services, databases, and

networks, all with the emphasis not on technology but on building a

sharing infrastructure that everyone in the organization understands,

and which “includes process owners, executive steering committees,

and knowledge creators and integrators,” according to John Peetz,

chief knowledge officer.

BUCKMAN LABORATORIES.

It is hard to understand the depth of Bob Buckman’s commitment to

knowledge management and transfer until you’ve seen the laptops at

each desk and workstation at his offices, listened to his commitment

and conviction in one of his many presentations, seen his electronic

network in action, and watched him participating in public meetings

and pecking away on his own laptop.

A clear indication of Buckman’s belief in the importance of transfer

was the 1992 creation of the Knowledge Transfer Department, which

he charged with:

Creating the Knowledge Infrastructure • 117

• Accelerating the accumulating and dissemination of knowledge

within the company

• Providing easy and rapid access to the company’s global knowledge

bases

• Teaching how to share best practices to all Buckman affiliates

The dedicated people who fill these roles complement an extensive

technological “pull” system called K’Netix (see case study in Chapter

14, page 144) as well as transfer-targeted recognition and rewards. The

“full service” approach has been paying off handsomely for Buckman.

Indeed, the company credits much of their 250 percent growth in sales

in the past decade to its system. Costs are down. Speed of response to

customers is hours, not weeks or days, and quality of response has risen

all over the world.

SEQUENT COMPUTER SYSTEMS

“We absolutely live or die by the knowledge of our people,” says John

McAdam, president and CEO of Sequent. His belief runs through the

company’s knowledge systems and underlies the design of Sequent’s

knowledge infrastructure.

For example, Sequent calls employees who produce knowledge

“publishers,” and employees who use it “consumers.” To know what

consumers want and need, the designers of the system brought to¬

gether groups of “influencers” from across the company to find out

what kinds of information was required and in what shape and form.

They found out that Sequent people wanted things they couldn’t

get elsewhere, such as outstanding presentations, scripts for sales calls,

and design documents. They also found out that some people wanted

information organized around the value chain, while others wanted a

functional view. Rather than forcing them to choose one or the other,

they do both; users of the Sequent Corporate Electronic Library can

click on a “value chain” view button or an “organization” view button.

To add value to knowledge transfer, change agents “listen in” on e-mail

conversations and pull out the “threads” from the conversations and

post them for later referral.

118 • If Only We Knew What We Know

The knowledge services and networks approach addresses many of

the transfer problems:

• It helps the organization “know more about what it knows.

• It facilitates “seeing” with databases and sophisticated search en-

gines.

• It taps into tacit knowledge by stressing personal relationships.

• It helps people overcome the “we’re different” block by using a

process focus that looks beyond one’s own functional silo, or firm, or

sector; and adds incentives based on sharing, such as performance

appraisals and recognition systems.

• It reduces the “we’re busy” argument by providing helpful services.

The major difference between this approach and the “next level up”

is that this approach to infrastructure does not necessarily include a

trained cadre of facilitators and change agents dedicated to action and

stationed throughout the organization.

APPROACH #3: FACILITATED TRANSFER

Descriptive words: Facilitators, change agents, implementers, pro¬

jects, technical assistance, consultants, guidance counselors, sup¬

port personnel, brokers, coaches, shepherds.

The third approach to a knowledge transfer infrastructure is the “full

service level.” We call it “Facilitated Transfer,” and it provides all of

the previous activities and services, plus more.

This approach designates specific persons—full or part time—to

stimulate, assist, and encourage transfer of knowledge and best prac¬

tices; some even help with implementation. These change agents go

by all sorts of names, from facilitators to progress directors. We’ve met

brokers and cross-pollinators, not to mention boundary spanners, light¬

ning rods, and champions.

Creating the Knowledge Infrastructure • 119

Companies that opt for this full-service level assign specific transfer

and facilitation responsibilities to full or part-time employees in plants

and offices throughout an organization. These change agents are trained

to assist other employees in:

1. Finding, capturing, codifying, and transmitting knowledge and

best practices to the knowledge base

2. Helping fellow employees to solve problems and improve

processes by helping them to use the databases and other knowl¬

edge services of their organization

3. In some cases, acting as internal consultants to facilitate imple¬

mentation

This approach not only informs and enables transfer, it provides di¬

rect facilitation and help in the field for process improvement. The

focus is on implementation.

Organizations taking the facilitated approach search for their facili¬

tators from a cadre of people already in the organization. Some of the

sources are:

• Internal “change agents” or consultants

• Team leaders and facilitators

• Trainers

• Benchmarking facilitators

AMOCO CORP.

Amoco began its “Shared Learning” program by assembling a group of

people recruited from across the company, selected for their interest in

organizational learning. This first group has now been expanded at

every Amoco business unit where dedicated “change agents,” called

quality/progress professionals, are charged with helping employees

make Shared Learning work.

Progress professionals coach fellow employees in process improve¬

ment techniques and how to participate in building the database; they

provide training and spread the word, answer questions, and make sure

all Amoco employees have access to tools. In addition, Amoco has

“Networks of Excellence” made up of individuals who share a common

120 • If Only We Knew What We Know

business-related interest, and who meet regularly to discuss how they

are facing challenges.

Amoco’s Shared Learning network includes a three-tiered database

structure. But Amoco is quick to point out that Shared Learning is not

so much about databases as establishing networks of people who share

knowledge. “Connecting those who have with those who need” is a fa¬

vorite quotation from Dave Ledet.

Does all this work?

Attend any of their “ShareFairs” where employees get together to

celebrate success, recognize each other, and exchange lessons learned.

We guarantee it will make a believer of you. Or else sit in on their reg¬

ular “Lunch and Learn” meetings in small groups across the company

and feel and hear the enthusiasm and commitment.

TEXAS INSTRUMENTS

TI’s late chairman Jerry Junkins established an Office of Best Practices

to provide an ongoing infrastructure/mechanism to facilitate transfer.

The vision was to move from an environment of “I have to solve my

own problems” to one of looking for existing solutions not only from

across TI, but also from benchmarking best practices in other busi¬

nesses and sectors.

When TI first embarked on the development of the Best Practice

Sharing project, like any pioneer, it knew very little about knowledge

management and transfer and collaborative architectures. But through

experience, TI quickly realized that simply creating a database of

knowledge and best practices, publishing knowledge and best practices

in memos or booklets, and attendance at conferences or seminars,

would not necessarily—or even often—lead to action. TI decided it

needed to create a mechanism for people-to-people, face-to-face trans¬

fers instead. Hence, as part of the structure of the Office of Best Prac¬

tices, TI established a “facilitator network,” headed by Cindy Johnson.

TI’s best practice sharing team initially selected 138 “facilitators”

worldwide; eventually their ranks swelled to over 200. These facilita¬

tors are TI employees who are improvement-minded, who naturally

Creating the Knowledge Infrastructure • 121

“trade” and “broker” knowledge, who are trusted by their colleagues,

and who are good communicators. They are “pollinators” that make

connections. And the more the better, for according to the “Law of the

Telecoms,” the value of a network is directly proportional to the square

of the number of users.

The typical facilitator profile is:

• A leader; a source people often turn to for help; a team builder

• Interpersonal skills—ability to network well with people

• Knowledgeable about the benchmarking process

• Flexible; has a tolerance for ambiguity

• Comfortable with technology

• Some training and consulting skills

Each facilitator is given training in skills such as interviewing, root

cause analysis, categorizing and documenting, and information search-

ing. In addition, they learn how to access in-house and external elec¬

tronic resources. Facilitators also rely on their own network for

assistance in finding critical knowledge, and attend a quarterly one-

day forum where they network with other facilitators, share success

stories, and receive advanced training.

Facilitators can be found at every site and every business. Their role

as facilitators is part-time only, ranging from 10 to 25 percent of their

time. They still have “normal” jobs and report to their line or staff

managers, and they get no extra compensation for being a facilitator.

Rather, they are charged with the following missions:

• Collect best practices in their own area of activities

• Translate these into usable/transferable information

• Assign appropriate keywords (categories, topics, processes, etc.)

• Champion and publicize best practice sharing efforts

• Access the “KnowledgeBase” (TI’s best practices database) for solu¬

tions to problems

• Ask questions, see patterns, link practices back to business problems

• Show others how to access best practices information; identify

people they may wish to talk to or visit

• Assist in implementation of the practices if asked to do so

122 • If Only We Knew What We Know

WHERE DO FACILITATORS COME FROM?

One great source of potential change agents is the quality improve¬

ment (TQM) department or some other internal consulting organiza¬

tion. Such central groups support and coordinate a network of internal

“change agents” or consultants, usually with representatives in each of

the units. Amoco has one. TI has one. Chevron and Citibank each

boast an internal consulting office, as do many others. This central

“node” or corporate hub of the “improvement network” is typically

quite small and performs many roles including:

• Maintaining the corporation’s common language and framework for

change and improvement (for example, “TI-Best”; “The Chevron

Way”; Amoco’s “Shared Learning” and Amoco Business Model;

Motorola’s Six Sigma)

• Providing central information technology architecture for best

practices and other databases (the “best practices database usually

resides here)

• Conducting internal quality assessments using the Malcolm Bald-

rige National Quality Award criteria or a customized version of it

• Providing training, professional development, and support for the

facilitators, coordinators, and changes agents out in the business

units

• Providing access to external resources

• Educating senior executives on improvement initiatives

• Communicating the above to the entire organization (usually in

conjunction with the internal communications function)

• Acting as a liaison with the training and communication organiza¬

tion

NEXT LEVEL: THE SUPERSTRUCTURE

At some companies, the three-dimensional design of the KM infra¬

structure is anchored by a “superstructure,” a top-level KM champion

with formal and enterprisewide responsibility for coordinating and di¬

recting the company’s multiple transfer initiatives.

Such designated leaders and champions seem to be located both at

the corporate and the business unit levels (see Figure 11.2). They are

Creating the Knowledge Infrastructure • 123

fl c3 a s o

a 4> U J- 0) cu

Level Unit Level

FIGURE 11.2

Knowledge Management Champion

partnering with their information technology colleagues to adapt the

appropriate information technology solutions to fit the need.

Most recently, the business media pundits have given a lot of press

to the creation of a few chief knowledge officers (CKO) and chief

learning officers (CLO) in some major corporations, such as General

Electric, Coca-Cola, and Monsanto.

At Sequent Computer Systems, the chief knowledge officer has a

global, corporate responsibility for all knowledge management pro¬

grams and infrastructure. The CKO and his team provide corpo¬

ratewide infrastructure, corporatewide formal educational programs,

various operational support and program management and coordina¬

tion for knowledge management initiatives inside the company. The

CKO is also responsible for the corporation’s patent portfolio and for

traditional corporate research services, such as the corporate library.

Sequent’s manager of corporate research services, who reports to the

CKO, is responsible for the corporation’s physical library and for

SCEL, which is considered knowledge management infrastructure.

At Coca-Cola, CLO Judy Rosenblum manages groups of facilitators

around the world who help identify and transfer successful practices.

Rosenblum has a dual reporting relationship to the vice president of

human resources and the CEO.

At GE, Steve Kerr serves as the CLO and vice president of leader-

124 • If Only We Knew What We Know

ship development. Kerr, who reports to the head of HR, says: “This is a

terrific opportunity for human resource executives. Now they can

apply their expertise to help the company get better at a major new

thing.” The new thing? Transfer of knowledge and best practices.

How does a CLO differ from a CKO? Their jobs are on a continuum,

with a lot of overlap. If a company has adopted the mantra of becom-

ing a learning organization, it’s likely to appoint a CLO. If its rhetoric

refers instead to leveraging knowledge as an intellectual asset, it may

opt for a CKO. To send the message that shared learning or knowledge

management is a high priority, the CLO or CKO reports to the top,

most often to the CEO.

So far, the CKO and his or her namesakes are the exception, not the

rule. But that does not mean companies are not serious about their KM

work. Andy Michuda, CEO of Teltech, says his firm’s research has

found no correlation between the existence of a CKO (or equivalent)

and the effectiveness of the KM projects. Indeed, he says, “some 45

percent of the companies we meet may not have a CKO, and they

‘may not talk the talk’ but they walk the walk; they may not have formal

management positions but they’ve got this readiness and grass-roots ef¬

fort.” Teltech’s research confirms ours: The real action in KM is not hap¬

pening at the CKO level. And when the media choose to focus on fancy

new titles they miss the main event: the grass-roots, organization-wide

proliferation of KM practitioners, from editors to facilitators, who pro¬

vide the real engine of knowledge.

WHICH APPROACH IS RIGHT FOR US?

First, understand that there isn’t a single right approach. The three de¬

sign approaches discussed here are more like steps leading to a “full-

service” or multidimensional solution. Leading KM companies have

elements of all three populating their infrastructure.

Second, critical infrastructure choices are (or should be) influenced

by geography, culture, money, technology, leadership, market struc¬

ture, philosophy—but most of all, by the answers to these three ques¬

tions:

1. How important is transfer of knowledge and best practices in the

strategy of the organization?

Creating the Knowledge Infrastructure • 125

2. How much assistance and intervention does the organization

think is required to make transfers happen and get results in their

organization in a timely manner?

3. How does the infrastructure address the six problems with the

flow of knowledge and its implementation?

We reiterate that the three approaches—self-directed, knowledge

services and networks, and facilitated transfer—are not discrete. They

depict a continuum, along which organizations choose the mix of ac¬

tivities. Which is better? Each organization has to balance the answers

to these questions with their resources, their strategy, and their belief

in the importance of knowledge and best practices transfers in the

years ahead.

Chapter 12 '

MEASURING THE IMPACT OF TRANSFER

The fundamental building material of a modern corporation

is knowledge. Using knowledge to make money is the real

challenge.

—Valery Kanevsky, Hewlett-Packard Company

Some experts wax sentimental about the benefits of transferring

and managing knowledge and the all-around warm and fuzzy feel¬

ing of sharing and learning. No argument here. We love sharing. And

learning is great fun. But the business world is not (only) about having

fun. Realists (and most managers are of that ilk) know that the key is

figuring out how to better manage knowledge assets to maximize their

returns.

It’s as simple as that. And it’s as complex as that.

“Maximizing returns” implies that we can: (1) measure the “princi¬

pal” investment, and (2) measure the yield from that investment on

some periodic basis. Both are debatable when it comes to putting fig¬

ures on things as intangible and ephemeral as the value of knowledge

and best practices sharing.

Debatable, perhaps. But not impossible!

And so, whereas measurement is the least developed area in the

126

Measuring the Impact of Transfer • 127

emerging knowledge management “discipline,” we believe it to be crit¬

ical, as well as an area where true leadership can make a real difference.

CLUES TO VALUE

Knowledge is beginning to be widely accepted as the atom-equivalent,

basic ingredient of today’s competitive advantage, and this presents

KM practitioners with a true challenge. Like early quantum mechanics

theorists, they know it is there but they cannot see it. “A scientist had

to imagine uncountable corpuscles banging invisibly this way and that

in the soft pressure of wind against his face,” writes James Gleick, in

Genius, his biography of physicist Richard Feynman. Like early physi¬

cists, exponents of KM must infer the existence of knowledge, assume

its potential and powers by observing its impact on other forces, such

as the forces of competition and innovation, stock market valuations,

employee morale productivity, and product excellence. Here are three

clues to potential “value” of knowledge within a corporation:

Clue #1: Stock Prices

The stock market routinely assigns higher valuations to “intelligent

firms” versus ones with stockpiles of hard assets, like brick and mortar.

The gap between the value of a firm’s tangible assets and its market cap¬

italization implies there’s some other asset at work: namely, knowledge.

Sometimes, knowledge is the only asset. For an extreme manifesta¬

tion of what knowledge can be worth, take the mid-1990s spate of suc¬

cessful “concept” IPOs (Initial Public Offerings). The Netscapes of

this world. These are technology start-ups, which include a bunch of

smart guys with a couple of great ideas. No revenues. Negative cash

flow. Yet they command hefty dollar valuations from such savvy—and

not uncynical—folks as venture capitalists and Wall Street bankers.

What gives? What investors pay for is their assumption about the

worth of the collective IQ of the new enterprise. Those who invested

in Netscape have not been disappointed.

The same applies to more mature firms such as Coca-Cola, Micro¬

soft, and Sun Microsystems. These companies routinely trade at many

times their book value—and usually at higher multiples than industry

128 • If Only We Knew What We Know

peers. The gap between them and the next guy represents a knowledge'

based competitive advantage. They take more intelligent action based on

what they know—about markets, about products, and about cus¬

tomers. About how to share and innovate.

The combined experience and “capital” of the organization—some

in the heads of workers, some in the brand loyalty of customers, and

some embedded into the structure of the organization—represent a

growing portion of what a company is worth. Often, the intellectual

assets and knowledge capital of a company are worth three or four times

its tangible book value!

When Apple Computer bought Next for $400 million, it did not

buy plants or buildings, points out Knowledge Capital™ measurement

expert Paul Strassmann, “It bought $400 million worth of knowledge.”

Clue #2: Performance Improvement

Stock market valuations are good only up to a point. We all remember

J. P. Morgan’s famous adage: “The markets will fluctuate.” And boy, do

they. When the market loses 10 percent of its value on a day in early

spring, do the underlying companies become instantly 10 percent less

intelligent? When the Dow Jones technology index drops 150 points

in an afternoon, does the collective IQ of these companies’ employees

suddenly plunge?

Obviously not.

But whereas market value proxies are beneficial (for starters, they

tell you just how much the market is willing to pay for smarts) they are

but a proxy for the Real Thing. The real potential returns on optimal

utility of knowledge assets lie in corporate performance measured

through a plethora of competitive, operational, process, and financial

measures.

Companies that know what they know can make better decisions—

faster. Those that have mastered their collective intelligence develop

better products and have a chance at capturing new markets—first. Let

us repeat these three mantras of modern-day competition: Faster! Bet¬

ter! First! “What used to take weeks now takes hours. What used to

take months now takes a week,” says Mark Koskiniemi, vice president

of human resources at Buckman.

Measuring the Impact of Transfer • 129

Clue #3: The Cost of Not Knowing = “CONK”

There are lots of examples where information and knowledge were

somewhere in the organization, but not where they needed to be to

avoid a disaster. When the Challenger space shuttle exploded, some en-

gineers at NASA knew the O-rings wouldn’t hold at low temperatures.

But the people who made the decision to launch did not. A company

may launch a new service in China, but if the new team can’t build on

past experience in that market, it will make all the same mistakes

again. A new movie comes out and preview audiences hate it, but that

information comes too late to fix it. Sometimes, the cost of not manag-

ing knowledge is easier to pinpoint than the positive contribution of

effective management. Easier. But can you afford it?

MEASURE VS. NURTURE

It’s not entirely surprising that measurement is the least developed as¬

pect of knowledge management and best practice transfer efforts. In

fact, many practitioners believe that trying to measure before you un¬

derstand how knowledge gets created and shared may lead you to focus

on the wrong things.

To date, there are two general schools of thought on measuring the

transfer of knowledge and best practices.

The first school can be called the “Nurturers.” Its adherents believe

measurement of knowledge management and transfer efforts is cur¬

rently premature. Proponents of this point of view hold that not

enough is yet known about the dynamics and the impact of knowledge

to justify elaborate measurement systems. Hence, measurement at this

early juncture can be risky and misleading, and the results suspect.

The second school, the “Quantifiers,” is suspicious of the first

school, and believes that measurement is important, both for under¬

standing and legitimizing investment. It wants to know where and

how to invest.

We probably edge over slightly into the second school. We believe

measurements are key to ensuring the sustainability and success of

transfer efforts over time. Top management will not continue to invest

if no tangible or quantifiable intangible results can be demonstrated.

To keep transfer efforts going, and centered, companies must continu-

130 • If Only We Knew What We Know

ously link their process to the desired outcome—their original value

proposition. Without that link, efforts will likely lose their direction

and purpose.

Are we “quantifiers”? Not quite. In fact, we are quite sympathetic to

the “nurturers.” It may be more productive in the early stages of trans¬

fer efforts to measure in order to observe, monitor, and nurture. It may

be better to celebrate successes, early and often, and only later work

out elaborate schemes for measuring knowledge management. Many

KM pioneers, like Sequent Computer Systems, believe that trying to

measure before you understand how knowledge management is work¬

ing in your organization may cause you to focus on, and reward, the

wrong things. So understand first; measure second.

Ultimately, we believe everything depends on what kind of measure¬

ments you are talking about and for what purpose.

If you want to know whether transfer efforts are achieving their ob¬

jectives, identify the business results that match your original value

proposition and measure those.

If you want to know which of your transfer tools and applications

are the most effective for the purpose of sharing practices and know¬

how, measure the level of transfer activity, and ask users how it has

helped them achieve business objectives.

Finally, for a comprehensive view of your knowledge management

activities: measure both!

A PRACTICAL APPROACH TO MEASUREMENT

Whereas there are evolving and highly sophisticated ways to measure

knowledge that go beyond traditional accounting, we believe that cur¬

rently the best way to measure the impact of knowledge and best prac¬

tice transfer is not by gauging the size of your knowledge capital base,

but rather the effect it has on your company’s performance.

And we are not alone. Indeed, measuring the benefits and results of

knowledge management per se does not seem to be prevalent in the

firms we have studied, which includes most of the leaders in knowledge

management (see Figure 12.1). Practitioners tell us it is more impor¬

tant to measure the success of the projects and business processes that

are being improved through the transfer of knowledge and best prac¬

tices. Hence they attempt to link the outcomes of these efforts to their

Measuring the Impact of Transfer • 131

Companies with Defined Measurements of the

Effectiveness of the KM Strategy

FIGURE 12.1

Companies with KM Mensures of Effectiveness

original value proposition. Each comes with its own set of “process”

and even financial measures. Hence we believe in measuring processes,

activities, operational outcomes.

As any statistician will tell you, however, if transfer and sharing is

made part of work, and performance improves, there are not enough

regression analysis tools in existence to help you figure out the relative

contribution of knowledge management versus other factors.

Although they may not be measuring the impact of their overall

knowledge strategy, most firms are using some combination of yard¬

sticks to gauge the success of their transfer projects. In fact, most can

clearly describe process improvement outcomes in projects in which

knowledge management approaches and applications were used. The

most commonly mentioned improvement outcomes were improve¬

ments in process cycle time and quality of products, and business

growth from the production and success rate of proposals and increased

customer satisfaction (see Figure 12.2).

MEASURING THROUGH OUTCOMES

Since most of the companies with which we work report that their

knowledge management efforts are tightly linked to business objec¬

tives and business needs, most prefer to measure process and project out¬

comes .

132 • If Only We Knew What We Know

7-r

6- /

5- Cfl

'2 03 > Cl 2 4- X 6 O

J 3' i /

2-

1-

0- z

"X"

n i i

r^r i i i i i i i i

Increased

Innovation

Business

Growth

Practice &

Process

Improvement

Increased

Customer

Satisfaction

Enhanced

Employee

Capability/Org.

Learning

Result

FIGURE 12.2

Outcomes Realized as a Result of KM Practice

For knowledge management applications to be perceived as success¬

ful by their users, they need to help users do their job or achieve an im¬

mediate business objective. Because this measurement is embedded in

the business case scenario, the success of the project serves as a proxy

for the success of the knowledge management application. In other

words, knowledge management is seen as a tool for enhancing a busi¬

ness or improvement process.

Companies like Buckman and Sequent have implemented enter¬

prisewide systems to support the sharing of knowledge and practices,

especially in the sales and marketing arena. They focus on the impact

on the value proposition, not the success of knowledge management

individually.

Example 1: Sales from New Products.

A good example of measuring business results is Buckman’s global

measure of sales from new products. In March 1992, Buckman Labora-

Measuring the Impact of Transfer • 133

TYING MEASURES TO THE VALUE PROPOSITION

When measuring the impact of transfer efforts, take into account your

original goal. Each value proposition comes with a set of logical “mea¬

sures” that help monitor your progress toward that goal, these could in¬

clude:

Customer intimacy

• Customer retention rates

• Number of calls handled per day

• Number of calls resolved on the first “sitting”

• Cross-selling penetration

• Increased revenue from existing customers

Product leadership

• Revenues from commercialization of new product

• Percentage of revenues from new products

• Time-to-market cycles

• Ratio of successful to unsuccessful product launches

• Number of launches per year

Operational Excellence:

• Cost per unit

• Productivity and yields

• Number of defects/poor quality

• Production cycle time

• Inventory carrying costs

• Environmental compliance

• Safety records

tories International introduced K’Netix, a corporate-wide knowledge-

sharing communication system. One of the value propositions behind

K’Netix was improving product commercialization by linking sales

people to customers and R&D. To measure how successful its efforts

have been, Buckman examines percent of sales generated from new

products (less than five years old). The results, which speak for them¬

selves, are shown in Figure 12.3.

Buckman Laboratories International links this improvement par¬

tially to the KMIT efforts:

• The speed of response to customers is hours, not days or weeks.

• The quality of response has risen all over the world.

134 • If Only We Knew What We Know

Percent of Sales from New Products

1988-1992 (prior to K’Netix): 23.6%

1992-1996: 33.3%

FIGURE 12.3

• Office costs are down; the office is anywhere.

• Employee competency has increased, since more employees are im-

mersed in high-intensity projects.

• The role of managers is not to manage the flow of information up

and down the organization, but to close the gaps.

Ultimately, Buckman believes it impossible to put a monetary value

on the knowledge network; it is fundamental to the way the company

operates. Most of the companies we work with are highly sophisticated

in the use of process, project, and enterprise-level measures. Many,

such as CIGNA., use a balanced scorecard or family-of-measures ap¬

proach.

Example 2: A Balanced Scorecard

The rationale behind the balanced scorecard (BSC) system was that

learning was directly linked to process performance, which in turn af¬

fected corporate performance. At CIGNA Property & Casualty, the

BSC is a means to measure and communicate with four aspects of the

organization’s effectiveness: financial, internal business, external busi¬

ness (customer), and learning and growth. To construct the model,

CIGNA P&C needed to provide a set of corporate goals and direct

performance measures as a framework.

To enhance the BSC results, CIGNA identified eight main compe¬

tencies that needed improvement. These included the abilities to:

1. select and enter new markets with above-average profit potential

2. select and attract the right kinds of customers

Measuring the Impact of Transfer • 135

3. select and manage the producers for those businesses

4. better identify exposures and hazards

5. drive pricing more accurately toward the right risk assessment

6. form partnerships with customers in order to reduce claims fre-

quency

7. reduce claims severity

8. manage claims performance to the required service level of each

business unit

CIGNA can now measure improvement by measuring change in

each of the eight categories. Overall, CIGNA’s consistently profitable

results point to a dramatic turnaround, a result of improved decision

making and faster work flow.

MEASURING THROUGH ACTIVITIES

Whereas management, customers, and employees care about how well

the process is working and the results that are obtained, the architects

of transfer initiatives must understand the activities behind the sys¬

tems. Hence, this second group of measures indicates how frequently

users are accessing, contributing to, or drawing on the information

technology tools that enable transfer enterprise-wide.

Most of these measures reflect an Internet or Web browser para¬

digm, such as hits per page or submissions per employee. There is value

in these activity measures. They can lead to a greater understanding

how, or if, a tool or support system of the transfer activity is being uti¬

lized.

A caution: though activity-based measures provide useful informa¬

tion on accessibility, utilization, content quality, and design features,

they do not provide information about the impact of these activities

on results.

Common activity measures include:

• user rating of effectiveness

• number of hits

• participation rates

• frequency of contribution

• frequency of use

136 • If Only We Knew What We Know

Arthur Andersen, for example, measures usage and user ratings of

effectiveness and solicits user “success stories such as faster report pro-

duction (by using existing knowledge) to illustrate the success of its

transfer systems. TI, like many other firms, counts the times employees

log in to its best practice databases. Most recently, the hits rate was

about 2,000,000 per month. Increases in hit rate indicate KM is pene¬

trating the consciousness of a greater number of employees and be¬

coming endemic to the way “work is done.”

Ultimately, to gain a comprehensive view of transfer efforts success,

firms rely on the combination package of outcomes and activity mea¬

sures.

ACTUAL COSTS AND RETURNS

To add yet another level of “measurement,” some firms track the actual

cost of KM and transfer projects. Those are notoriously hard to pin

down, because they are often dispersed throughout the organization,

and can “hide” in places like IT, marketing, HR and training, as well as

management time and the efforts of facilitators (see Chapter 11, which

treats the knowledge infrastructure).

However, since there are costs associated with developing support

systems, it helps to measure them. For example, at Sequent, the total

spending on KM infrastructure (people and technology) is well under

$1,000 per employee per year, including all direct costs and some por¬

tion of the indirect costs. Buckman spends about 3.75 percent of rev¬

enue on knowledge management and transfer activities, across the

board. Hence, the all-inclusive cost per employee is about $7,500 per

year for hardware, software, telephone, network, and staff support.

Obviously, the range is quite broad. And just because Sequent cal¬

culates a spending of $1,000 per employee does not mean its transfer

efforts are less enthusiastic or successful (as their case study in Chapter

17, page 170, illustrates). The difference in per-employee charges re¬

flect more accounting (what gets counted as KM-direct expense) than

effort.

Measuring the Impact of Transfer • 137

What Gets Counted?

Knowledge exchange and transfer applications (see Chapter 10 on

using information technology) incur less up-front hardware and soft¬

ware cost to deploy (if you have the Intranet and/or Notes infrastruc¬

ture already deployed), but keep in mind that they may require

significant ongoing support costs to be successful.

These support costs include:

1. facilitating the formation and health of communities of practice

and discussion groups

2. populating best practice databases

3. creating information technology standards for format and infor¬

mation and document management

4- advertising the existence of groups and experts

5. developing policies and procedures for appropriate use of infor¬

mation and dialogue

PERSPECTIVES ON MEASUREMENT

The bottom line is the bottom line. So if your company is doing much

better, overall, just how concerned should you be about measuring the

impact of transfer?

Very concerned.

Although measurement of the success of KM and transfer initiatives

is perhaps the least developed area in this evolving field, it is a critical

component of the creation an environment that both encourages and

sustains sharing. Without measurable success, enthusiasm from employ¬

ees and management will dissipate. And without measurable success,

you won’t be able to tell what works and what doesn’t.

Just how to measure the impact of transfer is a subject of much de¬

bate. We applaud the efforts of “quantifiers” who have managed to cal¬

culate what knowledge is worth. But we recommend the use of

practical process-outcome and activity measures, which, combined,

give you a realistic assessment of “how you’re doing.”

Ultimately, knowledge about your knowledge management initia¬

tives is important because it helps you:

138 • If Only We Knew What We Know

• Design future systems and applications

• Improve the current sharing processes

• Ensure the transfer effort stays on track (that is, delivering the value

proposition!)

As the role of knowledge managers within organizations grows,

these measures will provide the underpinning of their performance

evaluation and, hence, compensation.

If you cannot measure it, can you manage it? We don’t think so.

Part Four

REPORTS FROM THE FRONT LINES

PIONEER CASE STUDIES

Some companies have successfully adopted internal transfer of knovul-

edge and best practices as a competency and improvement strategy.

They’ve stopped dreaming, debating, and discussing it. They are doing it.

From our work with over seventy organizations, we have chosen four to

tell you about in detail in this section: Buckman Laboratories, Texas Instrw

ments, the World Bank, and Sequent Computer Systems. We chose them as

much for their differences as for their similarities. Their stories illustrate there

is no “right” way to leverage knowledge. No “cookie cutter” design. No one-

sizc'fitsmil approach. Each of these four pioneers has chosen a different path

to climb the mountain, but all four are setting their sights on the same peak.

What’s it like up there?

For starters, it’s getting crowded. While some companies attempt this

journey with little idea of what awaits at the top, this part gives you an early

and rare peek at the peak. The birds-eye view from the top. This is your

chance to glimpse how effective management and transfer of knowledge has

transformed four organizations. You can do it, too. We tell you more

about how in Part Five.

Chapter 13

THE VIEW FROM THE TOP

Consider these sky-high achievements:

Privately held specialty chemical manufacturer Buckman Laboratories

has retrained its global sales force and watched sales of new products climb.

How? By creating an electronic best-practice sharing engine that al¬

lows everyone to learn just enough, just in time. “Since it involves the

entire company to get an improvement of this magnitude, and particu¬

larly on the interface with the customer, I believe that this is solid evi¬

dence of an ability to satisfy customers better and faster. In other words

the productivity of the entire company went up,” says CEO Bob Buck-

man.

Giant'Size Texas Instruments has rediscovered growth and agility a la Sil¬

icon Valley startup.

How? By weaving a connective tissue of best practice sharing which

has breathed new life into its reengineered, downsized corporate body.

“We felt we need to find a new paradigm for reaching the next level of

improvement,” says Cindy Johnson, director of TI’s Office of Best

Practices. “We had to find ways to become more agile and learn faster

so that we can innovate faster than our competitors.”

The World Bank is redefining how developing nations attempt to resolve

poverty and redress poor living conditions.

How? By pulling regional resources and specialist know-how and

making it available to experts throughout the organization. They now

know what works and what doesn’t. They know what they know. “The

Bank Group’s relationships with governments and institutions all

over the world and our unique reservoir of development experience

across sectors and countries position us to play a leading role in [a] new

141

142 • If Only We Knew What We Know

knowledge partnership,” said president James Wolfensohn. . . We

need to become, in effect, the ‘Knowledge Bank.’”

Midsize Sequent Computer Systems has morphed itself into a big-time

player in the high end of the UNIX market.

How? By leveraging internal knowledge and transferring best prac-

tice among its front-line employees. “With the kind of competition we

have, we have to act and look bigger than we are, and the only way to

do that is by sharing information,” says Roger Swanson, manager of

Corporate Research Services at Sequent.

ALL RIVERS LEAD TO THE SEA

Look once . . . These four organizations seem to have little in common.

But now, look again . . . While different in some ways, all four have

adopted a methodical and strategic approach to the sharing of knowl¬

edge and best practices. They have defined the business objectives;

sketched out a technology and organizational infrastructure; targeted

the components of the knowledge-process (embodiment and dissemi¬

nation) they seek to manage. They have begun to manage and measure

the process of KM with a sensitivity to culture and an appreciation of

the possible pitfalls.

Perhaps the most intriguing commonality among these four organi¬

zations is the emerging outward focus. As Peter Drucker once said,

“The purpose of an organization resides outside the organization.”

While initially they focus on helping knowledge and best practices

flow seamlessly in-house, they plan ultimately to open up their collec¬

tive IQs to the browsing of external constituents.

Both Sequent and the World Bank recognize that distributors, al¬

liance partners, regional banks, outsourcing services providers, ven¬

dors, suppliers, and customers will gain from gaining access to their

knowledge bases. More to the point, they recognize they, too, stand to

gain much by forming more intricate, knowledge-intensive ties with

the external business environment. This is not about selfless sharing. It’s

about sharing in order to win.

And winning has never been tougher or more complex. As Brandeis

professor Ben Gomes-Cassares explains in his book, the Alliance Revo¬

lution, today’s competitive environment operates by new rules. “The

The View from the Top • 143

entire nature of the competitive environment has changed,” he says.

“Markets are more fragmented. Economies of scale more difficult to

achieve. Companies no longer compete one against the other. Rather,

they do battle in ‘constellations.’”

What holds these constellations together? What is their center of

gravity?

They are held together by the binding ties of knowledge. Those

companies that share more have longer-lasting, more intimate, more

successful relationships with the planets around them, be they cus¬

tomers, suppliers, or partners.

A PASSION FOR KNOWING

Finally, there is one more reason we wanted to share these four stories

with you. Their passion for knowing. At all four organizations, KM

leaders speak with a contagious enthusiasm about their work. And it is

that passion for knowledge and sharing which seems to permeate

throughout the organization that makes Buckman, Sequent, the

World Bank, and Texas Instruments ideal examples of knowledge in

action.

You should be able to find yourself in one or all of them.

Chapter 14 .

BUCKMAN LABORATORIES

EMPOWERED BY K’NETIX®

An off-the-shelf CompuServe application is the neural knowledge

network that energizes Buckman Laboratories. The Memphis-

based Buckman aims to arm each associate, particularly front-line sales

folks, with the collective IQ of the organization. And it relies on

K’Netix , the Buckman knowledge network, to do the job. Launched

in 1992, this electronic buffet of best practices, discussion forums and

“just-in-time” training is credited with speeding up response time,

increasing new product development, and boosting overall growth.

Psychologists often refer to it as the “Ah-ha!” moment; it is that

split second of realization when an insight hits and irreversibly alters

____our paradigm of how the world works.

Necessity is the mother of invention. For Buckman Laboratories, a specialty

___ chemical company, the moment of

enlightenment came in the late 1980s when Bob Buckman, chairman

of BuLab Holdings, Inc., parent company of Buckman, was at home re¬

cuperating from a back injury. Two weeks in bed had rendered the en¬

ergetic chairman utterly frustrated with his inability to “know” what

was going on at work. He realized that others must be feeling similarly

distressed by knowledge vacuums.

Buckman began to envision a knowledge-driven company linked

globally by an electronic network offering ready access to best prac¬

tices, experiences, and skills to Buckman’s 1,200 associates (employ¬

ees) in eighty countries. Upon his return to the office, Buckman

turned this bedridden fantasy into viable commercial reality—with

stunning results.

144

Buckman Laboratories: Empowered by K’Netix® • 145

Since the 1992 introduction of K’Netix®, Buckman’s KM and

transfer of best practices system, new product-related revenues

are up 10 percent and sales of new products are 50 percent higher.

Response time to customer queries is down to hours, instead of

days or weeks. “We have seen the payoff,” says Mark Koskiniemi,

Buckman’s vice president of human resources.

BUCKMAN BEGAT KTD

Buckman’s KM value proposition is providing associates at the point of

customer contact with instantaneous access to the collective IQ of the

organization. Thus armed, associates can make intelligent and imme¬

diate decisions that affect customer relations and—ultimately—rev¬

enue. “This [KM] system is dedicated to the front lines,” says Bob

Buckman. “It’s a library of the important information relevant to the

business of the customer.”

Buckman’s first structured effort to share and capture knowledge

began in 1988 with the initiation of a case history system. This elec¬

tronic best practice reservoir helps resolve customer problems faster

and more efficiently by first applying existing “Buckman knowledge”

about what works. By mid-1997, Buckman associates could access

close to 2,500 customer case histories—1,787 in English and 685 in

Spanish.

What did Buckman learn? That sometimes you need a KM-dedi-

cated group to get the ball rolling.

In March of 1992, Buckman reorganized further to optimize

knowledge sharing through the creation of a Knowledge Transfer De¬

partment (KTD). The KTD consolidated separate IS and telecommu¬

nications departments and merged with the technical information

center. KTD’s mission is to proac- -

tively respond to Buckman’s global What did Buckman learn? That

knowledge needs by planning and sometimes you need a KM-dedicated

managing the resources necessary to group to get the ball rolling.

rapidly disseminate collective indus- -

try, technical, and market knowledge to all associates. More specifi¬

cally, KTD is charged with:

146 • If Only We Knew What We Know

. Accelerating the accumulation and dissemination of knowledge

within the company , ,, , , . Providing easy and rapid access to the company’s global knowledge

bases

. Sharing best practices with all Buckman associates

AND KTD BEGAT K’NETIX

If KTD is the corporate “ nerve center,” then the neural network it em¬

ploys to speed knowledge throughout the corporate organism is

K’Netix. Launched in 1992 using the CompuServe platform, K’Netix

pulls together disparate technologies to allow twenty-four-hour, die -

of-the-icon access from the desktop (or laptop) to associates world¬

wide, Users can dial in using a local CompuServe number, log onto the

ANATOMY OF A SYSTEM

K’Netix is Buckman’s trademarked term for the way they use the Knowl¬

edge Sharing interface (K’Netix is not an application-it is a system of

applications, much like an Apple “Launcher” or the Windows 95 “Start

menu). It functions like connecting tissue which brings together differ¬

ent programs and systems, some old, some new, some compatible and

some not. It contains electronic forums, on-line libraries, a knowledge

base, electronic mail, Internet/World Wide Web, intranet, project track¬

ing systems, customer relationship management systems, groupware, bul¬

letin boards, virtual conference rooms, and databases that capture

institutional memory that is then made accessible to all employees. Ac¬

cess to K’Netix is available through the K’Netix Access Menu (KAM),

which allows users to tap into the following components:

• BL Systems contains component databases that relate to the business

operations and systems such as accounting and order management.

• The Buckman folder contains an accumulation of technical knowl¬

edge to solve customer problems. Within this “folder” are forums that

are the central point of the knowledge system—TechForum, Chem-

Forum, EuroForum, AAAForum for Asia, Australia, and Africa, and

ForoLatino. Each of these facilitates the communication of informa¬

tion throughout the organization worldwide. Because worldwide oper¬

ations necessitate communication on various timelines, associates can

communicate on-line, leave messages, and set appointments for real-

Buckman Laboratories: Empowered by K’Netix® • 147

Buckman server, and access everything from electronic forums to on¬

line libraries, electronic mail, the Internet/World Wide Web, project

tracking systems, customer relationship management applications,

bulletin boards, and virtual conference rooms. (See page 144.)

Technology purists may balk at this IT hodgepodge as well as at

relying on an external vendor for both network backbone and

some actual applications. It is certainly not “elegant,” in IT terms,

and proves a killer when it’s time for an upgrade. But “piggy¬

backing” off CompuServe’s existing network infrastructure meant

Buckman was ready to launch in only thirty days in 1992!

time interaction. The dialogue in forums proceeds without filtering,

but the best threads are captured by technical experts and put in the li¬

brary.

• Also contained in this folder is MIDAS (Marketing Information Data

Analysis System), which contains confidential information on cus¬

tomers and processes and how Buckman associates interact with these

customers to solve problems and improve their operations.

• Customer Forums allow a specific customer’s employees worldwide

the ability to communicate among themselves and with Buckman as¬

sociates. Access to these forums is restricted to people who are spe¬

cially nominated by Buckman and the customer. A separate “shadow”

forum exists for each; accessible only to Buckman associates, the lat¬

ter is invisible to the customer.

• The Customer Information Center (CIC) is an unstructured database

used by front-line sales associates to store electronic copies of docu¬

ments specific to a certain customer. Access to this data is restricted to

the account representative and his/her manager, and other marketing

support groups.

• The most recent addition to K’Netix is the BuLab Learning Center,

which facilitates Buckman’s distance learning project. This is in

essence a distributive learning tool with interconnected networks of

Notes databases. The purpose is to deliver asynchronous, just-in-time

education and training to associates throughout the world.

148 • If Only We Knew What We Know

Every IT architecture has its drawbacks. Buckman s choice of sys¬

tems and applications is dictated by need, not protocol. The IT archi¬

tecture has evolved over time, using what s available off the shelf

whenever possible to reduce capital expenses. “If we can buy it, we

don’t develop it,” says Buckman. Each application has been chosen to

fulfill a specific KM need, based on its inherent features.

An Internet approach, for example, would not work well for train¬

ing, since the latter requires large amounts of data and graphics—thus

--- long connect and download times. So

Never judge a book by its cover, or a for global access to just-in-time train-

KM suite by its platform, ing, Buckman relies on Leaming-

---Space from Lotus Development Inc.,

a distributive learning tool with interconnected networks of Notes

databases.

In contrast, Buckman’s BuLab TechForum, one of seven forums—

and “the engine of our knowledge sharing,” as Buckman refers to it—is

an off-the-shelf application provided by CompuServe. The application

allows up to twenty-one different message and library sections, mostly

topic-related.

Discussions proceed globally and directly, while at week’s end, a

“section leader” is charged with capturing conversation “threads” and

posting them in a “library.” By piggybacking CompuServe’s application

and global network reach, Buckman has worldwide connectivity at a

fraction of the cost. All 1,200 associates have CompuServe IDs and

passwords and they use the network for both intra- and intercompany

communication. “It works anywhere anytime. From the desktop and

from the laptops,” says Koskiniemi.

Although the buy-not-build philosophy keeps infrastructure costs

in check, K’Netix is certainly not cheap. Buckman spends about 3.75

percent of revenue on its integrated KM and IT activities each year,

not including senior management time. The per-employee price tag is

about $7,500, covering IT, telephone charges, hardware, and software

licensee fees. But the benefits, in terms of improved customer contact

and relationships, vastly outweigh the cost. “In total,” says Koskiniemi,

“we spend about $10 million. Can you spend more? Sure. Can you

spend less? Sure. The key is that there is business we got that we would

not have won if this system was not in place.”

Buckman Laboratories: Empowered by K’Netix® • 149

THE OTHER 95 PERCENT

Technology, however, is only 5 percent of the KM equation at Buck-

man. The idea of sharing knowledge, sometimes with across-the-ocean

strangers one has never met, is “90 percent culture, 5 percent technol¬

ogy, and the rest is magic,” says Buckman. Fortunately, Buckman labs

had a sharing-oriented culture even before KM became a corporate

strategy. The evidence is in its code of ethics:

1. The organization is composed of unique individuals with differ¬

ent capabilities, all necessary to successful operation of the com¬

pany.

2. Individuality will be acknowledged by treating one another with

dignity and respect and maintaining the focus on continuous and

positive communication among all employees.

3. The contributions and accomplishments of all employees will be

recognized irrespective of magnitude.

Whereas respect and recognition are critical, they are not neces¬

sarily formalized in the manner of fatter paychecks or bigger

bonuses at Buckman. Sharing is simply part of everyone’s work.

Forum Section Leaders, who either volunteer or are recruited for

this weekly editing job, receive no special form of compensation.

“The only status involved is the ‘SL’ at the end of your title,” says

Koskiniemi.

Top management sets the tone for sharing. “Some modes of encour¬

agement have been more subtle than others,” says Burrows. CEO Bob

Buckman is known to have printed -

out weekly lists of TechForum partici- Bingo! To get folks to share, show

pants. Associates who were not on them the way by sharing first.

the list would quickly get the message: -

“If I am not in there, sharing my knowledge, exactly what use am I to

the organization?” says Burrows. At the same time, 1993’s most active

forum participants got a surprise trip to a conference in Arizona plus a

$150 leather bag and an upgrade to an IBM Thinkpad 720. “Both have

since become status symbols unto themselves,” reveals Koskiniemi.

150 • If Only We Knew What We Know

The net result is that sharing has become endemic to the organization.

Of course, some associates are more dedicated than others. It certainly

helps that management leads by sharing: Bob Buckman was sitting in

the back of the room [at a meeting in Arizona] writing a synopsis as the

meeting was unfolding to post in BuLab News as soon as it was over.

MEASURING SUCCESS

Buckman began its systematic KM and best practice transfer efforts

with a clear goal in mind: It wanted to be able to get closer to its cus-

tomers, faster. Consequently, the measurement of its KM initiatives

centers on the interface with customers.

On an organizational (collective) knowledge level, Buckman mea¬

sures the system’s value in the number of customer contracts it helps

the company win. It also measures its effectiveness through the growth

in new product related revenues. In the four years prior to the launch

of K’Netix in 1992, the average sales of products less than five years old

was 23.6 percent of total sales. In the four years following K’Netix in¬

troduction the average rose to 33.3 percent. Buckman also credits

much of its 250 percent growth in sales in the past decade to its on-line

KM systems.

“Since it involves the entire company to get an improvement of

this magnitude and particularly on the interface with the cus¬

tomer, I believe that this is solid evidence of an ability to satisfy

customers better and faster. In other words the productivity of the

entire company went up,” says CEO Bob Buckman.

On an individual-knowledge level, Buckman monitors the education

of the portion of the work force engaged on the front line. Seventy-

nine percent of them now have college degrees, up from 39 percent in

1979. At the time, Buckman’s sales force numbered 79, compared to

over 600 today.

Operational measures provide additional evidence of success:

• The speed of response to customers has increased; it now takes

hours, not days.

Buckman Laboratories: Empowered by K’Netix® • 151

The quality of response has risen all over the world.

On-line remote access to knowledge means the office is anywhere.

Growth of talented people has in- -

creased; more employees are im¬

mersed in high-intensity projects.

Still skeptical? So were we, at first.

But not after we looked at this

list of big benefits! “What used to take weeks, now

takes hours. What used to take

months, now takes a week,” says Koskiniemi. “That means we are re¬

sponding to the customer, faster.” Ultimately, Buckman believes it im¬

possible to put a dollar-and-cents figure on its knowledge network.

Rather, it is more a central tenet to the way the company as a whole

operates. “Do you want to measure it before you get into it?” asks

Koskiniemi. “Not really. I think you just do it, and measure later. If you

wait for the measure to take action, you may find it is too late.”

Chapter 15

TI’S BEST PRACTICE SHARING ENGINE

Slimmed down, reorganized, and restructured, Texas Instruments

entered the early 1990s looking for a new way to generate perfor¬

mance improvement. Building on their initial success in sharing best

practices across semiconductor fabrication plants, TI’s CEO, president,

and chairman of the board, Jerry Junkins (now deceased), launched a

companywide improvement and transfer program in 1994.

The story of KM thinking at Texas Instruments is the of story of

post-reengineering management thinking in corporate America. Like

--- many of its peers, the Dallas-based

What do you do when you’re done electronics company (1996 sales:

cutting? You start managing. $13.1 billion) reengineered, reorga-

____ nized, and restructured during the

late 1980s and early 1990s. With over 50,000 employees in sixteen

countries worldwide, it has empowered its workers and delayered its

corporate body.

“We had cut and cut and cut,” says Cindy Johnson, director of Col¬

laboration and Knowledge Sharing and a fifteen-year veteran of the

company. “We had achieved high levels of efficiency across our opera¬

tions by streamlining our processes and fine tuning our people and re¬

sources.” Then, in 1993, with the last drop of fat squeezed out of its

organizational cost structure, TI began to look elsewhere for perfor¬

mance improvement. The company’s focus shifted from cost cutting to

opportunity growth.

With that shift, “we felt we needed to find a new paradigm for reach¬

ing the next level of improvement,” says Johnson. “We had to find

ways to become more agile and learn faster so that we can innovate

152

TI’s Best Practice Sharing Engine • 153

faster than our competitors,” she explains. “This is not about cutting

people or bringing in new machines. It is about facilitating the flow of

ideas, practices, and knowledge.”

KNOWING WHAT WE KNOW

In 1993 Junkins challenged the Quality Leadership Team (QLT, a

team of senior executives from each TI business) to create a common

and global approach to business excellence. “We had pockets of medi-

ocrity right next door to world-class performance simply because one

operation did not know what was happening at the other operation,”

explains Johnson. The QLT’s goal was to find a way to level the inter¬

nal playing field and leverage best performance across the enterprise.

Out of this need, TI-BEST, or the TI Business Excellence Strategy,

was born. TI-BEST deploys a standardized process of defining business

excellence, assessing progress, identifying improvement opportunities,

and establishing and deploying an action. “TI-BEST provides a com¬

mon methodology and a common language that can be used to provide

best practices across the company and to more rapidly accelerate best

practice,” explained Junkins at a 1994 leadership conference. “We

cannot tolerate having world-class performance right next to mediocre

performance simply because we don’t have a method to implement

best practices.”

Junkins laid the groundwork just before his tragic death in 1996 on

a business trip to Europe. But the effort continued full steam under the

guidance of incoming CEO Tom En- -

gibous, former head of the company’s After one year of sharing best

semiconductor group. Engibous em- practices, TI generated $500 million

braced TI-BEST, and in 1994 he chal- in “free” fab capacity. Need we say

lenged his managers of TI’s thirteen any more?

wafer fabrication plants to come up -

with a way to close down the wide gaps in their plants’ yields and pro¬

ductivity levels.

At the time, the semiconductor market was tight and new capacity

came at a premium. A new wafer fab would cost between $500 million

and $1 billion to construct. If TI could eke out the capacity equal to

that of a wafer fab by bringing laggard plants up to snuff, it would save

154 • If Only We Knew What We Know

significantly as well as be able to provide much-needed product to its

customers at a time of need.

The managers, along with a team of facilitators (trained in process

redesign theory and cycle time reduction techniques) compared best

practices across wafer fab plants globally. Combined with best practices

gained from external benchmarking, they then created a “virtual fab

(read: a best practice). They created an incentive to work together by

tying their bonuses to total capacity and productivity measures. These

improvements generated one “free” fab of capacity that same year

and then, they did it again in 1994 and 1995.

TRY, TRY AGAIN

Impressed with the success of the wafer-fab effort, top management

throughout TI began realizing the potential cultural and financial ben¬

efits of internal best practice sharing. The insight sparked a flurry of

enterprisewide activities.

By October of 1994, TI launched the Best Practice Sharing Initia¬

tive to help shift the internal culture from “I have to solve my own

problems” to one of looking for existing solutions—first. “Only invent

the wheel once!” was the rallying cry.

In October 1994, the QLT chartered a Best Practice Sharing Team

and the TI Office of Best Practices (OBP) to help facilitate, guide, and

manage the sharing of internal knowledge among the company’s vast

and widespread operations. The team’s mandate was to design the shar¬

ing process and IT infrastructure. The OBP serves as the “engine” for de¬

veloping and deploying training and technology; it is supported by a

team of thirteen champions and a network of 138 global facilitators.

“Facilitators devote about 10 to 50 percent of their time to sharing,”

Johnson explains. “Their role is to identify existing strengths within

TI, document those practices that provide improvement, and share

them throughout the organization. They are matchmakers,” says John¬

son. “They are trying to broker their businesses’ strengths to help oth¬

ers and in return, learn about practices in other businesses.”

SHIFTING GEARS

The role of a facilitator or “knowledge broker” was created in TI based

on the “free” wafer fab Findings as well as Gabriel Szulanski’s research

TVs Best Practice Sharing Engine • 155

conducted at the French business school, INSEAD, and sponsored by

APQC. A third party helps smooth the transfer of knowledge from

giver to recipient.

Indeed, TI’s greatest challenge has been motivating people to share

and accept other people’s ideas. In an organization where engineers

take pride in innovation, relying on someone else’s work has not ah

ways been respected practice. It was -

up to OBP to change that perception. Invested in reengineering? Do not

“What we see over and over again is despair! Lessons learnt once need not

that you have to provide the focus be learnt again. The result: a faster

and motivation, and you have to pro- step up the ladder.

vide the tools, and then you have to -

reward and create an incentive structure,” says Johnson of the cultural

transformation.

In some respects, TI was fortunate. Years of reengineering and cross-

functional teamwork provided a strong basis on which to build the

practice of sharing. “We understand processes, we talk to each other,

and we use the same language,” says Johnson. These are all critical

components of a knowledge-enabling culture. However, reengineering

had left its scars as well. TI—like many other organizations—has lost

an entire layer of middle management, and with it a critical link in the

communication chain. Fortunately, new network technologies have

stepped in to recreate the links.

“Technology has given us new ways to keep in touch with each

other and share information. We have to take advantage of that tech¬

nology to get better faster,” says Johnson. Before intranets and Lotus

Notes, knowledge bases were slow, tedious, and typically not easily

searched. “We could create enormous repositories, but never get any¬

thing out of them,” she muses. Search engines, digital signals, and data

compression have made high-speed communication cheap and easy.

The first release of the TPs internal card catalogue, in April 1995, took

only a quarter to design and roll out. “Previously, such effort would

have taken at least two years.”

THE BEST PRACTICE “CARD CATALOG”

The technology heart of TPs knowledge management efforts is a Best

Practices KnowledgeBase—a Lotus Notes application accessed through

156 • If Only We Knew What We Know

CREATING A KNOWLEDGE-ENABLING IT ARCHITECTURE

Based on her experience in helping develop and implement the knowl¬

edge-enabling technology at TI, Johnson offers the following five tips:

1. Find a renaissance person who thinks in business processes and is

also technology savvy. Work through a basic KM model: First cap¬

ture, then abstract, then disseminate.

2. Assess what tools provide the best capabilities in each area of the

process, using cost, maturity, and features as your assessment cri¬

teria. Recognize that you will need to stabilize new technologies

and deploy them worldwide.

3. Challenge your team “not to design the IT solution” but to focus

on the process, roles, and culture change requirements. After each

work session, go back and develop a prototype that illustrates the

process defined in the session. Force team members to focus on the

future state and not the technology.

4. Build an implementation plan around your proposed solution—

don’t take on training for the technology as a whole. TI’s initial im¬

plementation was a Notes-based solution, and training only

focused on the best practices sharing application and provided just

enough training to get everyone up and running.

5. Stand back and watch the darts fly over your choice of technology.

Be prepared to answer any and all questions about that choice.

Don’t oversell your technology or let it become a religious battle.

Stay ahead of your critics.

TI’s global intranet. The choice of Notes reflects TI’s belief that the

groupware provides the best categorization capability. In March 1995,

two hundred TI Notes users began using the KnowledgeBase. The fa¬

cilitators, the Best Practice Team, and the QLT loaded the original

supply of best practices; to date, the database contains more than five

hundred practices.

Each practice is documented using a title, a short narrative, and

contact information. Facilitators assign a quality criterion from a list of

eight derived from the Malcolm Baldrige National Quality Awards,

the European Foundation for Quality Management, and the Singapore

Quality Award Criteria. Practices are then categorized by process, from

TVs Best Practice Sharing Engine • 157

a list of sixteen high-level processes (based on a classification frame¬

work developed by the APQC and Arthur Andersen). Each practice is

also assigned keywords from a list of -

130 predefined keywords that apply to TI packs a “mean” one-two punch:

the process. (Users can perform key- One for Lotus Notes. Two for the

word or a full-text search to locate rel- intranet.

evant practices.) -

The Best Practice KnowledgeBase also contains external best prac¬

tices and benchmarking information. A Notes-based project notebook

helps teams capture learning and knowledge created during the course

of a project, and a Notes-based discussion database enables employees

to communicate business gaps and solutions.

While the Notes KnowledgeBase allows limited access, an intranet

site called “Sharelt” was launched in July of 1995 using the InterNotes

Publisher. The site allows TPs employees worldwide access to the best

practices listed in its database, as well as many of the Notes databases.

Using Domino to bridge the Notes/Web gap, users can access the

Notes material via any WWW browser. “Shareit” now gets more than

10,000 hits per month.

TI has launched another Internet application, a “yellow pages” web

site, Connect, which lists TI employees’ contact information, experi¬

ence, knowledge, and interests. It also contains want ads, which will

allow employees to promote new business opportunities and improve¬

ment projects for their peers to review, as well as to solicit ideas and re¬

sources for solving problems.

WE’VE COME A LONG WAY

After three full years of operation, the Office of Best Practices has mul¬

tiple successes under its collective belt: It has provided an integrated

mechanism for best practices sharing, built the initial supply of cata¬

logued practices to more than 500, and increased awareness and de¬

mand for the KnowledgeBase across the company.

In the process, it has learned four key lessons:

• Do what you are able to do.

• Don’t try to do it all at one time.

158 • If Only We Knew What We Know

. Have the infrastructure already in place and connecting the knowl¬

edge management activities to your business strategy.

• Put more emphasis on like-skilled communities (i.e., software

groups and purchasing expertise). Implementation is more likely to

take place where there is commonality.

---TI would like to pursue knowl-

The three pillars of KM excellence: edge management on a wider scale.

1. Focus 2. Focus 3. Focus In the early stages of its KM initia-

___ tive, TI focused on achieving opera¬

tional excellence by sharing best practices across businesses. However,

Johnson recognizes a shift is already underway toward collaboration for

SPREADING THE WORD: “SHAREFAIR”

TI’s Best Practice Sharing Program recognizes that providing knowledge¬

enabling technology does not guarantee people will share or use preexisting

knowledge. In many high-tech companies—and TI is no exception—the

inherent culture is one of designing new solutions, not reusing old ones.

Indeed, the latter may be considered antithetical.

In June 1996, TI held its first companywide ShareFair. The one-day

event drew over 500 people, who attended seminars on how to share

knowledge and visited an exhibit hall of more than fifty best practices

(the “practice that’s best for me”). Among them were best practices in

business process management, customer satisfaction survey techniques,

strategic planning process, and user-centered product development

process.

At the ShareFair, TI also handed out its first “Not Invented Here,

But I Did It Anyway” (NIHBIDIA) Award to the sources and recipient

of a transfer that exemplified the cultural change at TI through the trans¬

fer of a practice. Nominations for the award were evaluated for business

impact, business improvement priority, span of the transfer, and method

of transfer. The judges’ panel put greatest emphasis on high impact, high

priority, broad implementation, and team approach.

Fifty-two nominations were submitted with over 450 team members

cited for generating savings in excess of $1 billion. In the end, awards

were given to two TI organizations which best illustrated the spirit of

best practice sharing through their transfer of clean-room best practices

TI’s Best Practice Sharing Engine, • 159

innovation in product development and increased customer intimacy

as the focal points of its KM efforts. “The key message is the focus mes¬

sage,” says Johnson. “You can’t take KM and say we are to become a

‘knowledge-creating company.’ You have to find the key leverage

points before you do anything else.”

between a semiconductor wafer fab and the defense business gallium ar¬

senide wafer fab.

The 1997 ShareFair focused on Internet Technologies and the shar¬

ing of best practices in the use of the intranet, extranet, and Internet ca¬

pabilities. The 1997 NIHBIDIA Award generated fifty-one nominations

with over 480 team members cited. The award was given to TI’s Dallas

Wafer Fab Operations, and to a new TI acquisition, Silicon Systems

Incorporated, for their sharing of wafer fab practices.

The TI ShareFairs have helped employees recognize the value of

reusing existing knowledge and have attached prestige and status to the

practice. Future cultural transformation efforts will focus on helping em¬

ployees:

• Rely on the Best Practices KnowledgeBase first for a solution to their

business gaps

• Become aware of others’ problems to see if assistance can be offered

• Collaborate to solve a common business gap for which there is not an

existing solution document, instead of attempting to solve it separately

• Operate at the same level of excellence across the organization

To achieved these goals, TI recognizes it must change the way it re¬

wards and compensates its employees to reflect the great emphasis on

sharing and reuse. Indeed, the best practice sharing team hopes to make

Best Practice Sharing an integral part of individual, organizational, and

business review.

Chapter 16

BECOMING A “KNOWLEDGE BANK”

In its global battle against poverty and economic malaise, the World

Bank can afford to waste no time. The Bank has designed a compre¬

hensive KM initiative to improve the process of best practice sharing

among its regional groups. By formalizing hitherto informal communi¬

ties of practice, linking regional groups, and leveraging the unifying

powers of technology, the Bank wants to become a superior knowledge

performer in-house first. Ultimately, it hopes to be a clearinghouse for

economic development knowledge—“what works and what doesn’t

and the data to prove it”—throughout the world.

In the fall of 1996, James Wolfensohn, president of the World Bank,

pronounced knowledge management a strategic thrust for the twenty-

first century. Speaking before 175

Make note of this: Using a clear finance ministers, Wolfensohn out-

blueprint first prevents chaos later! Uned his vision for the World Bank

of the future. “The Bank Group’s re¬

lationships with governments and institutions all over the world and

our unique reservoir of development experience across sectors and

countries position us to play a leading role in [a] new knowledge part¬

nership,” said Wolfensohn. “To capture this potential we need to in¬

vest in the necessary systems that will enhance our ability to gather

development information and experience and share it with our clients.

We need to become, in effect, the ‘Knowledge Bank.’”

Within months of Wolfensohn’s speech, an institutional task force

translated his grand vision into a comprehensive action plan and de¬

tailed a road map for organizational restructuring. (Of course, they

160

Becoming a “Knowledge Bank” • 161

benchmarked best practices with APQC first!) The Bank’s new KM

system is in early stages of implementation, but management antici¬

pates that it will dramatically change the way in which it operates in¬

ternally and interacts externally. Based on early returns from a few

grass-roots guerrilla pilots, they are probably right.

Initially, the supranational is focusing on internal knowledge; it

wants to provide all of its employees with “just-in-time, just-enough”

knowledge to perform their work better.

What does that mean? It means advisory services would get answers

to customers, faster, by accessing existing knowledge and best practices

from around the globe. The institutionwide effort will also make it eas¬

ier for the Bank to consolidate its comparative advantage in providing

international best practice. Plus, the more efficient use of existing re¬

sources will increase its capacity for serving multiple stakeholders, be

they borrowers, donors, or foundations.

Ultimately, the bank hopes to become a clearinghouse of knowl¬

edge on economic development by opening up its collective IQ to ex¬

ternal browsers and interconnecting with universities, foundations,

and others.

This list of mission and business objectives would not surprise any¬

one familiar with the Bank. Though its mission tends toward the altru¬

istic, the Bank is managed for commercial viability. Headed by a

practical ex-Wall Street executive, this financial powerhouse gener¬

ates approximately $1 billion in annual profits and maintains a pristine

credit rating, while lending between $15 and $20 billion a year to

some eighty developing nations. The decision to embrace KM, there¬

fore, is rooted in sound business principles—turning internal know¬

how into commercial success, achieving operational excellence, and

forming more intimate ties with external customers.

“After fifty years of providing lending and advisory services in every

region of the world, the Bank Group is a substantial repository of de¬

velopment experience and best practice,” explains Stephen Denning,

program director, Knowledge Management. However, he points out,

“this knowledge is not always easily available to those who need it

when they need it, or in formats they find useful and accessible. As a

result, the effectiveness of our service suffers.”

162 • If Only We Knew What We Know

THINK BIG, START SMALL

Like other organizations, the Bank was not altogether a KM novice

when it embarked on this recent campaign. Over the years, the organi-

zation has taken steps toward more effective management of knowL

edge, including the highly successful African Region LDB, a smart,

- live database of economic indica-

This is one case where the tors and country-specific informa-

“more” is not the “better.” tion. All of the pre-1996 efforts,

-- however, were characterized by

their local focus and attention to the elemental aspects of knowledge

(i.e., data and information vs. action-oriented, best practice knowl¬

edge). Further, they did not provide the basis for a sustainable, inte¬

grated strategy. “The fragmentation associated with continuing or

further proliferation of local knowledge systems would delay by years

the Bank’s efforts to create the Knowledge Partnership,” says Denning.

The organizationwide initiative began in earnest after Bank execu¬

tives participated in an APQC’s KM 1995-96 consortium benchmark¬

ing study (APQC, 1996). Armed with Wolfensohn’s vision and newly

gathered best practices in KM, the Bank appointed Denning to the

newly created position of program director, Knowledge Management,

and devised a grand plan, involving new organizational facets, such as

sector networks and help desks (see page 166) and a phased implemen¬

tation plan that takes incremental steps toward the knowledge bank

ideal.

LAUNCHING PILOTS

By September of 1996, the Bank was ready to roll out its first sector

network: EKMS, the Education Knowledge Management System.

With a core staff of twenty and located within the Human Develop¬

ment Group, EKMS serves 250-300 education sector staff worldwide.

Its purpose is to facilitate knowledge synthesis, stimulate discussion,

and identify areas that need attention. For example, its staff organized

a popular e-mail discussion on the use of local languages in textbooks.

It then distilled key learnings which were added to the organization’s

knowledge bank.

Becoming a ‘‘Knowledge Bank" • 163

The EKMS staff identifies existing best practices, especially through

its advisory groups, and provides training for education staff. Groups of

two to four people work on each of nine strategic focus areas, such as

economics of education, educational technology, and effective schools

and teachers. In addition, groups of six to twelve of the Bank’s educa-

tion professionals located throughout the world act as advisers on each

of the focus areas as part of their full-time duties.

The EKMS network is prototypical to the Bank’s emerging sector

networks. At its nexus is the intersec- -

tion of technology and people. An in- Let us repeat this one: “Network

temal Web site provides staff with building is relationship building."

access to documented knowledge -

such as best practice examples, tools, ideas, links to other Internet

sites, key readings, and selected documents. An Advisory Service—a

help desk staffed by two information specialists—complements the

abundance of formal, organized knowledge by providing answers to

specific questions. Information specialists take responsibility for mak¬

ing sure internal “clients” get the information they need, whenever

possible—even performing specialized research.

“Many people initially said, ‘I don’t know how the Sector Network

can help me,”’ Martha Pattillo-Siv, coordinator for the Education Ad¬

visory Service, explains. “But once they saw they had someplace to

turn to get their specific questions answered, the network took on a

different texture for them.” Says Pattillo-Siv: “Network building is re¬

lationship building, and you really cannot build a relationship through

a Web site alone.”

As an added benefit, Advisory Service staff deliver detailed records

of questions and answers to enrich the organization’s collective IQ.

Previously, no such record existed, and each question ignited a “rein¬

vention of the wheel” process that significantly slowed down response

time to external customers. “By tracking this information, we ensure

that the momentum does not stop after every question,” Pattillo-Siv

says. “We see where the trends and gaps are. Then we can build knowl¬

edge development projects based on those gaps.”

About 15 percent of the World Bank’s education staff routinely uses

the Advisory Service each week. Though a relatively small percentage,

it sits well with the Bank’s “think big, but start small” approach. “We

164 • If Only We Knew What We Know

K

do not want to be like the restaurant that gets great reviews from the

New York Times and is mobbed with enthusiastic customers before it is

really ready to handle them,” Pattillo-Siv jokes.

Furthermore, the impact even on this smaller scale is already no-

ticeable. Pattillo-Siv recalls a recent case of a staff member in Nepal

who called the Advisory Service looking for implementation plan

models to help his clients in the Nepalese Ministry of Education pre¬

pare the next phase of a primary education project. “Through its net¬

works, the Advisory Service was able to find a generic implementation

plan as well as multiple contacts in countries that had done similar

projects,” recalls Pattillo-Siv. The Advisory Service located people in

Hungary and Turkey with best practice experiences in rolling out sim¬

ilar education projects. “The EKMS was able to rapidly pull together

resources from around World Bank, where in the past there would

have been no way to facilitate this kind of sharing.”

FIGURING OUT THE RESISTANCE POINTS

The Bank’s KM initiative is in the early stages of implementation. Its KM

staff, however, has been able to identify seven organizational resistance

points that could potentially impede their progress:

• Shifting the culture towards sharing

• Setting and implementing quality standards

• Maintaining the system to avoid knowledge “junkyards”

• Ensuring that the system stays demand-driven

• Balancing new information versus better access to current informa¬

tion

• Resolving external issues, such as confidentiality

• Achieving an integrated approach across the organization

BUILDING THE LAUNCH PAD

While the sector networks gear up, the Bank’s IT function is busy con¬

structing a supportive technology infrastructure. “Knowledge manage¬

ment is 80 percent human brain power and 20 percent facilitation by

information technology,” says Denning. “The challenge is to harness

the technology to link people together and to leverage its impact for

Becoming a “Knowledge Bank” • 165

development.” In practical terms, the Bank’s challenge is twofold:

pulling together disparate knowledge bases so that knowledge can flow

seamlessly throughout the organization and harnessing the right tech¬

nology for leveraging particular aspects of knowledge.

The World Bank’s KM technology infrastructure is not a single

monolithic delivery system. Rather, it is a suite of tools, mostly run¬

ning on Lotus Notes and Web-site platforms, which pulls together a

collection of integrated knowledge resources. The Bank tries to match

technology with knowledge “typologies” and is working to accommo¬

date multiple access alternatives, from faxes to telephones. In some

parts of the world, modem access is nearly impossible.

To accommodate both divergent and -

convergent knowledge processes, the For best results: take one spoon of

Bank’s IT architecture will include low-tech, and one spoon of high-tech,

(in descending order of “knowledge” mix and drink,

intensity): -

• Databases, having vast warehousing capacity and relational logic,

will offer access to codified knowledge, including terms of reference,

consultants, who knows what, lessons learned, key articles and

books, and reports.

• Knowledge bases, accessible via Web sites or Lotus Notes, will con¬

tain best practice reservoirs, including sector strategies, tool kits,

model outputs, analytical tools, sources of subject-driven informa¬

tion, and think-pieces.

• Help desks, low-tech and manned by experienced “humans,” will

provide the intelligent capacity to answer complex queries, and to

customize resource maps and information packets based on special¬

ized needs.

AN ARCHIPELAGO OF TECHNOLOGY ISLANDS

When it came time to create the technology skeleton that would sup¬

port the growing body of knowledge, the Bank ran into a common cor¬

porate problem: Years of as-needed technological and content

development have resulted in an IT landscape resembling a veritable

archipelago of disconnected islands. Each sector (department) has had

166 • If Only We Knew What We Know

its own somewhat different way of viewing, processing, and storing

knowledge.

For the Bank’s Information Technology Services (ITS) the chal¬

lenge has been to develop an architecture that would accommodate di¬

verse knowledge models but have enough commonality to allow best

practices to be easily transferred. Initially, it appeared the knowledge

A BLUEPRINT FOR THE FUTURE

Transformation without guidance often results in chaos. And the World

Bank was indeed careful to draw a detailed blueprint of the “knowledge

bank” of the future before it began tearing up departmental walls and de¬

signing enterprisewide systems. This sketch depicted a three-tier, knowl¬

edge-enabling infrastructure.

Tier 1: The Sector Networks. The key operating units of the KM plan,

the Sector Networks are formalized versions of the bank’s informal, in¬

ternational communities of practice. Whereas many sectors such as in¬

frastructure or environment—have had informal communities of

practice for years, they had been mostly focused on country or regional

expertise. The Sector Networks, in contrast, will expand and formalize

these relationships beyond geographic boundaries.

Sector Networks will take the lead in developing the requisite knowledge

bases, i.e., organizing operational staff so that the flow of global knowl¬

edge is facilitated. Each knowledge area will be led by a full-time knowl¬

edge manager and supported by subject specialists and other operational

staff, who will spend part of their time building and maintaining the

knowledge base—constituting communities of practice in their fields of

expertise—assisted by help desk personnel.

Ter 2: Central Knowledge Management Team. While the networks must

continue to do the bank’s day-to-day work, a Central Knowledge Man¬

agement Team with dedicated full-time staff will be charged with over¬

seeing the ongoing health of the knowledge management efforts. The

team’s responsibilities include: <*i

• Establishing and integrating knowledge management framework

• Setting priorities among multiple knowledge management activities

• Managing external access to the KM system

• Regulating the knowledge economy

• Managing issues related to the quality of knowledge

• Consolidating external knowledge partnerships

Becoming a “Knowledge Bank” • 167

domains were simply too differentiated and any compromise would, in

all likelihood, stifle creativity within each domain. At the same time,

merely overlaying this archipelago with a common Web-based browser

interface would provide some measure of pseudo-integration but would

deprive users of more systematic and powerful access of a truly inte¬

grated system.

• Facilitating the organizational cultural shift

Tier 3: IT Services. Finally, the Bank’s ITS group will support the

knowledge management system with the appropriate technological infra¬

structure, ensuring the knowledge management system is fully integrated

with other information systems in the organization.

Remaining Regional Role. While the networks and the central teams are

either new or newly formalized facets designed to cut across the global or¬

ganization, the existing regional structure will continue to play an impor¬

tant role, including:

• Network management: As members of the Network Councils and

Sector Boards, managers from the regions will play a crucial role in

championing the knowledge management initiative and in fostering

the culture shift.

• Network staff: The regional staff, who constitute the bulk of the net¬

work, will be the main actors in the knowledge management activities.

• Budget: Roughly one-third of the incremental budget resources for

networks will be devoted to knowledge management activities.

• Live data bases: The regions will play a key role in developing and

maintaining the live databases.

• Region- and country-specific knowledge: The regions will play a

unique role in capturing and managing region- and country-specific

knowledge. Details have still to be worked out.

• Mission briefing and debriefing: The regions should play a key role in

knowledge capture through mission briefing and debriefing. Details

have still to be worked out.

• Help desks: The regions will play a role (still to be defined) in orga¬

nizing help desk functions.

• External knowledge partnerships: Most of the regions are undertaking

or planning knowledge partnerships with organizations in their re¬

spective regions.

168 • If Only We Knew What We Know

A closer look revealed that while domains stored, processed, and

viewed data differently, they all used more or less the same basic build¬

ing blocks to classify their explicit knowledge. The solution was to add

in a third layer of user interface into the bank s IT architecture. This

interface is designed to fetch data from divergent databases, and pre¬

sent it within a context that is familiar to the particular user.

WEIGHING RISKS AND REWARDS

The Bank’s staff recognizes KM is in its infancy and presents tall chal-

---lenges—in particular in terms of

No one said it would be easy.... But no creating a new culture of sharing

KM is going to make it lOx harder. (see page 164). However, these chal-

___ lenges are dwarfed by the risks in¬

herent in failing to manage knowledge, and the benefits of managing it

more effectively, which include:

• Higher levels of knowledge reuse

• Reduction or elimination of rework

• Increased quality and quantity of client services through enhanced

access to Bank knowledge

• Strengthened comparative advantage in providing international

best practice

• Stronger capacity building

• A prerequisite for physical decentralization

• Better incentives for excellence

While some benefits may not show up for some time, a 1996 survey

of Bank employees revealed even early KM efforts to be yielding posi¬

tive results. Close to 90 percent of employees have found KM-related

products and services useful or very useful; some 84 percent rated the

turnaround as fast or very fast; and 70 percent find that it makes their

work significantly more effective.

The Bank’s initial focus has been on retooling its internal opera¬

tions to improve sharing of best practices. The pilot EKMS network

has already been joined by three other sector network pilot systems—

for nutrition, health, and population. By the end of fiscal year 1998,

the Bank planned to have seven or eight sector families taking their

Becoming a “Knowledge Bank” • 169

first systematic steps toward KM. The rest will join by the end of the

following fiscal year.

Longer-term, the Bank’s vision is fixed on building knowledge part¬

nerships with external stakeholders. “If we can do that for our in-house

staff, why not make it directly available to our clients and partners as

well?” asks Denning. “If we are working with joint information, we are

solving a joint problem. The whole concept of having genuine part¬

nership then becomes a real possibility.”

Chapter 17

SEQUENT COMPUTER’S KNOWLEDGE “SLINGSHOT”

ware that loyal customers value its knowledgeable sales force as

Ymuch as its top-of-the-line UNIX systems, Sequent Computers

began managing knowledge like an asset in 1993. The company s first

step was building an enterprisewide technology infrastructure. Next it

created a supportive organizational structure. Owning an earnings

growth rate of 20 percent per year, Sequent clearly has no time to

spare on reinventing the wheel.

For centuries, the smaller have relied on wits and agility to defeat

larger opponents. Add a modern-day slingshot in the form of top-

notch technology architecture, and you’ve got Sequent Computer

---— Systems: This $600 million David

• Think Big • Act Fast • Be Smart competes with Goliath-size HP,

IBM, DEC, and Sun Microsystems

in the high end of the UNIX systems market. “With the kind of com¬

petition we have,” says Roger Swanson, manager, Corporate Research

Services at Sequent, “We have to act and look bigger than we are, and

the only way to do that is by sharing information.”

Based in Beaverton Oregon, Sequent employs 2,700 in fifty-three

field locations in the United States, Europe, and Asia. Its gained a

stellar reputation for being the “freightliner” of the UNIX world—per¬

fect for those who require real “heavy-duty” machinery. But Sequent is

not selling “big iron.” It provides solutions, and often multimillion-

dollar solutions. As such, the expertise of employees has become as

critical to its success as the processing prowess of its computers.

“We recognized—or rather, our customers told us—that we were

valued for the ability to de-risk complex multi-vendor IT projects be-

170

Sequent Computer’s Knowledge “Slingshot” • 171

cause of our [knowledge-intensive] design and implementation exper¬

tise,” says Marc Demarest, ex-chief knowledge officer. “We [the COO

and CKO] became convinced very quickly that (1) we need to hus¬

band and nurture it [our knowledge] better, (2) we needed to amplify

its effectiveness by making the tacit explicit and embodied, and by dis¬

seminating the embodiments, and (3) we needed to treat it as the asset

it was from a corporate management perspective.”

LET THERE BE KNOWLEDGE

Sequent began by carefully analyzing its business model. The goal was

to identify and target knowledge-sensitive points where improvement

would yield the greatest results. Was it reduced cycle time? Lower

costs? Higher volume of customers, projects, and therefore revenue?

Risk mitigation? Or what?

The analysis revealed Sequent would do best by focusing on its di¬

rect sales channel which includes professional services communities

and sales teams. “Since these were

close to revenue, the customer, and Customer intimacy means pushing

our cultural heart, they made a good intelligence to the front lines,

starting point,” says Demarest.

An added bonus: They are also the closest to Sequent’s competi¬

tors—and hence a logical place to make a differentiating impact. “The

best and smartest people must be on the front lines dealing with the

customer,” explains Swanson. Indeed, Sequent had realized that the

productivity of its sales force was directly related to the number of

years they’ve worked at the company. Experienced sales teams drew on

their own knowledge reservoir to make future sales better and faster.

The goal, therefore, was to make this knowledge available to everyone

so that each front-line employee would be able to respond to cus¬

tomers with the collective intelligence of the organization.

PROVIDING THE UNDERPINNINGS

So, where to start? Based on a straightforward knowledge model which

tracks knowledge through construction, embodiment, dissemination, and

use, Sequent decided to focus on managing the two middle phases first.

172 • If Only We Knew What We Know

“Embodiment and dissemination are the two most tractable,” explains

Swanson. They are also the ones most clearly in the organizational

versus individual domain.

To implement its design, Sequent has followed a four-phased man¬

agement strategy:

1. Underpin the efforts by providing the IT infrastructure a com¬

mon space for storing and retrieving document-based knowledge

and some forms of electronic discussions

2. Observe the knowledge economy to understand what people use,

what they contribute, and the feedback they provide to help fine-

tune the system

3. Measure activities within the knowledge economy, such as pat¬

terns of use

4- Orchestrate by engineering more sophisticated social expecta¬

tions about use, adding new content, new producer communities,

and new tools

“We did one thing first: We built out the technology infrastructure

we would use for all our KM initiatives,” says Demarest, “because KM

infrastructure is a must if you intend to instrument your knowledge

economies. If you don’t build it first, it will be built in pieces by dozens

of people and projects and you will have a massive second-order inte¬

gration problem trying to hook disparate systems together, so massive

that its costs and inefficiencies might undo whatever improvements

you get from KM programs.”

At the core of Sequent’s knowledge-enabling IT design is SCEL.

Pronounced “sell,” this is the Sequent Corporate Electronic Library,

___— focusing on marketing and sales

Seeking balance? Sequent chose support and generally helping all

centralized IT to offset a free-market employees do their jobs better. The

KM philosophy. SCEL vision is: “The one place to

__- go to find anything you need to help

you do your job more effectively.” SCEL is an intranet site that con¬

tains corporate and individual knowledge domains (see page 173). The

site is integrated with other Sequent applications, such as an on-line

pricebook, the corporate “white pages,” and a customer database. “It’s

a single holistic platform,” says Swanson. E-mail discussions on shared

Sequent Computer’s Knowledge “Slingshot” • 173

“aliases” (discussion forums) are treated as “documents” and archived

in SCEL.

Whereas SCEL is critical to the dissemination of knowledge, IT and

KM are separate functions, one reporting to the CIO and the other to

the manager of corporate research services. The IT organization provides

the technology, and human and financial resources to support KM pro¬

grams (current costs per employee, per year, are under $1,000). The

manager of corporate research services has global responsibility for all

knowledge management programs and infrastructure. A cross-func¬

tional SCEL design team consists of librarians, a Web master, program¬

mers, a SCEL architect, a SCEL evangelist, and other members who

provide linkage to other parts of the organization. This SCEL design

team provides infrastructure, education, operational support, and KM

program management and coordination, and acts as the keepers and

guidance support system for the knowledge management effort. The

KM function is also responsible for the company’s patent portfolio and

for such traditional corporate research services as the corporate library.

WHAT’S ON SCEL?

We think of SCEL as a city: a marketplace in which many vendors

set up stores of various sorts and vend their knowledge to the

consumer community.

—Marc Demarest, former CKO

The Sequent Computer Electronic Library is an internal, Web-based

system for capturing, archiving, and providing information to Sequent

employees. SCEL includes a combination of database management

systems, full-text retrieval engines, file system storage, and complex

clusters of programs, all of which are integrated into Sequent’s world¬

wide internal Web and accessible to all employees through Web

browsers.

The choice of an intranet application was a natural one, says De-

marest. “Unlike Lotus Notes or other proprietary software, the Web is

owned by no one,” he says. As a result, Sequent is not beholden to any

vendor for the migration path of its knowledge-enabling tools. The

Web can be used for everything from real-time collaboration to elec-

174 • If Only We Knew What We Know

tronic commerce. “It’s the infrastructure of choice of computing inside

and outside the firm for the next decade,” he says.

SCEL works on a publisher/consumer relationship. Every employee

is a publisher and/or consumer if they use SCEL. Publishers put knowb

edge into the system and consumers use that knowledge. “We take, de¬

liberately, a laissez-faire capitalist approach to the ‘what content’

question: user demand and producer marketing savvy determine con¬

tent availability,” says Demarest. “We do not, and do not intend to,

regulate content centrally. We are responsible for providing a safe,

clean, well-planned city, not for acting as a knowledge police force.”

Indeed, the SCEL team relies on market mechanisms to control

quality: consumers complain when the content is bad. Every primary

section or page in SCEL contains a feedback button to allow con¬

sumers to immediately react to good or poor documents as well as sys¬

tem features. User feedback goes directly to the publisher. A copy also

goes to the SCEL team, who monitors it for action. When the market

mechanisms do not work, and the knowledge is considered high-value

by the librarians, the design team or the KM team will intervene.

When Sequent began to build SCEL in 1995, it brought together

groups of “influencers” from across the company to understand the way

people thought information should be organized. The design team un¬

covered two views. Business and marketing groups clustered informa¬

tion around the value chain and groupings such as “customer,”

“products,” “competitors,” and “suppliers.” Other groups preferred a

more functional view of the organization. SCEL was subsequently de¬

signed to accommodate both views, by relying on HTML documents

and hyperlinks. After they opt for a value-chain or functional

overview, users can browse or search, using a standard, key-word-

driven Web search engine.

The SCEL publishers began populating the content areas with

“magnet” content such as outstanding presentations, strategy and

scripts for sales calls, and design documents which people wanted but

could not get elsewhere except by going directly to the author. This

approach has positioned SCEL, from day one, as the place to find high

value knowledge. And, although in early days the SCEL team would

send e-mails suggesting users consult SCEL for particular knowledge

Sequent Computer’s Knowledge “Slingshot” • 175

needs, very quickly it has become corporate norm to expect to find im-

portant information in SCEL. Now, when users cannot find what they

need, they are quick to complain, thus proving the productive forces

of demand-side economies.

SCEL provides the dissemination infrastmcture for both semistruc-

tured and structured content. Structured content includes alpha-numeric

data which can be stored in conventional relational databases. Semi-

structured content, such as pictures, presentations, spreadsheets, and

mail messages is managed by storing the structuring containers and

metadata about the containers.

Metadata is captured on a per-document basis by the publication in¬

frastructure and is stored in a centralized database “behind” the SCEL

server itself. Obviously, standard metadata like author, title, subject,

keywords, document type, and so forth are captured. There is an effort

in progress on the Web, led by library scientists, to come up with a uni¬

form metadata classification scheme, and, when and if that effort pro¬

duces something, Sequent will consider adopting it.

SCEL also provides ways for disseminating or embodying un¬

structured knowledge. In the case of full-motion video, that means

metadata-based descriptions, until the software technology capable of

“querying” such containers is robust.

It also means creating mechanisms such as hypermail, which help

push unstructured content into the “searcheable realm.” Hypermail is

a simple archiving technology that captures conversation threads in

electronic chat forums based on their “subject” matter line. When a

conversation is started with an “alias” (the Sequent term for an e-mail

distribution list), the subject matter of the first e-mail becomes the

topic, and every subsequent message is threaded on to the main mes¬

sage, thus forming a small question-answer group of messages.

Each set of threaded messages is stored under its alias for reference.

When messages are read later for the purpose of information gathering,

the consumer may contribute to the group of messages by using the

same subject topic so that his response is threaded in that set of mes¬

sages. Messages are not edited for accuracy. If there is incorrect infor¬

mation in any of the messages, it is corrected by the group itself or by

others in the company using the feedback button.

176 • If Only We Knew What We Know

Release 3.0

Sequent has recently installed SCEL 3.0. The fifth iteration of the

original SCEL 1.0 roll-out in the first quarter of 1995, the new release

has much higher publishing capabilities. In addition, Sequent plans to

bring down SCEL’s external walls by creating other areas like SPEL

(the Partner Library) which serve constituencies outside the corpora¬

tion and allow them to have limited views of the company s internal

knowledge bases. This system is more porous and allows more publish¬

ing import and export than before but still maintains security and fire¬

walls.

A “LAISSEZ-FAIRE” KM PHILOSOPHY

In contrast with the planned and centralized manner with which Se¬

quent developed its organizational and IT infrastructure, the com¬

pany’s approach to content and practice is decidedly hands-off. Both

are allowed to evolve without any “government” (read corporate) inter¬

ference.

“It always amazes me that we want capitalism everywhere but inside

our own firms,” notes Demarest. “The whole point of KM is that no

one—most definitely not some wonk with the title CKO in corporate

HQ—is in a position to understand the hundreds of different, specific

contexts in which knowledge is required to satisfy a customer or busi¬

ness partner.”

Hence the knowledge economy at Sequent is regulated by the twin

market forces of supply and demand. An electronic hot link (the feed¬

back button) on each page allows readers to inform publishers whether

their knowledge meets quality standards, and how helpful it’s been.

Sequent uses a push-pull system (so does Texas Instruments). “‘Push-

based,’ centrally planned economies (such as the former Soviet Union)

don’t work well,” notes Demarest. Furthermore, “adult learning theo¬

ries suggest that people learn best when their learning experiences are

contextualized, that is, when they learn ‘just in time.’” Sequent’s “pull”

system is designed to put control of the knowledge bases in the hands

of consumers, and permits optimal learning to take place.

Not surprisingly, Sequent does not offer specific KM-related com¬

pensation. Prestige is the currency of greatest value. Prestige comes

Sequent Computer’s Knowledge “Slingshot” • 177

from being seen as an expert, and expertise is a “reward” for producing

high-value knowledge that is useful to consumers. Not using SCEL to

do your work would be as strange and antisocial a behavior as not re¬

trieving your voice- or e-mail.

OBSERVE FIRST, MEASURE LATER

By its own admission—and reflective of its libertarian bent—Sequent

has spent less time on measuring results than on understanding and

enabling the KM process. The company wants to understand the dy¬

namics of its KM model first, before it -

imposes exact measurements. Plus, The KM measure of the day: CONK

Sequent believes that the interactions or, the Cost of Not Knowing.

are too complex to offer a direct and -

indisputable causal link between KM and an item on the income state¬

ment. Indeed, imposing strident demands to cost-justify may misdirect

efforts and miscommunicate purpose. In fact, the real business case is

the cost of not getting the best information in the hands of sales teams.

Hence the objective of Sequent’s current measuring activities is to

help inform and discipline KM practitioners so that they better relate

their materials to the way the firm operates as a business. It does so by

tracking variations in demand for knowledge and perceptions of cost

and quality impact on its internal value chain. Recent employee sur¬

veys, in fact, reveal a growing perception of the utility and positive im¬

pact of SCEL and KM in general.

“This is not a search for the ‘truth’,” explains Swanson. “We are less

interested in how knowledge is defined than in how its helps us. We

want to focus on ‘what works,’ not on ‘what is right.’”

TODAY AND INTO THE FUTURE

Something is certainly working right. Sequent is enjoying record rev¬

enue growth. According to the company’s KM leaders, SCEL has

helped Sequent raise project average selling price, and reduce delivery

and response time at all stages in the sales and postsales process. It has

also increased the amount of customer-specific and generic knowledge

captured at the interface between the organization and downstream

178 • If Only We Knew What We Know

customers.'SCEL has focused the sales teams more effectively on

proper target markets and has made the assimilation process for new

employees more efficient. Finally, the company has increased the cus¬

tomer-perceived value of its offerings, in hard (financial) and soft (loy¬

alty) ways. Just like any successful consumer products maker, Sequent intends

to hone its product-development process through further insight into __—--its knowledge customers’ consump-

Watch out for that “Big Brother” tion patterns. “We are planning to

syndrome! do more monitoring, says Swan-

__ son. “Not from the big-brother per¬

spective, but to gain knowledge on how to make information more

useful for consumers.”

The company’s future plans include:

• Regularly reviewing the design of the physical and logical structures

of SCEL • Providing better browsing structures that are generated automati¬

cally through the use of metadata • Improving the security model to provide more control over who can

access what • Using SCEL as the front end for a variety of Web-enabled applica¬

tions

Sequent also plans to extend the reach of KM outside its corporate

walls. “We have distributors, suppliers, and manufacturing partners

who could benefit greatly from SCEL,” says Swanson. “Why not allow

them access?” Obviously, there are security and competitive issues in¬

volved. “We’re not sure how to do it yet,” he says. Ultimately, however,

the company believes it will gain more from sharing the most it can,

than by revealing the least it must.

Sequent Computer’s Knowledge “Slingshot” • 179

BUILDING THE “KNOWLEDGE ECONOMY”

Based on Sequent’s experience with SCEL, Swanson offers the following

key learnings:

1. Look for the business linkage. Think how knowledge can influence

the world of its consumers: for instance, sales folks are motivated

by faster close cycles.

2. Business means not just revenue generation, but also improving ef¬

ficiency internally through best practice in operational processes.

3. Technology is important. However, since more and more applica¬

tions are being developed with the Web technology in mind, KM

managers need not be preoccupied with the migration and develop¬

ment of new KM/IT tools.

4- Culture is very important. But don’t wait for the culture to change

to start implementing knowledge networks. “Look for areas of in¬

terest and work with those who are interested,” says Swanson.

5. Start small and don’t worry about the imperfections. In many

cases, “use itself converts the nonbelievers,” he says.

.

Part Five

THE FOUR-PHASE PROCESS

OR “WHAT DO I DO ON MONDAY MORNING?”

We’ve come a long way.

In Part One we laid the foundation for internal transfer and told

you how and why organizations leverage internal knowledge and best prac¬

tices. In Part Two, we delved deeper into the value propositions to achieve

dramatic improvements in customer intimacy, in operational excellence, and

in product-to-market excellence. In Part Three, we described the enablers in

detail. In Part Four, we added to our earlier wealth of examples by present¬

ing four detailed cases of pioneers.

We’ve told you a great deal. But in a way, everything we’ve said is simi¬

lar to telling a baby how it feels to be walking. It means very little until you

take that first step. Part Five is all about how to take the first steps toward

demonstrating the value of transfer. We will offer a simple four-phase process

that could get your initial transfer projects off the ground quickly and suc¬

cessfully. We will give you more examples, especially from the World Bank

and Amoco.

The following four-phase methodology can help organize the process of

making best-practice transfer and knowledge sharing a mainstay of your

company. This methodology (see the figure on p. 182) reflects the lessons

learned from our APQC KM studies, and our own experience working with

organizations to design knowledge management and transfer projects. One

caveat: Every organization is different and the actual steps and issues in the

methodology are never identical. The phases (plan, design, implement, and

scale up) continue to hold true, but excellent diagnostic and change manage¬

ment skills are critical to meaningful design and execution.

182 • If Only We Knew What We Know

1. Plan

The Environment

The Four-Phase Process

After all the “whys” and “whats,” it’s time for the “how-tos" of transfer.

Knowledge without action is useless, and we want you to walk away from

this book with a list actions that can advance and improve your organiza¬

tion’s performance.

Chapter 18

PLAN, ASSESS, AND PREPARE

PHASE 1

“Plans are nothing: Planning is everything. ”

—Dwight D. Eisenhower

The objectives of phase 1 (typically led by a small task force or in¬

ternal consulting group) include:

1. Assess your current opportunities for knowledge sharing

2. Discover your value proposition

3. Find a champion for the initial project(s)

4- Inform and prepare the organization

5. Define the business case

1. ASSESS YOUR CURRENT OPPORTUNITIES

How do you know where you’re going if you don’t know where you

are?

Answer: You don’t!

Hence, the first thing to do on “Monday morning” is to find out

precisely where you are today—so you can figure out where you might

be tomorrow.

There may be some areas of KM and best practice transfer where

183

184 • If Only We Knew What We Know

your company is already doing great. Other areas, however, may re¬

quire serious work. Even companies that have not been formally prac¬

ticing KM have initiatives underway that fit under its umbrella. Some

of these early efforts may well serve as the foundation for an initial

transfer project.

To get a quick size-up, answer these seven questions.

Question #\: Do you know what knowledge you have now? Who has it?

How can you get it, and—most critical!—which parts of it are truly valu¬

able? Few companies realize just how much knowledge they already

have in the heads and files of their employees, or know where to find it

and how to get it. Even fewer, however, understand how different

“types” of knowledge can affect their performance. Recognizing the

treasure chest of knowledge is the first step. Linking specific types of

knowledge to the firm’s value proposition (customer intimacy, prod-

uct-to-market or operational excellence) is the second.

Question #2: Do senior managers understand and support KM as a busi¬

ness strategy? Many leaders don’t understand that “knowledge” is an

asset in which they must invest and which they can create, manage,

and leverage. Others feel KM is but another fad. If knowledge manage¬

ment is not hooked directly into the way business is managed and work gets

done, it’s not likely to go anywhere.

Question #3: Are you systematically transferring knowledge inside your or¬

ganization? Even if an organization realizes where it has that knowl¬

edge, it still does not mean valuable knowledge is systematically

transferred from one part of the company to another—across depart¬

ments, divisions, regions, the globe. Sometimes it isn’t even trans¬

ferred across the room!

Question #4: Are you systematically acquiring outside knowledge? How?

From whom? And is it being used? Systematic acquisition of outside knowl¬

edge is frequently done through information searches, reverse engineer¬

ing, competitive intelligence, and sales-force feedback. Benchmarking is

another increasingly popular and powerful method. The latter involves

finding, learning, and adapting best practices from someone who is doing

Plan, Assess, and Prepare: Phase 1 • 185

the process better. Companies that rely on benchmarking often al¬

ready have a sharing culture that encourages knowledge transfer and

the implementation of best practices. So if your answer to this query is a

resounding “Yes,” your organization most likely has even more knowledge to

share.

Question #5: Are you leveraging knowledge as a product? In your prod¬

ucts? Leveraging knowledge as a product means selling your know-how

and making money. The most famous example is American Airlines,

which sells its travel know-how through its reservation system service,

SABRE. Sears sells its logistic management expertise through its Sears

Logistics subsidiary. IBM and Xerox have set up consulting services

based on their knowledge. Management consulting firms are the ulti¬

mate example. They sell knowledge exclusively and entirely. If you are

set up to take knowledge and turn it into product/services, your organization

is fulfilling the most important objective of KM: turning know-how into

commercial value.

Question #6: Are you using technology to acquire, disseminate, and trans¬

fer knowledge? To everyone? Everywhere? Anytime?

There are two errors you can make with technology investments in

knowledge management: The first is to overinvest—in computers,

software, modems, faxes, multimedia, cellular phones, video confer¬

encing, personal communication systems, and the like. There is no

question that these devices can assist knowledge transfer. They are al¬

most essential. The danger is that you place technology ahead of the

ability or the desire of people to use it. The second error is to underin¬

vest and undermine your transfer initiatives. Naysayers say, “Wait!”

fearing that a new technology will come along tomorrow. If your orga¬

nization has not linked its IT strategy to its KM and transfer strategy, it’s got

a long way to go.

Question # 7: Are you encouraging or discouraging knowledge sharing? Are

people sharing? If not, why not? In too many organizations, knowledge is

not being shared for two reasons. First, sharing often runs counter to the

culture. Until the culture is changed and incentives are created (more

effective use of time, rewards, recognition, promotion), it’s going to be

186 • If Only We Knew What We Know

difficult to get people to share. Second, most people aren’t in the habit

of sharing knowledge. The result is that valuable knowledge generated

every day remains locked up in the minds of individuals throughout

the organization. If the organizational culture runs counter to sharing, you

may be facing a tall—albeit not insurmountable—hurdle in implementing

transfer initiatives.

Obviously, answering seven questions is not a full assessment. We have

found that a fuller assessment emerges while discovering your value

proposition for transfer.

2. DISCOVER YOUR VALUE PROPOSITION

Finding the value proposition means answering this question: Where

can knowledge management principles and tools have the greatest im¬

pact on strategic initiatives, projects, and processes? Usually a KM task

force can find initial answers and direction through interviewing key

players in the organization. Generally, the team’s questions will resem¬

ble the following “to-do” task list:

Start with the business strategy:

• Revisit the organization’s competitive strategy

• Identify key processes and “drivers” that affect the success of that

strategy

Understand the current state:

• Which processes are suffering the most from knowledge gaps

now?

• Are there projects underway now to address these processes? If

the organization already is working on projects to support the

value proposition, then perhaps these projects can be enhanced

through the transfer of knowledge and best practices

• Understand the potential improvement (through internal and

external benchmarking)

• Is there valuable knowledge, and information that could be con¬

verted to knowledge, that could enhance this process if it were

accessible or used in a different manner?

• Assess, at a high level, the leadership and the cultural and

technical landscape that will help or hinder a KM initiative

Plan, Assess, and Prepare: Phase 1 • 187

Develop a KM framework:

• Develop a conceptual model of how KM would help this process

or project. You will use this framework to explain to people why

transfer of knowledge and best practices could lead to better re-

suits.

Make realistic choices:

• Pick something (a project, process, or problem) that the leader-

ship is committed to improving. Develop a value proposition of

how KM would help.

MATCHING SYSTEM WITH VALUE

A KM/transfer strategy must match an organization’s value proposition.

At Buckman Laboratories, for instance, customer intimacy is a top rev¬

enue driver and a strategic imperative. So when the company embarked

on its KM quest, it made sure its transfer projects were targeted at pro¬

viding associates at the point of customer contact with instantaneous ac¬

cess to the collective IQ of the organization. “K’Netix,” the resulting

architecture, is a virtual “library” of the important information relevant

to the business of the customer. The Knowledge Transfer Department

(KTD)—a conglomeration of IS, telecommunications, and technical in¬

formation center—is charged with:

• Accelerating the accumulation and dissemination of knowledge

within the company

• Providing easy and rapid access to the company’s global knowledge

bases

• Sharing best practices with all Buckman affiliates

VALUE PROPOSITION PRINCIPLES FROM SEQUENT

• Designers should start with the income statement, on Day 1, and

work backwards into the design of the knowledge management infra¬

structure.

• Focus on a specific revenue-oriented pressure point (and its processes)

within the firm’s value chain.

• Think big, have good models, and talk about knowledge management

internally, all the time, in practical language, using real-world exam¬

ples that make sense to the people asked to participate in the system.

188 • If Only We Knew What We Know

3. FIND A CHAMPION

It is critical to find a champion or sponsor who (1) understands the

need and (2) has the clout and resources to devote to supporting

knowledge management and transfer initiatives. Based on the value

proposition and the projects identified, it may be obvious who the

champion(s) need to be. Without a powerful champion, the effort will

probably stop here.

Once a champion has been found, the team can create a conceptual

business case for the project or initiative. Then they can get the go-

ahead to inform the organization, and resources to commence a design

phase with a design team.

4. INFORM AND PREPARE THE ORGANIZATION

This task focuses on preparing the organization to understand the role

of knowledge transfer and the project. The scale of this effort will de¬

pend on the scope of the project and how many people need to under¬

stand what is happening.

• Identify key stakeholder (internal customer) groups

• Articulate how they would benefit most from this project

• Inform and involve them

• Identify other successes already in your organization that resulted

from more effective access to and sharing of knowledge to help build

support and understanding

This information-exchange phase is by no means one-sided. By en¬

gaging in dialogue with key groups, you will be able to surface key is¬

sues and identify questions and barriers that will have to be addressed

in designing your KM project.

5. DEFINE THE BUSINESS CASE

At this stage, the team should have a good picture of what projects, or

transfer activities embedded in projects, will really support the value

proposition; what results could be achieved and what it will cost in

time and money. This allows them to define the business case for the

Plan, Assess, and Prepare: Phase I • 189

project, create the business case, and get management approval and

funding to proceed.

AMOCO’S APPROACH TO SHARED LEARNING

Amoco is an example of a KM work-in-process which illustrates the

four-phase process and many of the enablers required to launch a trans¬

fer effort. We will use Amoco as an illustration of the four phases as

they unfold..

In the early 1990s, Amoco’s top management was engaged in orga¬

nizational soul-searching. The goal was improving performance and

reinvigorating the work force. The method was a fresh examination of

the company’s values. The result was a realization that continuous im¬

provement is one of the values that is critical to the company’s present

and future success. The embodiment of this value was called Amoco

Progress, or just “Progress” as it’s known within the company—the cor¬

porate and personal commitment of everyone at Amoco to continu¬

ously improve.

The challenge for Amoco was developing a sustainable process that

would foster ongoing learning and growth of its employees and its busi¬

nesses. It needed a way to ensure that new and improved ways of doing

work would permeate throughout the organization. Not occasionally.

Not when there’s a big push. But every day. And seamlessly.

This was no small task, considering the size and scope of Amoco’s

operations. One of the world’s largest publicly traded diversified petro¬

chemical companies, Amoco has about 42,000 employees in three sec¬

tors—Exploration & Production, Petroleum Products, and Chemicals,

in multiple locations all over the world.

To get its intellectual arms around the potential payoff of sharing, as

well as the organizational elements required to make it work, Amoco

set up a task force in 1994. This “Shared Learning” team was an ad hoc

group of employees recruited from across the company. Members were

handed two mandates: First, examine the potential results of sharing

effective practices and knowledge. Second, examine the necessary sup¬

port systems (i.e., enablers) and structure that would make transfer of

knowledge and effective practices possible.

190 • If Only We Knew What We Know

Meanwhile, on a less formalized basis, networks of experts and com'

munities of practice throughout Amoco agreed to make this issue one

of their key focuses as well. “It was the beginning of a great speeding'up

of learning across all of Amoco,” says Bill Lowrie, president of Amoco

and executive sponsor for Shared Learning. ,

The plan the Amoco task force presented to management will be

described in the next chapter.

Chapter 19

DESIGNING THE TRANSFER PROJECT

PHASE 2

The objectives of the design phase are:

1. Decide the scale of the initiative

2. Use the learnings from others in “best practices in design”

3. Create the action plan and marshal all the resources (in-

eluding technologies, people, and communications)

1. DETERMINE THE SCALE OF THE INITIATIVE

We begin with the assumption that eventually you will want to embed

the transfer of knowledge and best practices into the fabric of your or¬

ganization (including the appropriate enablers: culture, information

technology, infrastructure, and measurement). But you’ve got to start

somewhere. And you ought to start somewhere that’s manageable.

The only thing worse than deciding not to do anything is deciding

to do everything at once and failing to show any success or results within

a reasonable time frame.

THINK BIG, START SMALL

At the outset of this new effort, you must decide how grand your initial

plan and scope need to be. There are two ways you could go: Start with

191

192 • If Only We Knew What We Know

projects and learn from them, always keeping in mind the ultimate ob¬

jective and need for an integrated approach; or alternatively, design a

grand plan and infrastructure from the start, and roll it out gradually.

Our KM study participants spanned the spectrum in their initial

scope (see How They Started, below). Some began by applying KM

and best-practice transfer projects to a specific need or project, as Texas

Instruments did with its Wafer Fabrication initiative, and expanded

from there. At Monsanto, for example, initial success in Ceregen, a

pharmaceutical division, provided the compelling business case for

organization-wide rollout. National Semiconductor began its transfer

of knowledge and best practices work as a grass-roots effort that is just

now getting renewed through top-level support.

Others, like Sequent, Buckman, and Arthur Andersen, took the

“Grand Plan” approach. They began years ago by creating a company-

wide architecture and sharing capability and improved it with experi¬

ence. Arthur Andersen’s KM architecture has changed and evolved

over a five-year period, but AA has had a firmwide knowledge capture

and sharing intention from the beginning. Still others, like CIGNA

Property & Casualty, used KM principles to complement enterprise-

wide reengineering work.

So, from a practical standpoint, the answer to “Where do I start? ”

depends on the firm’s KM readiness (indicated by the self-assessment);

what the business need is (radical improvement now, or steady move

up the curve?); how much buy-in exists into the KM concept at the top

echelons; available resources; and how much infrastructure can realis¬

tically be created to support the initiative.

BIGGER MAY NOT BE BETTER

Even when the resources and management buy-in are present, we still

recommend starting with a scalable project and a flexible overall design.

Why?

First, a “grand design” may be too restrictive for a field that is only now

evolving. “I do not know if any of the alternatives constitute the right

model,” writes Onno van Ewyk, of HCI consulting in Sydney, Aus¬

tralia. “There is a danger in choosing a model at this stage in that it en¬

courages managers to try to construct what looks like the right thing

without attending properly to any of the underlying dynamics,” van

Designing the Transfer Project: Phase 2 • 193

Ewyk continues. “Ultimately, successful business is a performance art

and the players are always the organization’s people (whether the peo¬

ple are employees or operating in some other form of relationship).

Knowledge management can only ever be a means of enhancing and

supporting their performance. If an organization develops KM with

this in mind then the right model will emerge.”

Second, such “grand scheme” may prove too overwhelming—unneces'

sarily. Such models or designs are never “entirely” executed. Spending

lots of time initially (instead of actually doing the work of transfer and

sharing) may lead to paralysis and delay the “good work.” Says Karl

Erik Sveiby, noted KM expert: “Do not try to create a ‘Knowledge Or'

ganization,’ or any other kind of ‘organization’ for that matter, because

models are always unrealistic. They are, at best, approximate represen¬

tations of their designer’s tacit knowledge. Try to ‘see’ instead, by using

these concepts as a language for dialogue and treat the suggested per¬

formance indicators as vision enhancing devices.”

In contrast, there are big benefits to beginning with “chewable”

chunks of projects:

• A project provides early evidence of value.

• Projects allow learning and experimentation on a small, manage¬

able scale.

• Projects help the organization to understand the IT requirements.

• Projects reveal the change management issues:

• The role of champion and sponsor

• The criteria for appropriate projects

• They build experienced and credible advocates

• They identify the required resources

• The design team learns how to learn, how to observe, what to watch

for, and what matters and what doesn’t.

• Finally, projects teach you lessons for future scale-up to the Grand

Plan.

2. BEST PRACTICES IN DESIGN

Whereas designing a KM or transfer project is different from, say, a

reengineering project, some tried-and-true design guidelines apply across

many projects. They include the following:

194 • If Only We Knew What We Know

A CHECKLIST FOR CHOOSING THE “RIGHT” PROJECT

1. Select projects that advance your business performance. Don’t do

“KM” projects. Projects must meet business needs in both appearance

and practice. Depending on your firm’s value proposition, project objec¬

tives might be:

• Reduce cycle times

• Reduce cost

• Enhance product development and innovation

• Increase reuse and efficiency

2. Select projects that have a high probability of success. Particularly

for your first efforts, you want early and clear success stories to help val¬

idate and propagate the cause of knowledge sharing. How can you tell

which one’s a likely winner?

• Can you articulate how knowledge management will contribute?

• What is the business importance outside the immediate user commu¬

nity?

• How strong is the business case?

• Is the project funded?

• Is senior management willing to champion the project?

• What’s the risk associated with failure (as relates to business critical¬

ity)?

• Is the implementation scalable?

3. Select projects which are appropriate for exploring emerging tech¬

nologies. This will give you a head-start on the scale-up while reducing

• Form a multidisciplinary design team of users, technology experts,

and change facilitators. It takes this kind of team to address all the

important issues in a project of any significance.

• Ensure the design team learns as much as possible about knowledge

management applications, barriers, and approaches, and how others

have addressed them. Draw on the experience and knowledge of

other knowledge management practitioners.

• Design the IT support backward from the business issue and objec¬

tive; don’t start with the technology.

• Address the change management questions.

• Identify how this project will help people do their jobs better.

Designing the Transfer Project: Phase 2 • 195

later anxiety about new systems and ways of doing work. To choose the

right IT for your project, assess the following:

• What’s the degree of passion, commitment, and comfort of users with

ambiguity and chaos (trial and error)?

• Does the project provide the necessary attraction (“rush,” reward)

that will enable the user community to sustain themselves through the

change?

• Is the entire culture (user, management, etc.) compatible with the ap¬

proach required?

• If we don’t apply the emerging technology (KM) now, how difficult

would it be to retrofit?

4. Select projects which have significant potential for building credh

bility for knowledge and best practice transfer as a discipline within the

organization. To make the right choice, ask the following questions:

• What is the potential reusability of captured knowledge beyond the

primary audience?

• What is the potential for learnings from this project to be used as a

template/blueprint for future projects?

• Does the process itself require high amounts of knowledge, skills, ex¬

perience, and expertise?

• How will this project help us broaden the depth and the breadth of our

use and understanding of knowledge management?

• Will the people involved in this project be able to help evolve our cul¬

ture toward one that is more “knowledge management friendly”?

• Create common standards for information technology infrastruc¬

ture.

• Develop an approach to capture and learn from your experiences,

and decide how that knowledge will be shared.

• Revisit the objectives and change management strategy constantly,

as the action plan takes shape.

3. CREATE THE PROJECT PLAN

Once developed, a project action plan will include the following infor¬

mation:

196 • If Only We Knew What We Know

1. Key milestones and phases/steps

2. How to involve others

• Whose involvement and buy-in are critical to success?

3. How to communicate and educate

• How to share lessons learned and proposed approach

• How to educate and train those who need to support this effort

4. Resources and budget

5. Assigned “owner(s)” for each major component or deliverable

from the design

IT DESIGN QUESTIONS

An information technology approach must be designed to reflect the

overall project design and approach. IT should not drive the design, al¬

though it will create constraints on what can be done now, while taking

into account future scalability and cross-enterprise compatibility.

The following are some of the questions that need to be addressed dur¬

ing the design phase regarding the use of information technology:

Content: What is the nature of the content (information or knowl¬

edge) that will be involved? How will it be used or shared? Different ap¬

plications are appropriate for different content. Some applications are, by

their nature, organization-wide. Customer databases are an example.

Others, such as discussion forums, may only affect the people in the

community of practice, or project team using the forum.

Audience: Who will be using the content or participating in the

process? How? How does this fit the way they work now?

Access and platform: What technology platform will have the greatest

ease of use and scalability for the future? Are off-the-shelf applications

available? Legacy systems and stand-alone databases present a challenge.

Use and interaction: How will the content be captured and refreshed?

How will quality be addressed? What are the norms of behavior? Are any

policies required?

People and roles: What support roles and people will be needed to en¬

sure results? This includes technology support, facilitators, coaches, and

information resources. (See Chapter 11 on infrastructure and facilita¬

tors.)

Designing the Transfer Project: Phase 2 • 197

THE DESIGN PHASE IN AMOCO

By November 1995, Amoco’s strategic planning committee accepted

the recommendations from the Shared Learning team, including:

1. Set expectations for the capture, transfer, and use of learnings

and best practices across the organization and from external

sources. Establish the strategic importance of doing so to improve

business.

• Define key strategic areas/processes where best practices should

be identified and used throughout the organization.

• Make the transfer of learnings and best practices a part of busi¬

ness strategy and develop annual performance plans to identify

and close performance gaps

• Assess progress in this area through milestones related to closing

performance gaps, such as the achievement of defined bench¬

marks and other stretch targets, through the ongoing monitoring

of metrics related to the capture, transfer, and use of processes,

and through insights gained and improvements achieved

through the Amoco business assessment process.

• Develop plans to overcome identified barriers to implementing

the recommendations of this study in their organizations.

2. Make a computer system available worldwide for storing and re¬

trieving learnings and practices, and manage the system to pro¬

vide the best available learnings and practices wherever and

whenever needed.

3. Define, encourage, support, and manage competency networks in

key strategic functional areas that can leverage knowledge, and

assist in managing the knowledge capture and transfer processes.

These networks should be well publicized and given responsibil¬

ity for providing “experts-on-call” services.

4- Incorporate effective learning/sharing behaviors and competen¬

cies into the Leadership Development Process, the Performance

Benchmarks of the Amoco Performance Management System,

and into Leadership and Progress-related training. Promote aware¬

ness of the many successful approaches that have been used within

198 • If Only We Knew What We Know

and outside Amoco for the capture, transfer, and use of best prac¬

tices.

5. Promote corporatewide recognition of individuals, teams, and or¬

ganizations who are role models of capturing, sharing, and using

learnings and best practices.

6. Commit additional resources at the corporate, business group, and

local levels to support implementation of the above recommenda¬

tions, in particular to create and manage the knowledge capture,

transfer, and use processes, and to further develop Amoco’s capa¬

bilities in these areas.

Based on these recommendations, the Amoco Steering Committee

launched its early projects in knowledge sharing and transfer and cre¬

ated a corporate infrastructure to support those projects. They moved

quickly to implementation, as illustrated in the next chapter (see page

205).

Chapter 20

IMPLEMENTATION

PHASE 3

The objectives of this third phase are action-oriented:

1. Launch the project

2. Provide support for both content and process

3. Observe and learn

4- Achieve results

The seeds of a successful implementation are sown in the first two

phases. If you have a compelling value proposition and clear busi¬

ness case, a passionate champion and a culture bent on sharing, ade¬

quate resources and a realistic plan—then implementation involves

taking the plan off the paper and into the trenches. It’s about turning

“knowledge” into action.

It’s also probably the toughest phase of the process because you will

find—as many have—the best plans falter in the face of day-to-day re¬

alities. To survive and excel, remember this: Keep your plan flexible and

adjust it constantly to emerging circumstances and oscillating emo¬

tions. And they mil oscillate.

1. LAUNCH THE PROJECT

Just how to implement a transfer project will be determined by the par¬

ticular project and the type of knowledge and practices being trans¬

ferred. The following are four prototypical projects.

199

200 • If Only We Knew What We Know

1. Learning from prior/new product launches. For example, if the trans¬

fer project is to capture and use lessons learned from prior new product

launches, implementation would focus on bringing the teams and indi¬

viduals with past experience together with those who are part of the

new product launch team. Initial meetings or conversations would re¬

volve around what worked and what didn’t, how past experiences re¬

late to the current initiative, what is different and what can be

adapted, and what should be stored in a database as universal knowl¬

edge” valuable enough to be stored in the organization’s long-term

memory.

2. Improving operational performance. If the project is intended to

identify and transfer best practices between similar operations, the ap¬

proach that often works best is forming cross-unit best practice teams,

composed of “experts” from the disparate locations. The teams go

through natural phases, and need facilitation at each phase: (1) identi¬

fying what they really mean by the process and creating a shared vo¬

cabulary (i.e., scoping and mapping the process); (2) identifying

current practice; (3) comparing performance to identify best practices;

and (4) adapting and implementing. Facilitators will be required to

help the teams throughout the process, then coach and monitor imple¬

mentation.

3. Reusing designs. If the project’s goal is to capture historical infor¬

mation or engineering designs that can be reused, then the implemen¬

tation challenge is far different. The planning and design phases would

have identified the types and nature of knowledge that must be cap¬

tured and codified, as well as the correct IT infrastructure to allow the

knowledge to be reused. But “theoretical” design—even the best

sort—is radically different from cajoling and training people to actu¬

ally work differently by getting others’ designs, and getting their con¬

sent that indeed, the effort (to change) is worth it. Again, the seeds for

success must be sown during the earlier phases. If there is a champion

with enough clout and a compelling business need, then the odds are

greater that people will go through the discomfort and frustration of

changing the way they work.

4. Building communities of practice. If the objective of the project is to

create ongoing networks of likeminded people or communities of prac-

Implementation: Phase 3 • 201

tice, then facilitators again play a key role in bringing people together

across functions and locations. It will be the role of facilitators to help

participants set objectives and learn to communicate and learn from

each other: This “leaming-todearn” process is the most critical part of

implementing a network-oriented transfer effort.

2. PROVIDE SUPPORT

No matter how “specific” the objective or effort, and whether it’s de¬

signed to improve product development or increase operational excel¬

lence, some “basic” truisms run through most implementation phases:

1. Face-to-face is critical—at the outset. Early “knowledge” exchanges

involve a wealth of tacit knowledge. Consequently, it’s not enough to

simply have a “historian” meet with the “past” teams to “record” what

they’ve learned. Recording is a pale and wholly inadequate imitation

of the kind of dynamics and problem-driven questions that a new team

is likely to ask if they are in a dialogue with the more experienced crew.

So let them meet, and interact, if you want to ensure your implemen¬

tation phase gets off on the right foot.

2. A one-shot deal is never good enough. “Old” and “new” teams—or

simply teams from various sides of the organization—will need to have

ongoing access to each other, in order to confront new issues and ad¬

dress problems as they arise. A one-shot “recording” session or even a

one-shot face-to-face meeting is unlikely to achieve long-term success.

To guarantee the effort remains an iterative process, build in constant

access and interaction.

3. Facilitators will earn their “money’s worth.” No matter whether the

implementation process involves different functional groups, or simi¬

lar groups but at disparate locations, it helps to have a champion or fa¬

cilitator present to bridge attitudes, vocabulary, and knowledge gaps.

A need for “brokers” has been driving the spectacular growth in the

management consulting business. It’s a lot easier to hear criticism and

new ideas if you do not perceive the “conveyer” as a threat. Plus, every

process needs a “master,” or at least a “nurturer,” to make sure it does

not fade into corporate oblivion. Designate internal “consultants” who

202 • If Only We Knew What We Know

are respected by employees or have a natural inclination to bridge to

help keep your implementation process on track.

4. Do everything possible to achieve early, visible results. The attention

span of today’s organizations is shorter than'ever, because the demands

BUCKMAN LABORATORIES LAUNCH “AIDS”

Bob Buckman, vice-chairman of Buckman Laboratories, was the key ar¬

chitect of the original design for their knowledge management and infor¬

mation technology system. Buckman realized in implementing the

knowledge sharing culture that it would not be easy to get all associates

motivated and involved. He offers the following tips.

Make it fun. On-line contests can be used to increase involvement.

Buckman has a Notice Board Section, called Breakroom, where all

associates go to ask questions on a wide range of social topics, e.g.,

items for sale or wanted, requests for a good local home-mainte¬

nance professional, interesting restaurants, helpful tips in arrang¬

ing a foreign vacation, any item of personal interest that does not

violate the Code of Ethics.

Lessen the stress. Buckman employees were allowed to access the sys¬

tem in a more relaxed environment, such as from their homes.

Nominate captains. At Buckman, two experts in each forum were ap¬

pointed to play the role of “answer-givers.”

Qet all associates on the system. With involvement of all associates,

the risk of losing untapped resources is minimized.

Renew the passport. Implement the system on a global basis so that all

cultures are comfortable and maintain involvement.

Provide functionality as well as portability. Because of global opera¬

tions, traveling staff need computer “survival kits.” These are

signed out based on a specific country and contain adapters, cords,

and other peripheral accessories.

Make it easy. The change is difficult enough so it must be made as

easy and user-friendly as possible. K’Netix operates in a Windows

environment and was designed to be both intuitive and flexible.

Provide training. Informal training, intensive training, and ongoing

training help with the transition. At Buckman, documentation is

provided in the form of a K’Netix training/reference manual.

Monitor progress. By whatever factor(s) indicate success in a given

system, that progress should be documented.

Implementation: Phase 3 • 203

are constantly changing. It is critical that projects selected for these

early transfer and KM pilots show visible results within at most six

months.

IMPLEMENTATION AT CHEVRON

In 1992, with a corporate imperative to cut costs (the value proposb

tion), Chevron set up thirteen best practice teams composed of refin¬

ery managers, operations personnel, best practice study personnel, and

volunteers. Together, the teams set out to examine “hard refinery

processes” like crude distillation, as well as corporatewide processes

such as energy efficiency and conservation; their ultimate goal was to

identify the “best” ones and then help others within the organization

implement the best practices. The teams’ initial focus was on areas

where improvement could yield short-term, high-value results.

Indeed, according to Jim O’Brient, manager of quality improvement

in Chevron’s USA production company, it should take no more than

six to twelve months to achieve some success and gain credibility.

Projected results from each project ranged from $10 million in sav¬

ings to $150 million. Front-loading success—and picking improve¬

ments that can yield visible change within a reasonable time—have

been critical to building overall credibility and facilitating later scale-

up.

By the end of the first year, the energy efficiency team generated

$150 million in initial savings; by 1997, they had achieved a total sav¬

ings of $648 million, by both following a model that takes each initial

“saving idea” and scaling it up across Chevron for greater results.

The gas compression team created $20 million in sustainable first-

year improvements on a $500,000 investment (mostly people’s time).

Similarly, Chevron has taken the initial thirteen teams and scaled

them up to help ignite change enterprisewide. Teams are focused on

specific business areas. Their work is credited with improving every¬

thing from catalytic cracking to plant maintenance. Each team meets

three to four times a year and sets aside time to share ideas and lessons

learned, as well as categorize knowledge into various levels of “applica¬

bility” from good ideas to best practices.

204 • If Only We Knew What We Know

<

To keep the improvement and transfer work on track, Chevron has

been relying on “experts” and “process masters.” These professional fa¬

cilitators are charged with providing continuity and keeping the process

moving as well as ensuring project goals and business goals remain in¬

timately linked.

The focus on best practice transfer continues to spread in Chevron.

For example, Chevron spends nearly $6 billion a year on capital pro¬

jects. In 1991, they benchmarked the best practices of sixty-plus com¬

panies and developed a world-class project planning and management

process. The task then became deploying the best practices methodol¬

ogy, known as the Chevron Project Development and Execution

Process process, to all the capital projects in Chevron. All the lessons

learned from earlier transfer work have supported this effort. The

CPDEP (“chip dip”) network of best practices in capital projects has

made a dramatic difference in project management timelines and costs

at Chevron. There has been over $816 million in savings in down¬

stream capital projects since 1992. Adding this to other savings,

Chevron has reduced its annual operating expense by over $1.5 billion

since 1992. On top of all the savings, Chevron is creating a culture

that values learning from others as a source of competitive advantage.

BRITISH PETROLEUM

In 1994, British Petroleum Exploration (BPX) launched a “virtual

teamwork program.” The project was designed to create effective ways

for members of teams to collaborate across different locations. The

goal: building a network of people, not a storehouse of data, information,

or knowledge.

BPX realized that with the level and wealth of technical expertise

within its organization, it would be far more efficient to give employees

access to other employees’ tacit knowledge than to capture and codify

that knowledge electronically or on paper.

To get the process rolling, BPX’s “Change Management Team” (the

design and implementation team) developed a coaching program to

show participants how to use the technology (when necessary) as well

as help them understand how reaching out to others could further

Implementation: Phase 3 • 205

their work. The company decided to call the implementation help

“coaching,” not “training,” to ensure it retains a touch of personal in¬

teraction like a coach working with players rather than a trainer pre¬

senting information to passive recipients. The coaching curriculum

included not only the “how to,” but “why” and “what.” About half the

budget was spent on coaching. After the pilot project was completed,

the Change Management Team was reincarnated as the Knowledge

Management Team, responsible for roll-out across BP (the eventual

scale-up).

AMOCO’S “FAST” IMPLEMENTATION

In February 1996, Amoco appointed Dave Ledet the director of

Shared Learning, and he began formal implementation. When the

company announced Ledet’s new role, Vice President of Progress Don

Tomberg sent a letter to leaders of all Amoco’s business groups and

units asking them to use the Shared Learning Study as a guide to ac¬

celerating transfer of knowledge across the company. It was clear that

Ledet and Shared Learning had support from the top.

Befitting a large-scale change initiative, “implementation” involved

a clear understanding and ongoing management of what Amoco iden¬

tified as the four key enablers of successful Shared Learning—focus,

alignment, support, and time (FAST).

Focus helps people determine what knowledge should be captured

and transferred. It is critical to successful Shared Learning because it

allows people to see a clear reason for this activity. Many groups at

Amoco focus their Shared Learning efforts on elements of their busi¬

ness strategy. “Focus is relatively easy at a local level,” Ledet says. “Or¬

ganizations know what they want to accomplish, whether that’s

growth or process improvement. What gets a little harder is determin¬

ing focus for sharing across internal organizations, but that is where

you break down boundaries and get the best benefits out of shared

learning.”

At Amoco, alignment involves the cultural and performance metrics

that help match individuals’ goals to those of the company. Instead of

issuing a corporate mandate (push) to share, Amoco focused on gener-

206 • If Only We Knew What We Know

ating demand (pull) for Shared Learning by raising employee aware¬

ness of it; shifting the company culture to make Shared Learning an

expected, normal part of work; and embedding it into Amoco s

processes. Ledet and his team dedicate significant time to working with

key personnel around the world, including corporate leaders whose

concerns and whose public and private speeches often set a pace for

the company. Other channels leveraged to generate demand for

Shared Learning include newsletters, e-mail bulletins, demonstrations

at internal conferences and meetings, and an intranet site. Ledet and

his staff are also pursuing alignment by working with Human Re¬

sources to include Shared Learning as an element in the company s

standard performance review processes.

“People don’t do this because we’re ‘dangling a carrot’ in front of

them. They do it because it fundamentally makes sense to them that

we need to know what we know. It makes sense that we need to de¬

velop new knowledge,” says Ledet. “The only way to make learning

systemic in the organization is by integrating learning principles into

processes, such as project management or strategic planning.

To support its people’s efforts to share, Amoco provides coaching,

tools, and processes in a wide range of areas, including developing a

change management framework, training in Shared Learning, and de¬

veloping and using database applications. “Once people are ready to

try knowledge management, you must be ready with resources to help

them,” Ledet says. “Otherwise, they hit a dead end. If that’s allowed to

happen, it takes five times the effort to get them to try again.”

Infrastructure support is guided by a corporate-level change man¬

agement team of two: Ledet and Julie Greer, progress consultant. A

battalion of change agents in every Amoco business unit bolster the ef¬

fort on the front lines. These quality/progress professionals help em¬

ployees make their Shared Learning processes work on a local level.

They spread the word. They answer questions. They make sure em¬

ployees have access to tools.

The other element of the infrastructure support picture is technologi¬

cal systems. In the original Shared Learning study, Amoco identified

Lotus Notes as the most appropriate tool for developing a corporation¬

wide repository of knowledge. The original format developed for the

database has been reused many times by special interest groups around

Implementation: Phase 3 • 207

the company. To find the information they needed, users accessed a

knowledge resource Index (KRI) which listed knowledge sharing fo¬

rums, such as databases, communities of interest, expertise lists, meet¬

ings, internal web sites, and others by topic. The next-generation

navigation tool—incorporating intranet technology—allowed users

not only to search for resources on a particular subject but also to ac¬

cess them directly through the navigation tool/browser. Amoco is

using Lotus Notes and Domino to make interactive knowledge sharing

databases available to everyone via the Netscape browser.

Employees may see the value and have the systems and training to

facilitate sharing, but more often than not, they don’t have the time.

“While we all recognize that learning ultimately results in a net sav¬

ings in time and work, fighting the alligators often takes precedence

over draining the swamp,” according to Ledet. To ensure employees set

aside time for sharing knowledge, Amoco has been promoting regular

meetings of communities of practice. It has also emphasized that those

who share will be recognized. For example, success stories from the

front lines were shared at a May 1997 global Amoco conference,

topped by a presentation of the Chairman’s Award for Excellence to

fifty teams best using applied Amoco progress principles for superior

business results.

Chapter 21 -

TRANSITION AND SCALE-UP

PHASE 4

The objectives of this phase are to:

1. Capture success stories and publicize early results

2. Use knowledge gained to expand the scale-up

3. Create a new organizational structure to oversee

the ongoing process

1. CAPTURE SUCCESS STORIES AND PUBLICIZE EARLY RESULTS

You’ve done it! You found the right project, identified the value propo¬

sition, designed a realistic plan, and implemented your first transfer

project. And it works!

Now what? Now, it’s time to figure out how to harness the engine of

KM and best practice transfer to turbo-power your entire organization.

The first order of business is leveraging the experiential knowledge

you’ve already got. It is important to have a record of early successes,

and elaborate anecdotal examples and testimonials about the contri¬

bution of knowledge management to the business results, so:

• Capture success stories; they can be used for learning and “selling.”

• Identify and list lessons learned.

• Codify the first good experience—and enlist the participants as mis¬

sionaries and coaches for future projects.

208

Transition and Scale-Up: Phase 4 • 209

Then:

• Use lessons learned to create an expansion and scale-up strategy.

• Identify the entity that will guide, support, and monitor KM on an

organization-wide basis.

It is the explicit need/inclination to observe, learn, and capitalize on

existing know-how that sets KM and best practice transfer apart from

other improvement projects. Hence, throughout the early stages, both

facilitators and participants must capture and transfer learnings about

the process of design and implementation so they can (1) enhance the

organization’s core competency at transfer, and (2) prepare for scale-

up—expanding the process to other topics and constituencies.

2. USE KNOWLEDGE GAINED TO EXPAND THE SCALE-UP

We’ve used the example of Texas Instruments throughout this book.

We go back to it here because we believe that what makes TI’s best-

practice transfer effort sustainable is the conscious effort to learn from

initial tryouts and apply these learnings to ongoing scale-up initiatives.

For example, during the assessment phase of TI’s first year of transfer

work—the “year of supply” as TI has dubbed it—the organization’s best

practice commando teams not only identified “best practices”; they

also identified gaps and strengths in the company’s efforts to share.

Both the gaps (need to do better or know more) and strengths were

documented by best practice facilitators. The knowledge captured dur¬

ing the first year led to the “year of demand”—the second phase of the

organization-wide program which involved a massive cultural change

initiative aimed at creating a demand for solutions to gaps identified

earlier.

At Buckman Laboratories, the vision of a full-scale effort, a way of

connecting associates with powerful knowledge bases, existed from day

one. Even so, Buckman started with a pilot project, focused on the

sales people, and only then scaled up. Now all 1,200 employees have

laptops and CompuServe accounts.

210 • If Only We Knew What We Know

3. CREATE NEW ORGANIZATIONAL STRUCTURES

In many cases, the scale-up stage also requires the formation of a new

organizational entity. A pilot project can possibly be managed part

time, or by a makeshift team of temporary facilitators. However, if you

intend to embed KM and transfer into the core of your operations, and

expand their reach enterprise-wide, the human/organizational infra¬

structure must be institutionalized to ensure continuity and focus.

• At Buckman, the scale-up led to the merger of IT and KM into the

Knowledge Transfer System (KTS) department which is currently

charged with managing all aspects of KM within Buckman.

• At Texas Instruments, early successes led to the formation of the TI

Office of Best Practices, which has two functions: (1) it defines the

best practice sharing process, facilitator and champion roles, as well

as technology infrastructure for sharing; and (2) it is the engine for

developing and deploying training and infrastructure supported by

the facilitator network and sharing process.

SEQUENT’S GUIDE TO IMPLEMENTATION AND SCALE-UP

Originally focused mostly on marketing and sales, Sequent Computers

took the following steps during design and implementation to support

eventual scale-up:

Start by providing a common space for storing and retrieving

knowledge and having discussions. Then, examine what people use,

what they contribute to, and immediate feedback they give to redesign

the system. Look at trend patterns of use. Then, begin to engineer

more sophisticated social expectations about use. Finally, roll the sys¬

tem out to more people and design new releases that incorporate learn¬

ing from initial experiences.

SUMMARY: THE “SCALER-UPPER” ACTION LIST

When it’s time to shift from small-scale to full-scale, most organiza¬

tions will need both to address learnings from the pilot and to prepare

the organization for scale-up. Some of these “preparatory” actions

might include:

Transition and Scale-Up: Phase 4 • 211

1. Providing a forum for lessons learned and information sharing/

collaboration on existing projects and initiatives that involve

KM principles

2. Reviewing and adjusting (if necessary) corporate reward and

recognition programs, including incentive compensation, to

focus attention on knowledge sharing with the goal of achieving

business objectives

3. Developing a means/process for advertising business and technob

ogy initiatives that focus on the collection and sharing of cus¬

tomer and employee information to reduce or eliminate redundant

efforts

4. Reviewing and adjusting corporate and business systems to en¬

able collaboration and communication within and among busi¬

ness groups

5. Designating a function to coordinate and maintain focus on

knowledge sharing to achieve corporate objectives

AMOCO’S TIME TO SCALE UP

Throughout 1996-97, Amoco made great strides in creating a culture

of Shared Learning. “My sense is that when we started to do this peo¬

ple weren’t sure what we wanted to do or that it was okay to do it,” ac¬

cording to Dave Ledet. “Although over 500 people were involved in

building the Shared Learning strategy, in retrospect, we should have

involved more to build commitment. Still, by getting Shared Learning

in front of employees and facilitating some successes around the com¬

pany we were creating an environment where people see Shared

Learning is something they need—and are expected—to do. Just as it

is expected that we show up at a certain time in the morning, it’s ex¬

pected that we share our knowledge.”

With the basic concepts in place, Ledet and his Shared Learning

team are working to enable the change to be integrated deeper into the

organization. While some groups within Amoco are sharing their

learnings instinctively, some have only taken the first steps to develop¬

ing a formalized change plan within their own groups. A systematic,

process-focused approach to Shared Learning is what is needed to

212 • If Only We Knew What We Know

capture the maximum benefits, and cause the corporation to think and

act as one. While Amoco has realized significant benefits and improve¬

ments due to their Shared Learning implementation (below), there is

still a lot of improvement left to accomplish.

IS IT WORKING AT AMOCO?

Evidence #1. In the Norwegian North Sea area, Amoco drilling opera¬

tions personnel recently had to quickly decide how to fix a crippling

problem with an operating well. Using a company-wide synthesis of

best practices in drilling, the team coordinated the complex work in

only one day. Without access to shared knowledge, the same project

would have taken five to ten days—at least.

Evidence #2. In Texas City, Texas, crews at the Amoco refinery are

using hand-held computer technology transferred from a plant in

Wyoming— yielding hundreds of thousands of dollars in working cap¬

ital gains through avoided equipment failures, increased run times, and

reduced maintenance costs.

Evidence #3. At Richard Elsenheimer’s office in the Amoco Chemicals

plant outside Alvin, Texas, the phone is ringing off the hook. Since

Elsenheimer spoke at a Community of Interest (COI) meeting of

Amoco maintenance managers last year, a steady stream of mainte¬

nance personnel from around the country have been calling him to

learn what his team has done and to cooperate for mutual gain.

Evidence #4. While planning the Purified Terephthalic Acid (PTA)

plant in Geel, Belgium, the team learned from other PTA projects at

Amoco, applied the learnings, and reduced the overall project cost by

10 percent from the original estimate. They also applied new technol¬

ogy that recovers energy from the process to allow the plant to actually

produce and export electrical power, and to significantly reduce waste

water treatment loads by recycling the solvent stream.

Transition and Scale-Up: Phase 4 • 213

Evidence #5. The Trinidad Upstream Development Project provides

for onshore and offshore facilities to produce 550 million standard

cubic feet of gas daily for a liquified natural-gas plant on the West

Coast of Trinidad. This project team also learned from other project

teams around Amoco, and applied a novel contracting strategy. They

formed a true partnership with six contractors and involved them early

in the project, for an estimated saving over the life of the project of

$85 million. The performance contracts allowed all of the members of

the partnership or alliance to make additional profits by realizing a

high level of performance.

Evidence #6. The Marlin project is Amoco’s initial step into deepwater

Gulf of Mexico, a $500 million tension leg platform (TLP) drilling in

over 3,200 feet of water. The team also used learnings from other parts

of Amoco and external companies to develop the project. As a result

of prior experience, they decided to have two separate engineering

firms develop different designs concurrently and then picked one to go

forward with. The cost of the additional unused design was far out¬

weighed by the cycle time saved, and the increased project efficiency,

both in design and construction. The overall estimated savings for the

project are $20 million.

In addition to these results, Amoco is also measuring progress on

knowledge sharing on a more formalized basis. The primary measure is

an annual survey which gauges, among other things, awareness of the

Shared Learning concept and employees’ application of it. “Basically,

about 60 percent of people in our company favorably respond to ques¬

tions like ‘Are you capturing and transferring your best practices?’ ‘Are

you using effective practices from other organizations?”’ Ledet reported

in 1997.

Ledet spends a good amount of time gathering more informal feed¬

back. Most Amoco managers tell Ledet that they know Shared Learn¬

ing is working when they see people applying lessons learned from

others. Amoco also collects activity measures, such as how often

Shared Learning databases are used. Nonetheless, most of the data

gathered is about project-specific reports on the success of Shared

214 • If Only We Knew What We Know

Learning. Many of these do involve hard financial results. For exam¬

ple, Amoco drilling operations expects to save $50 million a year

through Shared Learning.

Amoco’s Levels of Knowledge

Earlier we outlined the way AMP, TI, and Chevron define the “best” in

best practice (see What Does “Best Practices” Mean Anyway? on page

12). To make its burgeoning community of databases easier to manage,

Amoco also developed a multilayered approach that categorizes expe¬

rience and practice into three levels:

1. Tier I, or the “innovative knowledge level,” has the lowest degree

of structure and minimal criteria for material entered. The goal is

to allow users to freely input their experiences, successes, lessons

learned, or requests for information. Quality of the material input

varies, but by maximizing the amount of input the potential for

collaboration and virtual learning is also maximized.

2. The next level of the taxonomy is Tier II, or “proven knowledge.”

Data at this level can be the result of collaboration on Tier I. Al¬

though it may not have corporate-wide application, the transfer

of information at this level has proven value. It is also more struc¬

tured and meets criteria for inclusion. Tier II Shared Learnings

include local methodologies, policies and procedures, emerging

best practices, success stories, effective performance measures,

and lessons learned. However, the structure is still flexible

enough to encourage further collaboration and growth of knowl¬

edge capital. At this tier, collaboration enables identification of

best knowledge, including some industry best practices.

3. Tier III is “best knowledge.” This is where the company’s “secrets

of success” will reside, including methodologies, best practices,

core competencies, policies, and values. Information at this level

requires highest-level approval and rigorous screening.

THE FOUR PHASES IN ACTION AT THE WORLD BANK

In October 1996, World Bank President James Wolfensohn announced

at its annual meeting that the World Bank planned to capture and or-

Transition and Scale-Up: Phase 4 • 215

ganize its experience and know-how and make it available not only to

internal staff but also to clients, partners, and stakeholders around the

world. The World Bank was to become “the Knowledge Bank,” with

substantial transformation of the international organization which

aims at reducing poverty and improving the quality of life in develop¬

ing countries through lending and advisory services.

Using Stories to Drum Up Support

Introducing knowledge management was a significant challenge. At

the time, there were some pilot efforts under way in several areas, in¬

cluding the education sector, private sector infrastructure, and the live

data base in Africa. These efforts were perceived as successful and

helped make the case for introducing the approach more widely across

the organization. Many other key elements were missing. There was

little agreement on the strategy for knowledge management, and there

was no budget or institutional decision-making mechanism for knowl¬

edge management, and no way of tracking or monitoring activities.

In mid-October 1996, Stephen Denning was appointed program di¬

rector, Knowledge Management in the World Bank to help introduce

knowledge management on an institution-wide basis. A year later, the

main elements for implementation were in place. Knowledge manage¬

ment would be part of a strategic compact that was unanimously ap¬

proved by the board of executive directors in March 1997. A

substantial budget for fiscal year 1998 was passed and put in place. The

Bank established an institutional decision-making mechanism with

two representative forums—the Information and Knowledge Manage¬

ment Council and the Knowledge Management Board. They also now

had a method for tracking and monitoring the progress of implement¬

ing knowledge management in the Bank.

The use of stories was a key element in explaining the concept of

knowledge management. Stories were more effective than charts, or

reports. Dialogue was effective but difficult to use for large numbers of

managers and employees.

An early powerful story was that of a task team in a Latin American

country. The client country wanted quickly the Bank’s global experi¬

ence of the policy and political implications of education reform. The

Bank’s task team contacted the help desk of the Education Network,

216 • If Only We Knew What We Know

who quickly obtained the relevant material from the relevant commu-

nity of practice. The task team was able to meet the client’s needs more

quickly and completely than they otherwise would. This in itself was

an accomplishment, but knowledge management takes the process fur¬

ther. The material that was developed in this engagement has now

been identified as “valuable knowledge objects” to be edited for inclu¬

sion in the Bank’s electronic knowledge base so that they can be in¬

stantly accessed by anyone else in the organization. When the Bank’s

knowledge base goes external, the materials—suitably edited—will be

instantly available to clients and partners and stakeholders around the

world, thus enormously expanding and accelerating the reach of the

Bank’s know-how and expertise to groups that would otherwise not

have access.

Complex system ideas like this could be communicated effectively

and widely in this story format, whereas efforts to explain them in

charts, or reports, often ran into difficulties.

Denning found that stories work because they are efficient “carriers”

of high-impact tacit knowledge. They contain huge amounts of knowl¬

edge in tiny containers. They are able to show the interconnections

and the ultimate results.

Learning the Unexpected

Sometimes implementing a KM initiative is a lot less simple than it

may initially appear. For instance, simply “outlining” benefits, or illus¬

trating them on a flow-chart, may not convince people to change the

way they behave. To adopt new processes. To join the effort.

This was not the only practical lesson the Bank learned along its

one-year process of embedding best practice transfer into its core oper¬

ation. The Bank started out with a few “bad breaks” and “dumb luck.”

The bad breaks may sound familiar—they included a middle man¬

agement that needed a lot of convincing, a downsizing environment,

and the usual legacy of fragmented and inefficient IT systems.

These bad breaks were more than offset by four pieces of luck: (1) a

dynamic and charismatic leader who was on the lookout for a new

strategy; (2) a highly motivated professional staff with (3) a large un¬

tapped knowledge potential; and (4) the invention of the Web.

Luck is good, but never sufficient. So to really get things going, the

Transition and Scale-Up: Phase 4 • 217

Bank’s KM team designed the following steps: create a compelling

business vision of knowledge management; build a robust managerial

coalition in support of change; define knowledge management activi¬

ties in specific monitorable activities; map the knowledge “domains” of

the organization; and make the case for an adequate budget.

Finally, they learned from others! Benchmarking best practices in

an APQC consortium study helped the Bank learn and leapfrog to

being one of the leading practitioners after only a short time.

What did the KM team learn through its experience? A lot, and

most of it violated their initial assumptions:

Assumption #1: The assumption that “We know exactly where we

are going” was unsound, as the organization kept reinventing it¬

self and finding its way on the knowledge journey.

Assumption #2: “What we need is knowledge.” While knowledge is

valuable, they also found that there is much valuable know-how

that has not yet been authenticated as knowledge. In one sense,

says Denning, “Knowledge is yesterday’s innovation.”

Assumption #3: The idea that an intranet would largely obviate the

need for help desks proved illusory. The KM team found that

people-to-people contact remained crucial.

Assumption #4: “Systematizing KM will kill it.” The KM team hesi¬

tated to define KM activities in case it turned it into a supply-dri¬

ven bureaucracy. Although the danger is still there, Denning

reports that in general, describing KM as a system seems to have

helped people get comfortable with KM and encourage conver¬

gence towards a common institutional vision.

Assumption #5: “A mouse-click away is close enough.” This turned

out to be wrong. KM has to be embedded in the work process.

The Bank’s Scale-up Challenges for the Future

With all the learning it has achieved, the Bank is still facing four key

implementation challenges.

Making communities of practice work. Although the communities are

organized in a logical fashion, knowledge, being inherently un¬

tidy, does not always fall into their logical organization. It sprawls

across the boundary lines.

218 • If Only We Knew What We Know

Making the organizational culture shift. It seems like everyone in the

Bank is perfectly willing to share. However, Denning has found

that the problem with ensuring true flow of knowledge, across the

enterprise, is on the demand side, not the supply. People are will¬

ing to share—if asked. People simply don’t ask.

Getting best practices. The Bank has found that even when you figure

out what works, what works some of the time, and what almost

never works, a reported best practice invariably leaves out many

insights and is hard to keep up-to-date. Developing a dynamic

knowledge base remains a major challenge.

Figuring out external access. Since ultimately, the Bank wants to

open its knowledge treasure chest to external clients, it must re¬

solve questions of copyright, confidentiality, charging for knowl¬

edge, the internal impact of external access, and languages.

SUMMARY: PART FIVE

So what do you do on Monday morning? To summarize, let us be alle¬

gorical and cram a year into a fictitious day.

8:30 A.M. Start planning. Assess just where you stand on the KM

learning curve. Define your value proposition. You need to know

where you are now and where to go before you embark on any

change efforts.

9:00 A.M. Find the processes and projects that support your value

proposition, find champions to lead your KM quest, inform the

rest of the organization, and define a business case strong enough

to hold management’s attention and commitment.

10:30 A.M. Select a project that gives you a good chance of early

success and a testing ground for emerging technologies.

11:00 A.M. Begin your design work by following tried-and-true prin¬

ciples of design, such as employing multidisciplinary teams and

addressing process, change management, and business issues for

a given project.

12:00 Noon. Break for lunch! (And to get buy-in and understand¬

ing in the organization.)

1:30 P.M. Implement, launch your project, train the participants,

Transition and Scale-Up: Phase 4 • 219

provide support, observe, learn, and record results to help spread

the good word.

4:00 P.M. Organize your learnings from the pilot, and use the lessons

to create an expansion strategy that embeds KM and transfer into

every area of your organization.

5:00 P.M. Plan the next day—the expansion and scale-up.

‘

*

Part Six

CONCLUSION

There is no conclusion to managing knowledge and transferring best

practices. It is a race without a finishing line. And this is even more

true in the knowledge era as we move toward the Millennium.

But there is a conclusion to this book. Before we end, we want to share

with you some of the overarching themes of our three large-scale studies on

knowledge management and the transfer of best practices. These “consor¬

tium” (i.e., multicompany) benchmarkng studies led by APQC included

more than seventy of the leading organizations practicing knowledge man¬

agement in the United States and in Europe.

The organizations in our studies met together frequently, using APQC’s

benchmarking methodology, and shared their best practices in knowledge

management in their own organizations. Also, they sought out additional

“best practices” organizations (“partners”) across the United States and Eu¬

rope, site visited them, and included their knowledge in the findings. These

studies are probably the most comprehensive empirical collection of what ac¬

tually goes on in organizations in knowledge management and the transfer of

best practices. (APQC, 1996; APQC, 1997)

We have already encapsulated some of these findings in the preceding

chapters. But in this Conclusion, we share with you ten enduring principles

that emerged from the studies, and which will guide you as you move for¬

ward.

V

4

v.

Chapter 22

ENDURING PRINCIPLES

The APQC’s consortium studies on knowledge management and

best practice transfer were conducted during 1995-1997 and

other studies were under way as this book was published. Participating

companies represented a broad range of industries, from computers and

electronics to financial/investment and chemicals.

Each of our six-month studies gave us a unique and treasured

chance to “peek” behind the scenes of KM at some of the early work,

the experimentation, the success stories. Most of our benchmarking

participants had been actively involved for years in isolated, frag¬

mented KM efforts. What sets them apart is that they have, more re¬

cently, become more conscious and organized in their attempts to

manage knowledge and transfer practices. In fact, for some, it wasn’t

until we approached them about participation in a study that they la¬

beled their growing and successful efforts “knowledge management” or

the “transfer of best practices.” Our lessons and enduring principles

were detailed in final reports (APQC, 1996; APQC,1997).

The following principles will sound familiar because they resonate

throughout this book. They are intertwined in the case studies and

case examples; they echo the conclusions of many of the book’s parts.

Here, in a nutshell, are principles that we believe, for this moment,

to hold true.

1. BUSINESS VALUES DRIVE TRANSFER BENEFITS

Where KM and transfer worked, they worked because management

aimed efforts at a clear set of value propositions: improving customer-

related practices, speeding up product (including service) development,

223