Solve Business questions
Strategy Dynamics Essentials
Kim Warren
Strategy D ynam
ics Essentials - Kim
W arren
While many businesses may be well-managed, poor strategy choices and implementation lead to a perpetual, grinding under-achievement of potential, ill-advised initiatives, or avoidable failures. In other fields of human endeavor, we reduce the risk of serious failure by building models of what we want to do before trying it for real, and codifying how things are supposed to work. Learning from what we do, we revise the models and update those codified processes to improve performance further. But for the most important function of all—figuring out what the enterprise could achieve and how that might be done—most organizations still rely on qualitative judgement and superficial analysis. This is unacceptable.
The strategy dynamics method— in essence, the application of engineering control theory principles to enterprise systems—solves this problem. Strategy Dynamics Essentials explains the frameworks that accomplish this aim and shows how they are applied. The method can be used to build working, quantified models of any enterprise, or any part thereof, of any scale, in any sector—or of any issue that an enterprise may face. The book is written for executives responsible for any aspect of business performance, consultants and other advisors, business students at all levels, and business teachers. No advanced technical skills are needed—just the will and ability to think quantitatively about whatever enterprise or function you are concerned about. Even the earliest principles can be applied right from the start, and this fast ROI can be repeated because key structures can be usefully deployed to tackle strategic challenges in marketing, staffing, and other functions—which is less difficult and costly than trying to assemble coherent strategy and plans from current alternative approaches.
The book is supported by a full, Masters' level on-line course, “serious games” for training and education, and the powerful and easy-to-use Sysdea software for modelling strategy and performance.
In clu
de s F
ig ur
es
in C
ol or
STRATEGY DYNAMICS ESSENTIALS
STRATEGY DYNAMICS ESSENTIALS
Kim Warren
Strategy Dynamics Ltd
Copyright © 2010 Kim Warren
First edition published in electronic format only (PDF and Kindle)
Second Edition, print & Kindle, published 2015
Published by Strategy Dynamics Limited, Two Farthings, Aylesbury Rd Princes Risborough HP27 0JS, United Kingdom ` www.strategydynamics.com
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 without the written permission of the publisher. Requests for permission should be addressed by email to sales@strategydynamics.com.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners.
This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. Its contents do not represent recommendations for any specific circumstance, for which the advice of a competent professional should be sought.
This book contains links to external websites, which, although checked, the continued existence of these are outside the control of the author.
If you find errors in this book please let us know so that we can update the text: contact@strategydynamics.com.
Print versions ISBN:
Grayscale interior: ISBN-13 978-1505809053 ISBN-10 1505809053 Color interior: ISBN-13: 978-1512107753 ISBN-10: 1512107751
Cover image ©macrovector -Fotolia.com
To all the strategists, managers, educators and students who, with determination and open minds,
have embraced the strategy dynamics method and encouraged me to continue this work
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CONTENTS
A A 10 P 11
The method 12 Who is this book for? 12 What prior knowledge do you need? 13 What "return on investment" can you expect? 13 Other resources 14 Acknowledgements 14
C 1 - B P O T 17 1.1 A Life-Cycle Example: Blockbuster® Inc. 18 1.2 Strategic Management: Positioning versus Delivery 21 1.3 What “Performance” Do We Want To Improve? 25 1.4 Building Future Performance 27 1.5 Nonfinancial Performance Objectives 31 1.6 Levels of Strategy 32 1.7 Example: Low-fare Airline Ryanair 33
C 2 - H R D P 35 2.1 Strategy Methods Focusing on External Factors 36 2.2 Strategy Methods Focusing on Firm-specific Factors 38 2.3 Limitations of Common Strategy Approaches 40 2.4 Tangible Resources and Profits 41 2.5 From Performance to Resources 43 2.6 Resources and Nonfinancial Performance 47 2.7 “Stocks” of Resources 48 2.8 When Resources Themselves Are the Objective 51 2.9 Specifying and Quantifying Resources 52
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C 3 - R W L 55 3.1 Quantifying the “Bathtub Behavior” of Resources 56 3.2 Accumulation over Time 59 3.3 Consequences of Resource Accumulation 61 3.4 Resources Won and Lost by Ryanair 62 3.5 “Accounting” for Resources 64 3.6 Control over the Building and Retaining of Resources 65 3.7 Generic Behaviors and Their Drivers 66
C 4 - I S A 71
4.1 Competition and Other External Factors 71 4.2 Existing Resources Drive Gains and Losses 72 4.3 How Resources Drive Their Own Growth and Loss 74 4.4 The Role of Potential Resources 76 4.5 The Strategic Architecture 77 4.6 Functional Issues and Other Objectives 83
C 5 - R Q 87 5.1 Size Is Not the Same as Quality 89 5.2 Attributes of Other Resources 90 5.3 When Resources Bring Access to Others 93 5.4 Using the Quality Curve to Beat Competitors 95 5.5 Other Uses for the Quality Curve and Resource Attributes 98
C 6 - D R 99 6.1 Developing Staff 99 6.2 The Customer Choice Pipeline 101 6.3 Product Development 105 6.4 Deteriorating Resources 106 6.5 How Resources Develop in Noncommercial Cases 108 6.6 Boundaries of the Firm 109
C 7 - C R 111 7.1 Type-1 Rivalry 112 7.2 Type-2 Rivalry 114 7.3 Type-3 Rivalry 116 7.4 Further Issues with the Three Types of Rivalry 118 7.5 Competing with Intermediaries 119 7.6 Competing for Other Resources 121 7.7 Rivalry in Noncommercial Cases 123 7.8 Dealing with Multiple Competitors 124
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C 8 - S S P 127 8.1 The Difference between Good and Poor Strategies 128 8.2 Steering Strategy and Performance 131 8.3 Policy to Guide Decisions 133 8.4 Controlling Indirect Decisions and Interference 136 8.5 Conflicting Objectives 137 8.6 Goals and Policy in Noncommercial Cases 138
C 9 - I R 141 9.1 State-of-Mind Intangibles 143 9.2 Information-based Intangible Resources 147 9.3 Quality-based Intangibles 150
C 10 - C 155 10.1 Dimensions of Capability 156 10.2 Learning Develops Capability 160 10.3 Capabilities Not Linked to Resource-building 161 10.4 The Balanced Scorecard 163 10.5 Capabilities in Public Sector and Voluntary Organizations 164
C F S 167 Further resources 167
A 1: T S 168 A 2: P C 170
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
D
E
10
A
A
ABOUT THE AUTHOR
Kim is an independent strategy writer and teacher, and a developer of sophisticated dynamic enterprise models and "serious games" for MBA and executive education. After 15 years in senior corporate strategy roles, Kim joined the faculty at London Business School, teaching on MBA and Executive programs, where he was introduced to the long-established and rigorous discipline of system dynamics. Realizing serious limitations with the conventional strategy methods offered in business schools, he developed the powerful strategy dynamics frameworks, by translating system dynamics into simple managerial language. He has devoted 20 years of effort to codifying and communicating this practical, rigorous method, which enables organizations of any kind or size to make radical improvements to their planning and implementation of strategy. In addition to developing learning materials and publications, he continues to develop and extend the method's power and relevance by helping businesses across all sectors to deal with major strategic challenges. Most recently, the method has also proved invaluable for dealing with issues in the public sector and in non-profit organizations. He is author of the prize-winning Competitive Strategy Dynamics (Wiley, 2002), and a major strategy textbook Strategic Management Dynamics (Wiley, 2008). His work was recognized with the 2005 Jay W Forrester Award as the most important contribution to the field of System Dynamics in the previous five years, and he served as 2013 President of the International System Dynamics Society.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
D
E
11
P
PREFACE
While many of our largest corporations, as well as lesser-known enterprises, are extremely well-managed by skilled, experienced and dedicated leaders, those executives are ill-served in the critical task of developing and implementing strategy by crude and unreliable methods that are not fit-for-purpose. The price for this inadequacy is a heavy one. Poor strategy choices and implementation lead to entirely avoidable business failures, ill-advised initiatives that have to be abandoned, and a perpetual, grinding under-achievement of potential even for those organizations that do not succumb to failure. In the grand scheme of things, these failures sum up to the endless repetition of boom-and-bust cycles that afflict most industries and the economic recessions that cause government and society great hardship and cost.
In other fields of human endeavor, we have reduced the risk of serious failure with two related approaches. First, we build models—at one time, physical models; more often today, software models—of things we want to try, before creating the real thing, whether that is a building, an aircraft, or a drug. Secondly, we codify how things are supposed to work, to ensure reliable delivery of whatever it is we are trying to do. Since we learn from what we do, we revise the models and update the processes we have codified to improve performance further.
Both approaches are widespread in many fields of management. Manufacturing companies build models of production facilities, distributors model their supply- chains, retailers model the likely catchment of new stores, and organizations of all kinds, of course, model their financial prospects. And businesses operate reliably because they codify what they do, right from the most detailed operational details that you will find, for example, in franchise systems, up to the procedures
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
D
E
12
P
for undertaking, repeatedly, successful product-launches, market entry, and acquisitions of other companies.
Yet for the most important function of all—figuring out what we want an enterprise to achieve and how that might be done—most organizations still rely on qualitative judgement supported (at best!) by simplistic frameworks and superficial analysis. This is unacceptable, not only to investors, but to employees, customers and society at large.
The method
This book summarizes the strategy dynamics method for developing and implementing strategy. The method is made possible by deploying the rigorous, scientific method of system dynamics – well-established since the 1960s – to the task of strategic management. In essence, system dynamics is the application of engineering control theory principles to social systems, and since all enterprise are "designed" systems, those principles are directly applicable to their design and management. The challenge is to make those principles clearly understandable in terms that make it possible for executives and analysts to use them reliably in practice.
The end-result of this method is the creation of working, quantified models of any enterprise, or any part thereof, of any scale, in any sector—or of any issue that such an enterprise may face. But to get to that result requires, first, that we establish the fundamentals of what an enterprise is—the things that make it up, such as customers, staff, products and capacity—and the mechanisms by which those elements actually function as an integrated system, both to generate outcomes we want (such as sales and profits), and to enable the further growth and development of that same system.
This book therefore focuses on defining the elements and mechanisms we need for this task in clear, every-day language, and showing how these elements and mechanisms can be assembled to create those working, quantified models we need for developing and managing strategy.
Who is this book for?
The book is designed to help four main groups:
� Executives who have some responsibility, either alone or as part of a team, for the performance of an organization, a business unit or a function.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
D
E
13
P
� Consultants and others who advise organizations on how to improve performance, exploit opportunities effectively, or deal with substantial challenges.
� Business students, at Undergraduate, Master’s and PhD levels.
� Business teachers, especially those teaching topics related to strategy.
Chapter 1 clarifies that “strategy” is a task for all levels of management, in all parts of an organization—whether a profit-seeking business, or a Governmental or non-profit entity—and explains why that task needs a rigorous, time-focused approach. Chapters 2 to 4 set out the core principles of the strategy dynamics method and how to apply them. Chapters 5-10 add important further frameworks that are useful on their own, or as a part of a whole-enterprise model.
What prior knowledge do you need?
Fear not! You do not need advanced technical skills to learn and benefit from the strategy dynamics method. A good general education is all that is required, but you must be willing and able to think quantitatively about whatever enterprise or function you may be involved in. This does not imply advanced mathematical or statistical skills. Anyone capable of building simple spread-sheet models will be able to grasp the strategy dynamics principles. Even those who are not so comfortable with arithmetical analysis can still exploit the method by working with colleagues or support staff who do have that capability.
What "return on investment" can you expect?
Even the earliest principles in this book can be applied right from the start. So very little time (and almost no cost) is required before you can start to get value from the method. This fast ROI can be repeated because components of the method can be usefully deployed alone, to tackle strategic challenges in marketing, staffing, and other functions. Of course, to apply the method to modeling the whole of a complex enterprise will require more effort and practice, in conjunction with others who are building the same understanding. However, this is less difficult and costly than the alternative of trying to assemble a coherent strategy and plans from the plethora of isolated and unreliable methods and largely financial analysis that typically dominate organizations' attempts at strategy and business planning.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
D
E
14
P
Other resources
This book has been written to provide busy readers with the greatest amount of useful understanding in the shortest possible time. But the principles it describes are best-understood by seeing them in use, so extensive online materials are available to support this book:
� An on-line course, that can be taken in summary form or in full, and that may be taken in three separate parts. The course includes many easy-to-use working models, including most of the examples in this book. See http://sdl.re/sdcourse.1 (The course and models are free for registered teachers).
� “Serious games” for training and education, http://sdl,re/microworlds.
� Powerful and easy-to-use Sysdea software for mapping and modelling strategy and performance, http://sysdea.com. The software’s Help system includes many worked examples—see http://docs.sysdea.com.
� A free fortnightly email briefing-note series at http://sdl.re/briefings.
The book deals with strategy for individual businesses or business units, and functions or departments. The principles and frameworks also apply directly to public service, voluntary and other not-for-profit organizations. Extensions to cover additional issues of corporate strategy in multi-business firms are planned for the future.
Acknowledgements
The list of people whose profound knowledge, experience and generosity of spirit inspired the decade of effort that has gone into developing the ideas in this book and the materials that support it, is a long one. It includes both outstanding academics and exceptional practitioners—and many people who are both!
I must start by thanking the person who first introduced me to the concepts of system dynamics, and who transformed my understanding of business and strategy; John Morecroft of London Business School. And none of this would have been possible without the rock-solid practical principles of the system dynamics method developed by Jay Forrester of MIT Sloan School. Other outstanding thinkers and teachers helped, inspired and encouraged my own learning, including John Sterman (MIT/Sloan), David Lane (Henley Business
1 The main source for all materials is http://strategydynamics.com. http://sdl.re is a shorthand redirect address for the main site.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
D
E
15
P
School) and George Richardson (University of Albany). Many other academics not only contributed to this learning but also showed how the knowledge could be practically applied in challenging real-world business situations, including Peter Milling (University of Mannheim), Markus Schwaninger (University of St Gallen), Carmine Bianchi (University of Palermo), Bob Cavana (University of Wellington), Jac Vennix and Etiënne Rouwette (Radboud University, Nijmegen) and Jürgen Strohhecker (Frankfurt School of Finance).
A large number of practitioners continue to help organizations of all kinds and sizes deal with large and complex challenges that no other concepts can help with. I have been fortunate to work with just a few of these outstanding professionals, including David Exelby, Steve Curram and Siôn Cave (DAS Ltd), Lars Finskud (Vanguard Strategy) and Maurice Glucksman (McKinsey & Co). Hundreds of students and executives have also contributed more than they will ever know. Every class and every project-case they worked on has added to and consolidated the ideas in this book and provided the extraordinary range of real-world examples in this book and its related materials. Few, though, has done more in this regard than Justin Lyon (Simudyne Ltd).
The production and delivery of the wide range of materials now available, and the support for the many teachers who use them, has been entirely due to the tireless efforts over nearly 20 years of my dear wife, Christina Spencer. Finally, none of us could use—or teach others to use—strategy dynamics as easily, quickly and reliably as is now possible without the brilliant Sysdea browser-based software, developed by Chris Spencer.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
D
E
H
R
D
P
16
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 1 BUILDING PERFORMANCE
OVER TIME
Although definitions of strategy can get very detailed and sophisticated, a concise statement will suffice for now:
An organization’s strategy is how it tries to reach its objectives.
This definition implies that the management of strategy consists of the following tasks:
1 Choosing objectives for the organization. Objectives may be financial, such as growth in cash flow, or non-financial, such as reaching a target number of customers by a certain date. Objectives can also evolve as conditions change.
2 Positioning the organization relative to other organizations. In business cases, this involves deciding which customers to serve, which products and/or services to provide, and how this will be done, usually in comparison to positions chosen by competitors.1 The choice of position implies that certain resources and capabilities will be needed. Nonbusiness organizations also choose positions, deciding which services to offer, for example, and to which beneficiaries.
3 Steering the organization’s progress over time. Having decided on a position the organization believes will be successful, management has the continuing challenge of developing effective policies and making
1 Constantinos C. Markides, All the Right Moves. (Boston: Harvard Business School Press, 2000), 27–112.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
18
good decisions in order to build the resources and capabilities it needs and to steer its strategy and performance.
These three elements of strategic management are most clearly visible in the case of a business trying to make profits for investors, but strategy is also important in public service, voluntary, and not-for-profit organizations.
1.1 A Life-Cycle Example: Blockbuster® Inc.
To see how these aspects of strategic management develop and adapt, consider an organization that has lived through its entire life, from birth, through growth and maturity, to decline and death - the video-rental business Blockbuster Inc.
Videocassette recorders (VCRs) were the first devices to allow consumers to both time-shift live TV and to watch movies at home. By the mid-1980s, 30 percent of US households owned a VCR, a fraction that was rapidly increasing. This was before DVDs and Blu-ray, before Internet services such as Netflix enabled movie rental from home (www.netflix.com), and long before it was possible to download movies or TV shows online. Figure 1.1 shows the entire history of Blockbuster’s store numbers, revenues, and operating profits, from its start in 1985 to 2008.
Blockbuster’s history can be divided into four main phases:
1 Start-up. In 1985 entrepreneur David Cook saw an opportunity to rent videos from stores in residential neighborhoods to consumers with VCRs. The stores were large and offered a wider range of movies than other rental outlets, which were mostly, small independently-run businesses. IT systems based on a customer membership card enabled control and ensured the most popular movies were available.
2 Growth. After just two years, Blockbuster was bought by another successful entrepreneur, Wayne Huizenga, who set out to dominate the market by opening hundreds of new stores each year. Eight years of rapid growth gave Blockbuster strong buying power with movie distributors, and hence low costs and early access to new titles. These factors made the company highly profitable, generating over 30% return on sales by 1995. Three big initiatives accelerated the company’s growth:
○ A franchising scheme allowed independent operators to invest their own capital in new stores, using Blockbuster’s systems and marketing in return for a share of revenue.
L
- C
E
: B
® I
.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
19
○ Several acquisitions of other significant chains extended the company’s market reach and buying power.
○ International expansion took place through acquisitions in the United Kingdom, Australia, Japan, and other countries.
3 Maturity. After being acquired by Viacom in the mid-1990s, Blockbuster’s profitability fell significantly, due to a combination of turbulent market conditions, loss of focus on the management of operational details, and a resulting high turnover of CEOs. Still, the basic business remained attractive, boosted by a new product range— software titles for videogame consoles, such as Nintendo, PlayStation, and Xbox. Thanks to video game rentals, store numbers and revenues continued to grow, if rather more slowly. The company also had to respond to the new service provided by Netflix, which enabled consumers to use the Internet to rent DVDs from home and return them by post.
4 Decline. By 2005 the first signs of the final phase of strategy’s life cycle—decline and ultimate closure—were evident. Netflix had become a major force in the marketplace, making 2007 revenue of $1.2bn, and similar services from Amazon.com and others led to intense price competition, driving down profit margins. Blockbuster’s vast store network, once a source of its dominance, became a large
0
2,000
4,000
6,000
8,000
10,000
1985 1990 1995 2000 2005
Stores at year-end
0
2,000
4,000
6,000
8,000
1985 1990 1995 2000 2005
Revenue ($millions)
0
200
400
600
800
1000
1985 1990 1995 2000 2005
Profit * ($millions)
Figure 1.1: Blockbuster Inc.’s Performance History
0
2,000
4,000
6,000
8,000
10,000
1985 1990 1995 2000 2005
Stores at year-end
0
2,000
4,000
6,000
8,000
1985 1990 1995 2000 2005
Revenue ($millions)
0
200
400
600
800
1000
1985 1990 1995 2000 2005
Profit * ($millions)
* Profit before interest, tax, depreciation, and exceptional items
Source: Company reports
L
- C
E
: B
® I
.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
20
and costly liability. To these pressures would soon be added the threat from video and TV download services as wider availability of fast broadband services made this new distribution mechanism a practical alternative. In November 2013, the business finally closed the high street stores.
So how did the three key elements of strategic management change over Blockbuster’s life?
� The most remarkable feature is that the basic positioning of the business remained largely unchanged for over 20 years—providing a wide range of family-oriented movies through neighborhood stores, using a membership card system to ensure control and availability of popular movies. The addition of gaming software merely built upon that basic position. Nor did the resources and capabilities required to succeed in that position alter—the stores, the brand, the types of customers and staff, the systems, and the types of product remained essentially unaltered in nature, although of course they changes considerably in scale.
� Objectives clearly evolved as the market developed, and competitive conditions and business performance changed. Cook’s aims during the first year or two were to get Blockbuster established and start generating profits. Huizenga raised the ambition considerably, aiming to open hundreds of stores each year, and drive rapid revenue growth, as evidenced by the compound annual growth rate (CAGR) averaging nearly 50 percent per year from 1987 to 1995. Growth in scale was key to Huizenga’s aim to grow profits, which persuaded Viacom to pay
years
Profit
start-up growth maturity
choose the initial strategic position
add to the strategic position
extend the strategic position
decline if no change
change the strategic position
Figure 1.2: Generalized strategy life cycle, illustrating extensions and strategic-change events
years
Profit
start-up growth maturity
choose the initial strategic position
add to the strategic position
extend the strategic position
decline if no change
change the strategic position
L
- C
E
: B
® I
.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
21
handsomely for the business in 1995. From then on, choice of objectives became less easy. The previously automatic link from store openings to revenue growth was broken, and management had to balance demands for delivering current-year profit targets with the need to invest in growth.
� Steering the strategy from year to year dominated the strategic management of the company throughout its life. This required continual decision-making on the range of movies to offer, price levels, staff hiring and training, marketing spending and message, and—as is critical as for any retailer—the rate, location, size, and design of new store openings.
1.2 Strategic Management: Positioning versus Delivery
One observation stands out from the Blockbuster story:
Changing a strategic position is a very rare event.
It is not surprising that organizations rarely change their choice of who to serve, with what, and how to do it. If you find a position that works today, something quite substantial must change for it to stop working tomorrow. If your positioning does not work, then you need to find one that does, quickly, before financial losses put you out of business. We certainly do not observe businesses—or indeed voluntary or public service organizations—constantly shifting their positions, despite what appears in business papers and journals about strategic innovation and transformation. (Note that strategic innovation and transformation are not to be confused with product innovation).
Most well-known businesses demonstrate great longevity of a successful positioning. Low-fare airlines such as Southwest in the US and Ryanair in Europe have maintained a strategic position unchanged for decades, as have IKEA (furniture retailing), BMW (automotive), and Mittal Steel (steel production). The resources and capabilities these firms require have also remained largely unchanged over decades.
This stability of strategic position is not limited to old technology. In the internet era, firms such as Amazon.com, eBay, and Expedia quickly found and exploited a position that worked. They may have made small adjustments, such as Amazon’s addition of affiliates into its business. They may also have added layer upon layer of extensions to their basic proposition—for example, Amazon’s expansion into CDs, DVDs, and other valuable, nonperishable products. Nevertheless, what these
S
M
: P
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
22
and most successful businesses actually do is find a successful position, and then relentlessly drive it forward. They do not keep changing their minds.
This is not to say that the positioning question never recurs. A strategic position may need to change for either positive or negative reasons.
Extending into new positions
Businesses often see new opportunities, whether to serve new customers, to provide new products or services, or to deliver their goods and services in a new way. Some such extensions of position make a big difference, as in the case of Amazon’s Marketplace offering, which allows many other suppliers to offer their products alongside Amazon’s own. More often, strategy extensions are small, compared with the existing business. Europe’s easyJet has added more low-price business models in car rental, cruise holidays, bus travel, and hotels, but all are trivial compared to its vast airline business.
Extending into new positions can also be done by acquisition. In 2005 eBay acquired Skype to strengthen its reach in the global marketplace and its payments platform. Cisco Systems takes this mechanism much further, having acquired over 100 companies1 in its strategy development. Not all such moves succeed— Cisco has benefited strongly over many years from this stream of acquisitions, while eBay’s purchase of Skype ended in break-up in 2009.
Avoiding threats to an existing position
Blockbuster suffered from two big threats—the emergence of postal-service substitutes to its store-based service and the increasing viability and availability of online streamed delivery of movies and other media content. Management has no choice in such cases but to review its established position and look for ways to adapt or add to their original position, or even to move to a new strategic position. Advances in technology are not the only source of threat. Powerful new competitors can also force a rethink, withdrawal, or effort to replicate the challenger. Established airlines, such as British Airways and American Airlines, started up low-fare operations in response to the burgeoning growth of Southwest, Ryanair, and the like. Others simply withdrew from the short-haul routes these innovators threatened.
Occasionally, the threat may be so serious that the position ceases to be viable and is abandoned. This has been the fate of many store-based travel agencies and music stores, as it was for Blockbuster. In cases like these, investors will be best
1 http://en.wikipedia.org/wiki/List_of_acquisitions_by_Cisco_Systems
S
M
: P
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
23
served by returning to them whatever cash can be generated during a gradual winding-down of the store’s business.
Although switching to an entirely new strategic position is very rare, it is not completely unknown. Cell phone manufacturer Nokia for example, was founded in 1865 as a paper producer before moving into rubber products around 1900, and then cable manufacturing and a host of electrical products. Nokia’s first electronics department started in 1960, leading to its move into telecommunications, which only really started in the 1970s.
The Blockbuster story offers a general model for the strategy life cycle. Figure 1.2 illustrates the four broad phases of the strategy life cycle, starting with the initial choice of strategic position, a period that is often unprofitable. The chart also shows the impact of initiatives to add to the original strategic positioning, both by acquisition and extension, and a change of that position when it becomes unsustainable.
Given the extreme rarity of changes to strategic position, then, what does “strategic management” actually do throughout the long intervening periods?
Most of management’s strategy work consists of delivering the strategy, also known as implementation or execution.
Taking big initiatives, such as acquisitions or entering new markets, is sometimes termed strategic decision-making, but seemingly small decisions can also have serious impacts on future performance. Hiring too few skilled staff, for example, can hold back business performance for years afterwards.
Actions, choices and decisions by an organization, and changes in external conditions are “strategic” if they significantly affect medium- to long-term
performance.
Making skilled choices that will steer strategy well from period to period is therefore critical, but is often overlooked or even dismissed as mere “operational effectiveness”.1 However, companies can massively outperform competitors on this basis, rather than by choice of strategic position. Blockbuster delivered revenue and cash-flow growth that was orders of magnitude larger than many of their nearly identical rivals. The same is true of the dominant low-fare airlines who have delivered performance that is orders of magnitude greater than weaker competitors who chose precisely the same strategic position, let alone the large number who have failed and gone out of business. Continuous strategic decision-
1 Michael E. Porter, What Is Strategy? Harvard Business Review, 74(6), 61–78.
S
M
: P
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
24
making is equally critical in public service, voluntary, and not-for-profit organizations.
The observation that strategically well-managed firms can radically outperform competitors has another important implication: performance prospects are not dominated by external conditions.
In most cases, management has considerable scope to drive strong performance.
Strategic issues are not exclusively long-term in nature, and the annual planning processes common to many companies are not adequate. Two considerations considerably reduce the timeframe over which both overall strategy and specific issues or initiatives should be examined:
1 In some industries, conditions change very quickly. Consider how often new cell phone handsets are launched or how quickly online social networking services develop, and you will see that substantial changes can occur from quarter to quarter or even month to month. Strategic plans that offer only annual forecasts cannot be adequate in such cases—it will be far too late to adjust strategy if you have to wait twelve months before knowing how well it is working!
2 A focus on short timeframes is also needed when events occur that will have long-term consequences. The success of a new product is often determined within a few months of its launch, so its impact on overall performance needs to be tracked continually, not just in an annual review. Such initiatives certainly have significant implications for long-term performance and may also need the coordinated efforts of several functional groups.
The start-up of a new rival is another event that may need a short-term focus. Such threats require quicker responses and adaptation to strategy than an annual perspective can handle. In one case, a division of GlaxoSmithKline saw its dominant position in the market for travel vaccines under attack by a rival product launch that would succeed or fail over just ten weeks, fundamentally threatening the business’s entire future. There was no value in annual plans here—the episode was unforeseen when the year’s budget was put together and was over before the next planning round started!
S
M
: P
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
25
1.3 What “Performance” Do We Want To Improve?
Strategy textbooks, being largely devoted to commercial businesses, generally focus on some indicator of financial performance, so “profit” is a sensible place to start. Two elements of profit must be distinguished. First is the normal profit that investors expect for the use of their capital, given the level of risk. This leaves a second element, the “economic profit,” which is the surplus that remains after the costs of all inputs (including capital) have been paid out. Economic profit, or the closely related “economic value added” (EVA), has been widely adopted as a management tool by many large corporations.1
Another performance measure that commonly gets attention is the profit return on invested capital (ROIC) that a company generates. Different firms in an industry typically exhibit a range of ROIC—a few suffer losses, others make a modest return, and a few are exceptionally profitable. In Figure 1.3, Company X and a few others are in this happy position and can be expected to achieve a superior return as a result of their “competitive advantage.” This advantage is said to be “sustainable” if we can continue making this superior return year after year.
The distribution of ROIC varies widely among industries. Some industries feature a large number of firms, many of which are very small, while others are more concentrated with fewer, larger firms. Both the average profitability level and the variance of that profitability also differ in a characteristic manner among industries. Cell phone service providers, for example, are few in number and struggle to generate strong returns on the costly assets of their networks. Pharmaceutical companies, in contrast, are still relatively numerous, in spite of
the many mergers that have taken place in the industry, and generate higher returns. Extremely diffuse and segmented markets, such as a city’s restaurants, may feature hundreds of competitors and a very wide range of profitability.
1 Al Ehrbar. EVA: The Real Key to Creating Wealth. (Chichester: John Wiley & Sons, Inc., 1998).
0-10 10 20 30
Profitability Return on invested capital ROIC
Company X
Number and size of
companies
Figure 1.3: Illustrative profitability distribution among firms in an industry
0-10 10 20 30
Profitability Return on invested capital ROIC
Company X
Number and size of
companies
W
“
P
” D
W
W
T
I
?
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
26
Although a focus on profit levels and profitability ratios is understandable, it poses an obvious problem. We can nearly always boost profits now, by simple changes such as raising prices or cutting expenditure, but we will regret these actions if they damage future profits. In spite of this obvious conflict, it is common for management—under pressure from investment analysts and their own performance incentives—to focus on delivering short-term results, even if it damages prospects for the future. The banking crisis of 2008–2009 is an extreme example of this problem.
Superior ROIC is not a good indicator of competitive advantage or strong strategic management.
Investors’ interests are in fact best served by strong and sustained growth in the future stream of profits, rather than the profits being made right now. Even if investors sell their shares, the price they get will reflect what the buyers of those shares expect future profit growth to be. A company’s share price increases when it announces higher-than-expected profits, not because that profit is in itself worth a great deal, but because it suggests that the company can continue delivering increasing profits.
Investors don’t get their hands on all of a company’s profits. The actual money available is the cash flow generated by the company’s operations minus any cash needed to make the future growth of cash flow possible. Cash may also be needed for more working capital, such as inventories, to support a growing business. What investors should really want, then, is growth in free cash flow: 1
Free cash flow = profit, after interest and tax are deducted + depreciation − capital invested − increase in working capital
It is not even necessarily bad for a company to generate negative profits and cash flows for long periods, if the company is spending money now to exploit opportunities to make strong future cash flows. Figure 1.4 illustrates the profits and cash flow generated by Amazon.com from 1997 to 2008, both of which were negative for several years. If ROIC were the appropriate indicator of strategic success, we would have wanted to know why Amazon was so unsuccessful.
1 A clear explanation of why free cash flow is of critical importance to investors is in the letter to shareholders from the founder and CEO of Amazon.com, Jeff Bezos, on pages 3– 5 of the Company’s 2004 Annual Report.
W
“
P
” D
W
W
T
I
?
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
27
To put a value on a firm’s strategy, then, or to value an investment or other initiative, we need an estimate of the trajectory of future cash flows, not just the cash flows for a single period of time. But that value is not simply the sum of future free cash flows. People value cash that they will receive soon more highly than the cash they
may receive far in the future because it has alternative uses and because of the uncertainty of long-term results. Each period’s cash flow is therefore discounted by the firm’s cost of capital to arrive at its “present value.”1
1.4 Building Future Performance
The principles outlined to this point provide the main focus for the strategy dynamics approach:
Strategic management is about building and sustaining performance into the future.
This is not a new idea in strategy. We have known since the 1950s2 that superior profitability is neither interesting in itself, nor usually sustainable. A firm entering a new market will spend money to do so and will therefore reduce its profitability before making gains in the new market. Having successfully entered one new market, it might do so again and again, and could therefore persist in making lower returns than its stay-at-home rivals. Investors, though, would welcome its efforts because of the increasing free cash flows.
From this point on, we will refer only to operating profit or cash flow when discussing the financial performance of commercial firms.
Three distinct but related questions lie behind how organizations perform through time (see Figure 1.5):
1 Tom Copeland, Tim Koller, and Jack Murrin. Valuation—Measuring and Managing the Value of Companies, 4 ed. (Chichester: John Wiley & Sons, Inc., 2005).
2 Edith Penrose. The Theory of the Growth of the Firm, 4 ed. (Oxford: Oxford University Press, 1959), 78–91.
-1,000
-500
0
500
1,000
1,500 Cash flow
Operating profit ($millions
Figure 1.4: Amazon.com operating profits and cash flow, 1997–2008
-1,000
-500
0
500
1,000
1,500 Cash flow
Operating profit ($millions
Source: Company reports
B
F
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
28
� Why has past performance followed the time path that it has?
� Where will future performance go if we carry on as we are?
� How can we improve future performance?
While the question of past performance won’t be applicable to new ventures just starting out, in most cases, it is critical to address all three questions in order to understand an organization’s past, present, and future performance. Figure 1.6, illustrating the performance of coffee-store chain, Starbucks, from 2002 projected to 2012, shows why these three questions are important.
1 Why has past performance followed the time-path that it has? The growth in profits up to 2007 resulted from a rapidly expanding network of stores, which involved entering the markets of new countries, built on an attractive profit model for each location. This expansion was made possible by strong sales to a reliable customer base, supported by excellent service from well-trained, loyal staff. 2008 profits suffered from both difficult market conditions and from some poor strategic decisions made earlier, notably the raising of prices and over-expansion between 2002 and 2007. (The profits for 2008 were actually lower than shown, at $504 million, but included a large provision for the closure of stores).
2 Where will future performance go if we continue as we are? If Starbucks had failed to act in 2008, profits in the following year would likely have been lower than they were (dashed blue line). Continued loss of customers and reduced spending by those who
Where?
time
Performance
now
Why?
How?
Figure 1.5: General questions of performance for strategic management to address
Where?
time
Performance
now
Why?
How?
0
500
1,000
1,500
2002 2004 2006 2008 2010 2012
history
preferred
feared
$ millions
Figure 1.6: Starbucks’ profit history to 2008 and alternative futures
0
500
1,000
1,500
2002 2004 2006 2008 2010 2012
history
preferred
feared
$ millions
B
F
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
29
remained would have further depressed sales revenue. The stores’ fixed costs would have caused even a small fall in revenue to drive a heavy fall in profits. Further damage could then have set in. Customer loss would have undermined Starbucks’ image as “the” place to go for a great coffee experience, leading to still further loss of sales. Staff morale could also have fallen, worsening the service experience, increasing staff turnover, and raising the costs of hiring and training. In this “feared” future, problems would have multiplied, and profits would have declined steeply.
3 How can we improve future performance? Figure 1.6 shows a “preferred” future (dashed green line), in which profits recover. Part of this “how” was already in progress by the end of 2008, with cuts to overhead costs. Unprofitable stores were closed in 2008 and 2009, and it may be necessary to reduce prices to support sales volume, which would hit margins. If the cost cuts do not damage service quality, store closures effectively cut out loss-making branches, and consumer spending recovers, the company may see profits grow again from 2010. It could also continue opening branches in still-promising locations. Taken together, these outcomes could lead to Starbucks’ preferred future profits.
Starbucks’ story clearly shows that history is of more than academic interest. This company, like all organizations, is moving along a path through time whose trajectory is already strongly determined by policies, decisions, and other events from the past. Starbucks would not, for example, have had hundreds of loss- making stores in 2008 if it had not previously opened them!
To appreciate the implications of history’s role in driving strategic performance, imagine that Starbucks made exactly the same operating profits in 2008 as it
actually did, but had reached this point by a quite different path. The answers to our three critical questions would be entirely different (Figure 1.7).
1 Why has past performance followed the time-path that it has? The company could have opened fewer, but high- performing stores through 2002–2007. It might have raised
0
500
1,000
1,500
2002 2004 2006 2008 2010 2012
history
preferred
feared
alternative
$ millions
Figure 1.7: Starbucks—hypothetical profit history and alternative futures
0
500
1,000
1,500
2002 2004 2006 2008 2010 2012
history
preferred
feared
alternative
$ millions
B
F
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
30
prices less to maintain customers’ perception of good value and set higher staff levels to ensure good service. These choices would have held down profits until 2007, but benefits could have shown up in 2008. Profits would not have been depressed by loss-making stores, and the company would have saved the heavy exceptional costs of closing them (not included in Figure 1.6). With more attractive pricing and service, customer numbers and purchase frequency may have fallen less heavily. Profit growth may still have slowed, but may not have turned negative.
2 Where will future performance go if we carry on as we are? With this more healthy history, the likely future would have been entirely different. Instead of collapsing as more stores became unprofitable and customers left due to poor value and service, there could have been a less severe decline, one due only to reduced consumer spending (blue dashed line). This brighter future would still have come at a cost, however. Up until 2007, the company would have made nearly $700 million less profit but would have saved most of this amount by not investing in stores that would have failed. Starbucks would also be making back those lost profits twice over in the few years following 2008.
3 How can we improve future performance? The preferred future presented in Figure 1.7 is still more attractive when starting from Starbucks’ alternative (hypothetical) history. Sustained good-value pricing and service levels could allow store profits to pick up again in 2009, and enable continued store openings in under-served territories. Without the crisis of 2008 to 2010, the company could realistically aim for continued solid profit growth. (It turned out that Starbucks improved profits much more than shown, even in this preferred future, delivering profits of $1419 million in 2010).
The key message is that history matters—a lot! This has important implications for what you do today, because “today” will be “history” when viewed from the future, and past decisions will determine the path of the organization into that future.
B
F
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
31
1.5 Nonfinancial Performance Objectives and Not-for-profit Organizations
Management often sets targets for nonfinancial measures—customer growth, market share, staff numbers, and so on. Some even set aims for intangible measures, such as reputation. This does not mean they ignore financial performance. Rather, they focus on these other factors because they are believed to drive financial results. Investors and analysts also pay attention to such measures—airlines report passenger journeys sold, cell phone companies report on subscriber numbers, fast-moving consumer goods (FMCG) firms report market shares, and so on.
At Skype, the voice-over-Internet phone (VoIP) service, attention has focused on the growth in user numbers (Figure 1.8). Having users is not especially useful, however, unless they are making calls, so a secondary indicator is usage, measured as the number of call-minutes the business serves.
Nonfinancial performance aims are understandably common in public sector, voluntary, and nongovernmental organizations (NGOs). Governments try to decrease unemployment rates, police forces try to reduce crime rates, and charitable organizations try to reduce or eliminate suffering among target groups. For example, the World Health Organization’s Global Polio Eradication Initiative1 aims “… to ensure that no child will ever again know the crippling effects of polio.” As recently as 1980, there were thought to be approximately 400,000 cases worldwide. Here, the preferred and feared futures are reversed—the aim is
1 For more information on the Global Polio Eradication Initiative, see http://www.who.int/topics/poliomyelitis/en/.
0
200
400
600 User accounts millions
0
10
20
30 Call traffic billions of minutes
Figure 1.8: Growth in users and call tr
0
200
400
600 User accounts millions
0
10
20
30 Call traffic billions of minutes
affic for Skype’s VoIP service
Source: eBay Company reports
Note: Skype was owned by eBay Inc. between 2005 and 2009.
N
P
O
N
-
-
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
32
a decline in the performance indicator, preferably to zero (Figure 1.9).
Objectives to reduce, rather than to increase, something also arise in corporate settings—many businesses would prefer lower rates of staff turnover or shorter lead- times for product development, for example. Targets of zero are also common, such as for product failures or customer complaints.
1.6 Levels of Strategy
So far, we have focused on the overall performance of a self-contained organization. Strategic management, though, applies in other contexts, as well.
Within an organization, each function or department may need a strategy— marketing, research and development (R&D), human resources (HR), and so on. Such departments commonly have their own objectives, such as capturing customers or launching products. It is of course important for those aims to be consistent with those of other functions. Winning more customers is not helpful, for example, if HR does not aim to develop the staff numbers to serve them.
Also, some businesses are not independent, but are part of a multi-business corporation. In these cases, a business may also not be entirely self-contained, but share support functions such as finance or R&D with other businesses owned by the same group.
This means that strategic management is required at several levels (Figure 1.10). For clarity:
� the term “business” will refer to organizations with one main purpose, either a self-contained, independent business, or else a subsidiary, or “business unit” of a larger group
the term “corporation” will refer to organizations consisting of several such business units, and “corporate strategy” will refer to the particular strategic challenges that arise in that context. (Note that this essentials book will not cover these more complex corporate strategy issues.)
0
500
1000
1500
2000
2500
1995 2000 2005 2010 2015 2020 2025
history
preferred
feared previously
4-7,000 per year
Figure 1.9: History and alternative futures for the number of cases of poliomyelitis worldwide
0
500
1000
1500
2000
2500
1995 2000 2005 2010 2015 2020 2025
history
preferred
feared previously
4-7,000 per year
L
S
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
33
Public service and other not-for-profit entities also have functional departments needing strategic management, and large organizations may consist of several operating units with distinct purposes.
1.7 Example: Low-fare Airline Ryanair
This chapter has explained the main tasks of strategic management and the kinds of performance objectives that strategy is designed to deliver. It has also shown that strategic management is fundamentally a “dynamic” issue—improving performance continually over time. This is the challenge that the strategy dynamics method has been specifically designed to address.
To show the build-up of a strategy dynamic analysis, we will use Europe’s successful low-fare airline, Ryanair. This case has the advantage of being focused on a single business activity—providing low-cost air travel on short- to medium- haul routes in a single region
Being a commercial firm, Ryanair certainly has financial aims, and as a publicly quoted company, there is a focus on the profits it makes each year and each quarter. We will focus on its earnings before interest, tax, depreciation, and amortization (EBITDA). To increase its results, the company must grow revenues, but since higher pricing contradicts Ryanair’s strategic positioning, revenue growth must come from increasing sales of passenger-journeys.
The last item to address before setting out the quantitative history and aims of Ryanair is time scale. Since conditions are not changing too quickly, we will look at the company’s progress in annual terms, and over several years. Ryanair’s own management undoubtedly looks at performance more frequently—to make plans to open new routes, for example—but an annual perspective is adequate for our purpose.
Corporate strategy (the multi-business company)
Business-unit strategy
Business-unit strategy
Business-unit strategy
Business-unit strategy
Marketing strategy
Human-resource strategy
Manufacturing strategy
Information- systems strategy
other function strategies
Figure 1.10: Functional, business-unit, and corporate levels of strategy
Corporate strategy (the multi-business company)
Business-unit strategy
Business-unit strategy
Business-unit strategy
Business-unit strategy
Marketing strategy
Human-resource strategy
Manufacturing strategy
Information- systems strategy
other function strategies
E
:
L
-
A
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
B
P
O
T
34
Figure 1.11 shows a five-year history of sales and profits and shows plausible objectives for these two indicators—the preferred future—plus a less attractive feared outcome. Ryanair’s recent past includes two strong external impacts—high fuel prices during 2005–2007 and the severe recession of 2008–2009, which hit demand for air travel especially hard. The company coped well with both of these challenges, sustaining growth in demand and remaining profitable. Demand was only sustained during 2009, however, with an 8 percent fall in prices, which led to lower profits.
In both the preferred and feared futures, market demand is assumed to recover only after 2010. In the preferred outcome, demand recovers strongly, and Ryanair is able to capture new demand by continuing to open many new routes. Economic recovery also allows prices to recover and profit growth to resume. In the feared outcome, market demand recovers more slowly, prices still cannot recover, and the company is unable to find many new routes that will be profitable to serve.
The remaining chapters of this book will discuss the logic and key elements of strategy dynamics analysis by building on the strategic performance indicators illustrated in the Ryanair example1.
1 This example will be developed further in later chapters All charts and figures include data from the company’s Annual Reports and statutory returns, where reported, up to and including the financial year ended March 2009. Other data and projections to 2014 are illustrative estimates by the author
0
500
1,000
1,500
2005 2007 2009 2011 2013
history
preferred
feared
profit (EBITDA) € millions
0
20
40
60
80
100
2005 2007 2009 2011 2013
history
preferred
feared
passenger journeys sold millions
Figure 1.11: Five-year history and plausible objectives for low-fare airline Ryanair
0
500
1,000
1,500
2005 2007 2009 2011 2013
history
preferred
feared
profit (EBITDA) € millions
0
20
40
60
80
100
2005 2007 2009 2011 2013
history
preferred
feared
passenger journeys sold millions
E
:
L
-
A
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 2 HOW RESOURCES DRIVE
PERFORMANCE
Before we can understand how profits change over time, we need to clarify what explains profits at any point in time. Strategy research seeks explanations of two broad types:
1 How market features, competition, and the wider external environment constrain profitability. Typically, these factors are used to choose a strategic position that offers the potential to be sustainably more profitable than average.
2 How characteristics of any specific firm, such as its resources and capabilities relative to competitors, enable it to achieve and sustain superior profitability.
The following discussion gives only a short summary of common approaches in these categories. More detail on these and other methods can be found in popular strategy textbooks.1
SWOT analysis. While now largely viewed as inadequate, SWOT remains the first method most executives think of for developing strategy. SWOT is an acronym for the four main issues that the method considers. “SW” refers to the Strengths and Weaknesses of a business, relative to competitors, and to what is needed to succeed— i.e., internal considerations. “OT” refers to Opportunities and Threats that the organization might face—features of the external environment.
1 See for example: Robert M. Grant. Contemporary Strategy Analysis, 5th ed. (Oxford: Blackwell Publishing, 2005).
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
36
2.1 Strategy Methods Focusing on External Factors
Industry forces. The dominant model for assessing a firm’s business environment is the so-called “5-forces” framework.1 This model helps explain industry profitability and its variance (Figure 1.3), enabling management to find a strategic position where a sustainably high rate of profitability may be possible—specific customer-segments to serve, distinctive products or services to offer, or particular marketing channels to use, for example. The five “forces” making up this framework are:
1 Activities of competitors who, in trying to capture sales, compete away profitability through pricing, or by offering costly benefits to customers—lower fares or better service by competing airlines for example.
2 The potential for new entrants to start operating in the industry, if they think they can make good profits, such as the start-up of new airlines.
3 Availability and appeal of substitutes—i.e., goods and services that, while not the same as those sold by the company, offer the customer similar benefits. For example, trains are a substitute for many airline routes, and videoconferencing is a viable alternative for some business travel.
4 The buying power of customers to drive prices down by switching between competing suppliers. The ease with which customers can switch between alternative airlines and the incentive to save money on this high-cost purchase pushes down prices and profitability.
5 The power of suppliers to command high prices for critical inputs. Major airports control access to a large fraction of the travel market and so can command high prices from airlines who want to fly there.
In principle, the more intense the competitive pressures, the lower will be the likely average profitability of an industry and the less the scope for any firm to do better than this average. However, it is difficult to define and quantify each force sufficiently to arrive at a robust analytical prediction of profitability and its variance in any particular setting.
1 Michael E. Porter. Competitive Strategy. (New York: Free Press, 1980), 3–33.
S
M
F
E
F
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
37
PEST analysis Competitive forces are in turn affected by wider features of the external environment, commonly assessed under four categories:1
1 Political factors include influences arising from the actions, policies, and attitudes of governmental and nongovernmental entities. The influence may be explicit and direct, such as the 1980s legislation that opened up competition in the US airline industry or more generalized, such as trading agreements between countries.
2 Economic conditions have various effects on business prospects and hence on strategy and performance. General economic growth of course helps stimulate many markets, enabling the growth of companies’ sales and profits. Economic conditions may, however, have specific effects on particular industries, such as the severe drop in business air travel in 2009 as companies in other industries sought desperately to cut costs.
3 Social factors also influence how markets develop. The potential for any product or service may change in size due simply to population growth, age distribution, and other demographic effects. Other social mechanisms reflect behavioral factors, such as the increasing popularity of short-break vacations.
4 Technological developments cause two main effects. First, there is the improved functionality of products and services—the ever-increasing capability of cell phones and other electronic devices, for example. Secondly, technology can reduce unit costs. The extreme efficiency of web-based sales compared with the slow and costly alternative of using retail travel-agencies, for example, enabled new airlines to reach customers at very low cost.
Environmental concerns and Legislative impacts are sometimes added to this list, extending the acronym to PESTEL analysis. Once again, while a useful checklist, these elements cannot be defined and quantified sufficiently well to provide more than generalized conclusions regarding an industry’s attractiveness or future prospects.
1 V. K. Narayanan and Liam Fahey. “Macroenvironmental analysis: Understanding the environment outside the industry,” in The Portable MBA in Strategy, 2ⁿ ed., Liam Fahey and Robert M. Randall, eds. (Chichester: John Wiley & Sons, Inc., 2001), 189–214.
S
M
F
E
F
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
38
Scenario planning.1 This important strategy method looks at how the external environment may change. (Note that scenario planning should not be confused with “forecasting.”) Scenarios are plausible alternative stories of how an industry’s wider environment and competitive conditions might evolve into the future. Management then assesses how demand, competitive conditions, and other factors might change under these alternative versions of the future. These conclusions are then used to develop a strategy that can both exploit opportunities that may arise while at the same time being robust enough to account for any dangers that may threaten those futures. While commonly used among very large corporations, the principles of scenario planning are also valuable to smaller companies and not-for-profit organizations. Scenario planning is a vital component of strategic management but is frequently neglected—many firms may have saved themselves considerable pain had they carried out this planning prior to the recession of 2008–2009.
Competitor analysis.2 In addition to the above frameworks for thinking about the wider external environment, several techniques exist for understanding one’s competitors. Understanding how competitors’ strategies work, initiatives they may pursue, and how they might respond to one’s own actions is a further critical task of strategic management, but one that is rarely done well.
2.2 Strategy Methods Focusing on Firm-specific Factors
Strategy research has now largely concluded that industry conditions are less influential on firms' profitability than are the choices made by and the characteristics of firms themselves-you can do well in intensely competitive industries, and badly in more benign sectors. We have explained part of why this might be the case in Section 1.2. The following are some common frameworks for strategic assessment of the organization itself.
Value-chain analysis3 This approach looks at how a business adds value to the inputs it uses in order to create products and services for which it can charge more than the cost of those inputs and its operating costs. A restaurant, for example, has to pay for the food and drink it buys, and hopes to sell the resulting
1 Kees van der Heijden. Scenarios: The Art of Strategic Conversation. (Chichester: John Wiley & Sons, Inc., 1996).
2 Craig S. Fleisher and Babette E. Bensoussan. Strategic and Competitive Analysis: Methods and Techniques for Analyzing Business Competition. (Upper Saddle River, NJ: Prentice Hall, 2002).
3 Michael E. Porter. Competitive Advantage. (New York: Free Press, 1985), 33–52
S
M
F
F
-
F
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
39
meals for a higher price than the cost of the ingredients. That margin must also be sufficient to pay for the real estate the restaurant occupies-another bought-in input-and for staff, fuel, and other operating costs.
Internal costs are divided between "primary" costs-sourcing raw materials, converting them to finished products, delivering them to customers, marketing and selling to customers, and providing customers support-and "support" costs, such as finance and IT support. Different industries vary widely in the mix between these cost elements. Manufacturing companies may be dominated by the cost of converting raw materials to finished products, wholesalers feature a high proportion of logistics cost, and cosmetics companies spend heavily on marketing.
Value-chain analysis can be used to identify opportunities for competitive advantage in either of two ways-by eliminating costs or by seeking ways of raising the price customers are willing to pay. It is also possible to combine the approach with an analysis of customers' value chains in order to make savings or generate more value, for example, by offering products that may be higher in price but save customers more in other ways.
The value curve1 The value curve is a simple tool for finding a potentially strong competitive position for a business by understanding the relative importance of different customer benefits. This analysis can help identify new combinations that are more attractive to customers than those offered by existing firms. Low-fare airlines first succeeded, for example, by offering exceptionally low prices, and frequent point-to-point services, without offering many of the services of the full-fare airlines who had previously dominated the market. If a powerful and fundamentally new value-proposition can be found and implemented quickly, then it can be difficult for competitors to copy, leading to a sustainable competitive advantage and the profitability that this earns.
Resource and capability analysis2 Since we now know that a good strategic position alone is not enough to guarantee success, research has tried to identify
1 W. Chan Kim and Renée Mauborgne, Creating new market space, Harvard Business Review, 77(1), 83–93; W. Chan Kim and Renée Mauborgne. Blue Ocean Strategy: How to Create Uncontested Market Space and Make the Competition Irrelevant. (Boston: Harvard Business School Press, 2004).
2 Jay B. Barney. Gaining and Sustaining Competitive Advantage, 2ⁿ ed. (Upper Saddle River, NJ: Prentice Hall, 2001), 127–169.; David J. Collis and Cynthia A. Montgomery, Competing on resources: Strategy in the 1990s, Harvard Business Review, 73(4), 118–128.
S
M
F
F
-
F
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
40
the internal factors that may provide sustainable advantage. That effort has identified resources (things an organization possesses) and capabilities (activities it is good at doing) as two internal factors leading to sustainable advantage. Note that these factors are sometimes confusingly lumped together under the single term "resources."
To provide competitive advantage, it is argued, a resource should be valuable, rare, hard to imitate, and embedded in how the organization operates-the so-called VRIO criteria. Tangible factors, such as cash, capacity, and staff, are considered to be both obvious and easily copied or bought by rivals; therefore only intangible factors can provide sustainable advantage. Examples of intangible factors include a company's reputation and the strength of its customer relationships. These principles now form the core of the so-called "resource-based view" of strategy (RBV), a perspective that dominates current academic thinking.
Analysis of resources and capabilities, then, aims to: (1) specify the resources and capabilities that could provide advantage; (2) assess these resources and capabilities against what competitors possess; and (3) identify those resources and capabilities that need to be developed in order to win and keep winning. Unfortunately, like other strategy concepts, these factors are not easy to specify or quantify, making it hard to work out how much, and of what, needs to be done, or what impact any efforts may have on performance. Furthermore, there are some fundamental problems that current RBV thinking does not adequately address.
As has long been known, even tangible resources are difficult, costly, and time-consuming to accumulate, and so cannot be dismissed as irrelevant to strategy and performance. Blockbuster's stores, for example, were fundamental to its competitive advantage and market dominance for two decades. Furthermore, resources are not independent of each other-a great sales force cannot win customers without good products, and they cannot keep customers if service capacity is inadequate. It is the entire system with all its interdependencies that generates performance. As a result, no simple listing of resources and capabilities can be adequate, even if these could be specified and quantified
2.3 Limitations of Common Strategy Approaches
While useful to some degree, most of the principles and frameworks for strategy analysis suffer from three important limitations:
1 They offer guidance on choice of strategic position—who to serve, with what, and how. As explained in Section 1.1, such decisions are made
L
C
S
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
41
infrequently. Strategic management, being an ongoing task rather than an occasional challenge, needs tools for steering strategy and performance through time.
2 They are based on evidence for what might cause superior profitability ratios, while management and investors should be more concerned with improvements in cash flow over time and other absolute performance indicators, such as customer numbers and sales volume.
3 Most rely on abstract and ambiguous terms that are hard to specify or quantify, and therefore offer little confidence as to their impact on real performance.
2.4 Tangible Resources and Profits
Even if tangible resources are easy to copy, which is not always the case, they are clearly involved in explaining an organization’s performance, so we start by clarifying that connection. Figure 2.1 picks up the example of Ryanair from Chapter 1, illustrating the firm’s income statement for the financial year ended (y/e) March 2009 in causal form. Profits arise from its revenue minus costs, with revenue arising from passenger fares and ancillary items, such as ground transport and in-flight sales. Costs fall into major categories, such as staff, aircraft, operating airports and routes, and marketing.
2593.3 1402.1
309.3
286.6
443.3 139.1
12.8
2343.9
348.5
2942.0
598.1
Total operating costs
Fare revenue
Aircraft costs
Staff costs
Profit [EBITDA] €m p.a.
Total revenue
Ancillary revenue
Route costs
Airport costs Other costs
Marketing costs
Figure 2.1 Figure 2.1: Causal Structure of Ryanair Profits for Year Ended March 2009
2593.3 1402.1
309.3
286.6
443.3 139.1
12.8
2343.9
348.5
2942.0
598.1
Total operating costs
Fare revenue
Aircraft costs
Staff costs
Profit [EBITDA] €m p.a.
Total revenue
Ancillary revenue
Route costs
Airport costs Other costs
Marketing costs
Figure 2.1 T
R
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
42
The causal relationships here are simple:
Profits = total revenue − operating costs total revenue = fare revenue + ancillary revenue operating costs = aircraft costs + route costs + airport costs + staff costs + marketing costs + other costs
If this causal explanation for profits was true for y/e March 2009, it was also true for every previous year. It will also continue to be true in future, so long as the firm conducts the same business. We can therefore connect the time charts of the above-listed items in the same layout (see Figure 2.2). Each chart in this figure portrays the historic values for the items named and provides a plausible time path for how they might develop in future. Although this may be an unfamiliar view of a company’s income statement, it is simply illustrating in a graphical, causal layout the same data we normally see in spreadsheet form.
The features of this diagram and others that will follow are important. Each item includes a quantified scale on the vertical axis, and a specific timescale on the
Figure 2.2: Explanation of Ryanair Profits, 2005–2009, and plausible future to 2014
Source: Company Reports and author’s estimates.
T
R
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
43
horizontal axis. The current value “today” (the latest time for which data are known) is highlighted in green, just above a vertical dashed line for the time at which it applies. The time path of historical data is shown as a solid green line, and forecasts are denoted by a dashed line.
Word-and-arrow diagrams are commonly used in management books to illustrate a general relationship between connected items. Here, every such link has a more precise meaning—if arrows link items A and B into C, then the numerical value of C can be calculated or estimated from the values of A and B at each point in time. Figure 2.2 follows this rule rigorously.
2.5 From Performance to Resources
Next, we need to know what causes each of the revenue and cost items presented in Figure 2.2. Conventionally, revenue would be explained in terms of market size and market share:
revenue = size of air travel market × market share
Indeed, this is exactly how respected finance books recommend analysts to forecast a company’s revenue and how many firms start developing their business plans and budget forecasts. But while this equation is mathematically correct, it is not a causal explanation of revenue. The equation could equally be stated the other way round—i.e., market share = revenues ÷ market size. The reality, not just in this case but in virtually all others, is that revenue comes from sales volume and price, and sales come from customers. This statement might seem obvious, even simplistic, but that does not make it any less true! And since sales and profits are not normally forecast in this way, it is worth pushing on to find out the consequences of pursuing a more rigorous causal logic.
A true explanation for profit over time requires working back from the income statement, through rigorous causal connections, until we get to factors that management can influence. In Ryanair’s case, this works as follows:
annual fare revenue = passenger journeys sold × average fare paid
The number of passenger journeys sold is not given by market size and market share, but instead by:
passenger journeys sold per year = number of customers × journeys per passenger per year
F
P
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
44
Journeys are taken by people! And it is the choices of those people—to become customers, to cease being customers, and to buy more or less often—that a company tries to influence with its choices of product offering, marketing, and sales effort. Figure 2.3 shows these relationships, over time, for Ryanair, following exactly the same principles as in Figure 2.2.
There is a critical reason for placing the “Estimated customers” graph at the far left of Figure 2.3 in a box. All other items in Figures 2.2 and 2.3 are measures that describe what happened over each year as a whole—the fare revenue of €2,343.7m was made over the period of April 2008 to March 2009, for example. In contrast, “customers” are a quantity of something that exists at each point in time—like the amount of water in a tank, which is what the box symbolizes. Typically, such items are measured at the start and end of a reporting period—how much “stuff” we have at the end of the financial year. This is well understood in the case of cash. We start the year off with a quantity of cash, incur cash flows in and out during the year, and end the year with a different quantity. This is simply the link between the balance sheet and the cash flow statement, but equivalent relationships apply to customers and other items, as well.
Generally, factors of this kind are known as “asset stocks” or simply “stocks.” In the context of strategy, such items are called “resources,” and their behavior is critical to how performance changes over time. This will be explained in Chapter 3.
Customers’ asset-stock character makes it necessary to calculate the average number of customers during the year and their average travel frequency during the year to accurately match the quantity of customers with the rate at which they
Figure 2.3: How customers drive sales for Ryanair
Source: Company Rep[orts and author’s estimates.
F
P
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
45
are delivering sales. If customer numbers were to change significantly over a twelve-month period—as is likely in the case of Ryanair—taking a simple average of year-start and year-end numbers may not accurately explain total sales over the year. The more frequently this average is determined—each quarter or month, for example—the more accurately it will explain the annual sales result.
Some additional details in Figure 2.3 are needed for completeness and accuracy:
� The company also gets additional, ancillary revenue from sales of food and drink, ground transportation, and other services—also driven by customer numbers.
� Customers do not take all the journeys they buy—about 7% are “no-shows,” so fewer journeys are flown than bought.
� The number of customers is not precisely knowable. Many people habitually and repeatedly use this airline, perhaps because it is the only one serving their desired routes, or simply because they prefer it. Others use it only occasionally. These differing customers could be split out into two or more segments, each of which has its own journey frequency and fares. The structure in Figure 2.3 would therefore be copied for each group, and total fare revenue calculated from the sum of these segments.
� While this structure focuses on the firm itself, rather than being driven by market size and market share, it is not immune from external factors. Customers’ average travel frequency is clearly affected by economic conditions, as many airlines found when sales collapsed in the 2008–2009 recession.
Exactly the same principles apply to the company’s costs. Costs are conventionally explained in terms of the company’s success in reducing the percentage of revenue expended on each item. So, for example:
staff cost = revenue × percentage of revenue spent on staff
Again, though, this equation can expressed the other way round: percentage of revenue spent on staff = staff cost ÷ revenue × 100. And once again, this equation does not provide a causal explanation. While we may want to limit such ratios, the reality is that staff numbers drive the cost of staff, and we actually control those costs by hiring and firing people and/or by changing their rate of pay. The true causal explanation of staff cost is therefore:
staff cost = number of staff × cost per staff member
F
P
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
46
Similarly, aircraft numbers drive the costs of operating aircraft for any level of utilization, the number of airports served drives airport-related costs, and the number of routes drives the cost of operating those routes. Everything else being equal, having more or less of each of these resources—staff, aircraft, airports, and routes—will raise or lower the corresponding cost item accordingly. Figure 2.4 shows this analysis for Ryanair’s routes.
Again, some important details and adjustments are needed in practice:
� Some fraction of these costs is driven by adding or removing a resource, as well as by simply having it. It costs money to open a new route. Hiring and firing staff also costs money, as does starting or ending operations at an airport. Ryanair buys its own aircraft, rather than leasing them, so the large cost of these purchases is a capital item and does not appear on the income statement.
� Costs, too, are affected by external factors. The average aircraft cost, for example, climbed sharply during in 2008, due to increases in the prices of oil and jet fuel, then fell back once those conditions eased.
� We have the same problem here with point-of-time values for the resources versus. whole-period values for the costs that they drive. Aircraft numbers, staff, airport costs, and routes are reported as at the end of each reporting period, but the costs are incurred in total during those periods. As with customers, then, we again have to take the average quantity of each resource during each period to explain the total costs for the period, and the more frequently these values are computed, the more accurate is the causal calculation.
The point of this discussion is to provide the tools to help management steer strategy and performance over time. Figure 2.2 offers not only a causal
Figure 2.4: How resources drive Ryanair’s costs, e.g., routes
Source: Company Reports
F
P
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
47
explanation of the company’s performance up to 2009, but also plausible targets for future years out to 2014. This projection makes explicit three key assumptions:
4 Revenue growth is expected to come from increasing sales volume (passenger journeys sold) plus slight price increases. It is not expected that average prices will fall. The fares are still very low in comparison to competitors, however, as this is fundamental to the firm’s strategic positioning.
5 Growth in journeys sold will come mostly from increased numbers of customers—we will see later what is expected to cause that increase.
6 There is expected to be some increase in average customer demand— journeys per passenger per year. In the early years, this will be helped by recovery from economic recession and by continued, modest growth in consumers’ general desire to travel.
2.6 Resources and Nonfinancial Performance
The same logic as we discussed above applies to other kinds of performance aims. The quality of service for a telephone call center, for example, reflects a balance between the incoming call volume and the capacity of the center’s staff to answer those calls. Demand is driven by the population of people who might call, and the rate at which they do; capacity is driven by the number of call center staff and the rate at which each can deal with calls.
This makes the resources-to-performance link directly applicable also to public services, voluntary, and not-for-profit organizations. Figure 2.5 provides an example of causal relationships between resources and performance that are common in many charitable organizations. This particular organization recruits volunteers to support people suffering from a serious, progressive, and ultimately fatal disease. Volunteers provide advice and emotional support, for which they need specialized training. There are currently (time = 0) 356 volunteers supporting 2,511 patients, which is not enough to fulfill the required workload, so calls on patients are sometimes rushed or missed.
Twelve quarters previously, volunteer capacity was only 60% of demand, but a recruitment effort over the last eight quarters has increased volunteer numbers. Planned future recruitment over the next four to six quarters is expected to raise capacity to the point where all patient calls can be handled effectively, with adequacy approaching 1.0. However, it is feared that the organization will face a further rise in patient numbers, due to increasing prevalence of the disease, earlier
R
N
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
48
diagnosis, and longer life expectancy. By Quarter 16, then, the organization fears that it will again be unable to provide fully adequate support for its patients.
2.7 “Stocks” of Resources
The number of patients and volunteers in Figure 2.5, like Ryanair’s customers and routes in Figures 2.3 and 2.4, are shown in boxes because these too are resources, or asset stocks. The number of patients and volunteers are quantities of items that drive demand and capacity, respectively, or are useful to the organization, and are collected over time.
Resources are quantities of items or materials driving demand and capacity—and hence income and costs— that are owned or reliably
available to the organization and are built up over time.
To clarify the distinction between resources and other items, imagine that time is standing still. There would be no revenue or journeys, costs or profits for Ryanair, and no workload for the charity from Figure 2.5, because time has to pass for these items to exist. Indeed, they are actually measured in units per time period—journeys per year, profits per year, workload per quarter, and so on.
Figure 2.5: Resources driving service performance in a voluntary organization
“S
”
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
49
Resources need not be owned or directly controlled to be useful. They need only be somewhat reliable—if they are available today, they are likely to be available tomorrow. On this basis, many organizations find customers are more reliable than staff—fast food chains, banks, and sports teams, for example. The average McDonalds employee stays for only a matter of months, while customers may remain loyal for years or decades.
Table 2.1: Standard categories of tangible resources
Ta ng
ib le
re so
ur ce
Co m
pu te
r ch
ip -m
ak er
La w
fi rm
Ai rli
ne Co
ns um
er br
an d
Ch ar
ity
De m
an d
si de
Fi na
l c us
to m
er s
Co m
pu te
r bu
ye rs
Cl ie
nt s
Cu st
om er
s Co
ns um
er s
Be ne
fic ia
rie s
In te
rm ed
ia rie
s Co
m pu
te r
m ak
er s
Re se
lle rs
Re ta
ile rs
Su pp
ly si
de Ca
pa ci
ty St
or es
Pr of
es sio
na l
St af
f Ai
rc ra
ft Pr
od uc
tio n
pl an
t (P
hy sic
al as
se ts
) St
af f
Pr od
uc tio
n st
af f
Se rv
ice s
ta ff
Sa le
s fo
rc e
Vo lu
nt ee
rs
Su pp
lie rs
Ai rp
or ts
Pr od
uc t r
an ge
Ch ip
ty pe
s Le
ga l s
er vic
es Ro
ut es
o ffe
re d
Br an
ds o
ffe re
d Se
rv ice
s of
fe re
d Ca
sh Ca
sh Ca
sh Ca
sh Ca
sh Ca
sh
“S
”
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
50
The resources displayed in Figures 2.3 and 2.4 are specific to the airline business but are also examples of general types of resources to be found in most organizations (see Table 2.1). Some resources drive demand for an organization’s products or services, while others make up its ability to supply them.
Note that the table 2.1 specifies only a key staff group for each example, although a complete analysis would need to cover all relevant staff groups. Table 2.1 is not an exhaustive list of resource categories—intangible factors and capabilities are also relevant, for example. However, the direct causal relationship between these tangible items and performance leads to a critical observation:
It is not possible to explain performance without information on the quantities of the tangible resources that drive it.
A few details to note about the items listed in this table:
� Not all kinds of firm have all types of resources. Companies that sell directly to the final users of their product or service, such as a retailer, do not have intermediaries (wholesalers, dealers, or agents).
� Some organizations depend on two quite different groups of customers, either or both of which may generate revenue. For example, media companies have viewers or readers but also serve advertisers, and eBay depends on having both buyers and sellers.
� Where intermediaries are involved, they may well be the company’s true customers—that is, the organizations that actually buy from the company. For example, Procter & Gamble (P&G) sells its products primarily to retailers, rather than to consumers. Nevertheless, end- customers are a critical resource, so are vitally important to strategy. P&G certainly depends on, and needs to know how to influence the end-consumers of its products.
� For many service organizations, staff are their capacity to provide their service, rather than physical plant or equipment resources.
� Depending on the situation, not all resources need be made explicit, even though they exist. Many companies have such easy access to suppliers who want to serve them that there is little to be gained by specifying those suppliers. (A publisher’s authors or an airline’s airports, though, are both hard to obtain and critical to success.) Investors, as suppliers of capital, may not usually need to be identified but do need special attention in some cases, such as with fast-growing new ventures.
“S
”
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
51
� Specific cases may involve types of resource that are uniquely important to those situations. Oil companies and other raw-material producers, for example, are fundamentally dependent on their reserves.
Public service, voluntary, and not-for-profit organizations feature very similar resource categories. They usually have a population that drives demand, even if that population does not generate revenue. Criminals drive demand for policing, patients drive demand for health care, malnourished people drive demand for aid organizations, and so on. Not-for-profit organizations also feature staff, of course, and very often those staff make up much of the capacity to deliver their services—police officers, doctors, and aid workers for example. Finally, many voluntary organizations require a special category of resource—the donors who provide their funds.
2.8 When Resources Themselves Are the Objective
There is one case in which the rule that resources drive performance is not relevant—when the objective that the organization is pursuing is itself a resource. This can arise at different levels:
� A particular function or department may aim to build some quantity of resource by some point in time—numbers of suitably qualified staff, or numbers of products on the market, for example.
� A whole business may have an objective to achieve some target quantity of a resource. Many companies have goals to reach a certain number of customers by some point in time. This may be more than simply the responsibility of sales and marketing departments if, for example, product performance, service quality, or other factors could affect customer retention.
Objectives for intangible resources are also common—staff morale and market reputation being common examples. Again, such aims may apply to the whole organization or just to specific parts of them.
When an objective concerns a resource itself, rather than some other performance outcome, we can forget about the principle that resources drive performance for now, and go straight on to how such resources are won and lost, discussed in Chapter 3.
W
R
T
A
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
52
2.9 Specifying and Quantifying Resources
To understand performance accurately enough to make good strategic choices, it is not enough to simply note that certain resources are involved and move on—the resources must be quantified, as in Figures 2.3 to 2.5. This requires them to be properly specified.
For many resources, accurate specification is easy. There is no problem counting Ryanair’s aircraft, for example! Staff and customers, too, are easy to define and count. There is more to think about with both staff and customers, though.
It may be important to specify different groups of customers. We already noted Ryanair’s regular passengers and other occasional customers. Distinct staff groups may also need to be specified, such as production staff versus sales staff versus service or support staff.
Defining and counting customers is not always easy. Counting customers when there is a contractual relationship (e.g., for cell phone companies or electricity suppliers) is easy, but how about a bank? Quantifying the total number of account-holders may not be a problem, but some account-holders may have little money in their bank accounts and conduct so few transactions that not meaningful customers. This case would best be handled by segmenting customers into the most important groups and, as with staff, leaving less-important customers in an “other” category.
Of the items listed in Table 2.1, capacity can be the most problematic. For an airline, it is simple enough to count the number of aircraft, from which the passenger-carrying capacity can be calculated, depending on the length of routes that are offered. Often, though, it is not helpful to count the physical units of capacity. A chemical plant’s capacity, for example, is not the sum of the number of pipes and pumps of which it is made up. A simple solution is to express capacity in terms of the physical output that can be produced—tons per hour, for example. This also works in cases where capacity involves information technology. In banking, a key capacity is the number of transactions that can be processed per hour, not the number of servers or terminals.
Product range can also be tricky to define and measure. For an airline, this is the number of routes it operates. Retailers count “stock-keeping units” (SKUs), pharmaceutical firms have a number of drugs on the market, and car makers offer a range of models. But is a single drug offered in adult and child formulations one product or two? Are different variants of a single car model different products?
S
Q
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
H
R
D
P
53
So, it is not always an easy task to specify and measure resources, but there is value from the very effort of trying to do so. Many companies benefit from thinking carefully about the nature and number of customers they serve or about the products they offer. New products, for example, are often added without considering the impact on those already offered, leading to unhelpful and costly proliferation. Management can make better decisions about what to do if they have a clear and quantified specification of the resources that drive the sales, capacity, revenue, costs, and profits of the business. Professional service firms, for example, may struggle to agree on strategy because they have not clarified exactly what “products” they offer or might develop. A law firm that treats each client engagement as a unique service will often do better to define and specify a clear set of services that it offers. This not only makes it easier to market and sell those services, but helps control workloads and costs, increasing profitability.
The resource-based view (RBV). The discussion and specification of tangible resources in this section is not the same as the textbook resource-based view of strategy, discussed in Section 2.2. The strategy dynamics approach differs from RBV in three principal ways:
1 Since tangible factors comprise the heart of any business or organization, it is necessary to make them explicit, quantify them, and connect them to the organization’s performance outcomes. Intangible factors and capabilities will be added to the analysis if their influence operates through the impact they have on the tangible factors that directly drive performance.
2 Strategy dynamics does not limit analysis to resources that are owned or controlled by the organization, but includes any that are in any way reliably accessible, customers being the main example.
3 Lastly, the approach makes explicit and quantifies the “complem- entarity” amongst resources—how they work together as an integrated, functioning system, to generate performance. This will be explained further in Chapter 4.
.
S
Q
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
54
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 3 RESOURCES WON AND LOST
Chapter 2 explained that resources are useful items that have been built up over time, and that there is a direct, quantitative relationship between the quantities of those resources and performance at all times. The next logical question, then, is what determines the quantity of resources at any point in time? The causal relationships we have discussed so far have come in the following form:
If we know the value of A and B at any time, then we can calculate or estimate the value of C at that time.
Resources and other asset stocks do not obey any such relationship. We explain quantity of resources as follows:
The quantity of resource X today is the total amount of X that has ever been added up to this time, minus the quantity that has ever been lost.
This is not an opinion, a theory, or a finding from statistical analysis of large data-sets—it simply is how the world works. Nor is it an approximation. The amount of cash in your bank account is precisely, to the cent, the sum of every amount ever paid in minus every amount ever taken out. This is the critical feature of resources, without which no sound understanding of strategy and performance is possible.
Resources accumulate and deplete (fill and drain) over time.
The implications are profound. The number of customers you have today is not explained by anything else—not by the prices charged, products offered, marketing dollars spent, or sales effort expended. The number of customers is
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
56
precisely the sum of every customer ever won, minus every customer ever lost, since the day the business started. And if we cannot explain the number of customers in terms of anything but its own history, then neither can we explain anything that depends on that number—such as sales and profits! So be very skeptical of any business frameworks based on claims to have found that factor X drives profitability.
Likewise, the number of staff today is not explained by salaries, career prospects, or how well they are treated. It is precisely the sum of every person ever hired, minus every person who ever left or was fired. The same logic applies to the number of products offered, the amount of capacity in place, and the amount of cash. Cash, of course, is the one resource that is explicitly reported in this way, in a company’s balance sheet and cash flow statements.
Fortunately, we do not have to know all of a resource’s history to make use of the concept. If we know how much of resource X there was at the start of any period, and how much X was added or lost during the period, then we know precisely how much X there will be at the end of the period—and that quantity will then be how much there is to start the next period.
This principle can now be added to our emerging theory of performance:
1 Performance at any time depends on resources, and
2 Resources accumulate and deplete over time
3.1 Quantifying the “Bathtub Behavior” of Resources
Although the calculation of resource quantities is quite different in nature from most other causal relationships, we already know how to work out what happens with accumulating stocks. If we are filling a bathtub and want to know how much water it will contain at the end of the next minute, we measure how much there is at the start of the minute, add the amount that flows in through the faucet and subtract how much runs away down the drain. This is why the way resources change over time is sometimes termed “bathtub behavior.” The quantity of resources is built up by the flow of new resource into the stock—the inflow. And resources are depleted by quantities of the resource flowing out of the stock—the outflow—whether through misfortune, the actions of others, or by deliberate intent. This applies to intangible factors as well as tangible resources, so:
� Winning customers adds to a customer base; recruiting new employees increases our staff resource; marketing our products raises awareness among potential customers; training our staff enhances their level of skill.
Q
“B
B
”
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
57
� Customers are lost to competitors; resignations and firing reduce our staff base; discontinuing a product reduces our product range; intense pressure on staff reduces morale.
The mathematical behavior of this process is termed “integration,” and although the term may not be familiar to you, you probably know how to do it. You likely know more or less how much money was in your bank account at the end of last month, and have some idea how much will be added to it during this month, and how much you will pay out of it. You can therefore work out what will be left at the end of the month.
Figure 3.1 quantifies this idea. Think of the box in the middle as a bathtub containing your cash, and the wide arrows as pipes carrying cash into and out of the tank. Think of the ovals as pumps that determine how fast that cash is flowing. The ovals are highlighted in yellow because what they represent turns out to be critically important.
If the numbers in Figure 3.1 continue through time, you will have $1,200 at the end of next month, $1,400 at the end of the month after that, and so on. It is also easy to work out what will happen if your rent goes up from $1,500 to $1,800 per month.
The relationship between a resource stock and its flow rates shown in Figure 3.1, known as a stock-and-flow diagram, is trivially simple, but just asking about these basic numbers can transform strategy and performance. One popular
Salary 350
Investment income
150 Rent 200
Food 100
Clothes, fun … 150
Cash in bank account
$1,000
Cash in ( )$ per month
500 450
Cash out ( )$ per month
Figure 3.1Figure 3.1: Drivers of flows into and out of your bank account
Salary 350
Investment income
150 Rent 200
Food 100
Clothes, fun … 150
Cash in bank account
$1,000
Cash in ( )$ per month
500 450
Cash out ( )$ per month
Figure 3.1
A working model showing how changing flows of money into and out of a bank account change its balance over time is at http://sdl.re/mmgs. This is an example of working models included in the strategy dynamics course.
Q
“B
B
”
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
58
over-the-counter drug1, had been slowly losing sales and market share for many years, and no marketing efforts had succeeded in changing this trend. About 22 million people now use the product, a number that is increasing by just under a million each year (meaning average purchases per person are falling). This information is not enough, however, and the example illustrates a critical issue:
It is important to know, separately, the inflow and outflow rates for a resource.
Winning one million customers a year and losing none is not the same as winning five million a year and losing four million, even though the resulting customer base is the same. This mature pharmaceutical product might be expected to feature little change in its customer base, but it turned out that 3.7 million new consumers were won each year, and 2.9 million were lost! Marketing strategy shifted from winning new consumers, which was already working well, to retaining existing users, especially heavy buyers. The sales decline was halted, and with a reduction in marketing spend. Similarly valuable insights have been found in a wide variety of industries, from soft drink producer Coca-Cola to software, car manufacturing, and consulting firms.
A common response to such examples is “That’s obvious.” Well, yes it is, but this firm had been worrying about its sales and market share problem for many years, and never found the solution because it never looked at these numbers. If you don’t ask the question, you will not find the answer, no matter how obvious it seems afterward.
Surprisingly few organizations track these numbers or know what to do with them. One large bank, for example, studied hundreds of pages of data about all the different kinds of financial products each customer held, segmented by age, income level, region, and just about every other criterion you might imagine. But their analysis included not a single number on how many customers were won and lost by each product each month. The data could be found, but no-one had ever asked.
This stock-and-flow mechanism is not just critical for understanding customers and sales. In some cases, especially for professional service firms and others that rely on scarce, skilled employees, winning and retaining staff is vital to strategy and performance. Infosys Technologies, a world leader in IT services based in Bangalore, India, offers an interesting statement on the front of its 2007–2008 Annual report2.
1 The product cannot be named for competitive reasons.
Q
“B
B
”
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
59
“Our core corporate assets walk out every evening. It is our duty to make sure that these assets return the next morning, mentally and physically enthusiastic and energetic.”
The first eleven pages of that report go on to talk, not about market conditions, customers, or financials, but exclusively about its people and what the company is doing to develop and retain them.
3.2 Accumulation over Time
Chapter 2 pointed out that the direct relationship between performance and resource quantities is true at all times in the past and future, so long as the same activity continues. The same is true of the relationship between the inflows and outflows of a resource, and how its level changes over time.
In Figure 3.2, a business wins twenty customers per month, and initially loses only twelve per month, but this loss rate rises in successive months to fourteen, sixteen, eighteen, and so on, until by the end of month 12, customers are leaving at the rate of thirty-six per month. There is a constant win rate and a straight-line trend on customer losses. But the stock of customers does not follow a straight line, but instead follows a curved path through time, peaking at 120 during month 5 (when twenty customers are won and another twenty are lost), and then decreasing ever more rapidly until the year ends with only sixty-four customers in place.
2 http://www.infosys.com/investors/reports-filings/annual- report/annual/Documents/Infosys-AR-08.pdf
Figure 3.2: How changing rates of customer losses affect customer numbers and sales
T
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
60
This figure illustrates why it is difficult for the human mind to intuitively grasp how resources will change over time. Even highly numerate people find it hard to estimate what will happen to the level of a resource over time, given simple changes to win and loss rates. Such misunderstandings can be serious. Cutting your spending rate will not, for example, cut your borrowing unless the cut is so large that spending falls below your income. Similarly, cuts to greenhouse gas emissions will not reduce atmospheric levels of these gases unless the cuts are so deep as to bring emissions below the planet’s natural absorption rate. Unfortunately, this absorption rate is a small fraction of the emission rate, which is why climate change will not even start to be reversed unless emissions are cut by at least 75% from 2009 rates.1
This non-intuitive behavior makes it vital to lay out the relationships between a resource, its flow rates, and factors that depend on that resource in an arithmetically accurate way. This structure must also be linked to performance outcomes, such as the sales chart in Figure 3.2, if those too are to be understood. Each item that changes over time needs to be displayed on a chart with a quantitative scale and clear time frame. In Figure 3.2, the path of total monthly sales shown by the dashed lines can only be understood if the rising sales per month to the average customer are displayed alongside the changing customer base.
While Figure 3.2 might be an unfamiliar way of looking at a situation, it is perfectly easy to capture exactly the same relationships in a simple spreadsheet. However, diagrams provide a clearer image of what is happening than a page of rows and columns. They provide a strong focus for a team to address what needs to be done, and how much impact their choices will have, and by when. In this case, discussion could focus on the same three critical questions highlighted at the end of Chapter 1.
� why the loss rate has been rising at the rate it has
� where the future customer numbers and sales will go in future if the loss rate (a) continues to rise, (b) stays at the same high rate, or (c) is brought back to a lower rate once more
� how the company might reduce the loss rate, how quickly it might be cut, and what will then happen to the number of customers and future monthly sales.
1 See http://climateinteractive.org for extensive information on this case.
C
R
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
61
3.3 Consequences of Resource Accumulation
This defining characteristic of resources—called “asset stock accumulation”—has long been known to be critical to strategic performance. A widely cited article introduces the bathtub metaphor, and concludes that “a key dimension of strategy formulation may be the task of making appropriate choices about strategic expenditure … with a view to accumulating required resources and skills.”1 The article points out why a firm with already-developed resources has a substantial competitive advantage—in our terms, the ability to deliver stronger sustained growth in cash flows than competitors:
� Time compression diseconomies. This simply means that it takes time to accumulate resources. For example, no matter how aggressively a new competitor might try to copy the branch network of McDonald’s or Starbucks, it will take years or decades to match the sheer scale of that resource. Doubling the effort does not halve the time taken, because of limits to how quickly things can be done. This means we can readily explain the long-term leadership of firms, with no need to rely on abstract “strategic” resources and capabilities claimed by RBV to be critical. Such factors may add to that advantage, but are not essential.
� Asset mass efficiencies, or, “the more you have, the more you get.” This applies to many resources and will be explained in Chapter 4. For now, consider this. If you owned a great retail store location, would you rent it to Starbucks or to a complete unknown in the coffee store market? If you are a skilled young professional, would you be more likely to join the leading global consultancy firm McKinsey & Company, or an unknown consulting firm?
� Interconnectedness of asset stocks. In simple terms, building any resource depends on other resources already in place. Taking McKinsey as an example once more, major firms seek advice from this firm because it has the brightest business analysis brains available, and the brightest young business brains join the firm because it serves the most important clients. This interconnectedness is crucial to the performance of sector-beating firms, and will also be explained in Chapter 4.
1 Ingemar Dierickx and Karel O. Cool, Asset stock accumulation and sustainability of competitive advantage, Management Science, 35, 1504–1511.
C
R
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
62
� Asset erosion. Many tangible and intangible assets deteriorate unless effort is made to maintain them—i.e., resources suffer outflows. Physical plant wears out, staff skills become obsolete, products lose their appeal, and customer loyalty to a brand decays. This deterioration threatens the competitive advantage that a firm has from its superior resources, a challenge that can vary widely for different firms in the same industry.
� Causal ambiguity. It can be hard, even for the firm itself, to know why exactly a resource accumulates and depletes at the rate it does. Some changes, like adding capacity, simply reflect management’s decisions, but it’s not so easy to know why customers are won and lost at a certain rate. As Chapter 2 explained, much of what causes performance outcomes is simply arithmetical, back to the point where resources drive sales and costs. This chapter has shown that the current value of each resource is not just influenced, but totally determined by resource flows over time, so that that relationship is not ambiguous either. Causal ambiguity must therefore lie in explaining what causes resources to be won and lost at the rates observed.
These are powerful arguments for the critical importance of asset stock accumulation to strategy and performance. It is remarkable, then, that no useful models exploiting the mechanism have emerged from the strategy field in the last twenty years. In addition, stock-accumulation has a particularly serious consequence – it destroys the validity of regression-based methods typically used in the quest to explain business performance (see Appendix 2).
3.4 Resources Won and Lost by Ryanair
Ryanair explicitly reports some of the resources it adds—aircraft added and new airports served each year, for example—but not others. We know how many staff were employed at the end of each period, and so can work out the net increase. However, we do not know separately how many employees were hired and lost. Where staff turnover is modest and not a problem, as in this case, this unknown may not be an issue, but it can be critical in other cases.
Figure 3.3 shows the history and a plausible future for the growth in Ryanair’s customer base, fleet, and route network. The changing number of customers is the most problematic item. Not only is the company not explicit about how many distinct customers use its service—it may not even know—but we know nothing at all about the number of new customers won each year or the number who decide not to use the airline again. The values shown in these charts, then, are
R
W
L
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
63 only illustrative. However, if accurate, they would imply that by year ended March 2010 there would be 14.4 million customers (13.3 + 3.2 – 2.1).
Although not shown separately in Figure 3.3, the company pays close attention to how many aircraft are acquired and disposed of each year, not just the net addition. A faster renewal rate keeps down the average age of the fleet, with benefits of customer comfort, reliability, and fuel efficiency.
For the route network, the separate issue of opening and closing routes is also important. Opening a route is costly, so deciding to close it, due to disappointing passenger volumes, is undesirable. The bigger the company becomes, the more of the potential market it captures, but the more problematic the issue of route openings and closures becomes; it becomes increasingly difficult to find viable routes that are not already being served. The difficulties of this particular issue explain why the company, though keen to boast of the large number of new routes it opens each year, is rather quieter about how many it closes.
Figure 3.3: Growth in Ryanair’s customer base, aircraft fleet, and route network
Source: Company Reports and author’s estimates
Figure 3.4: Changes to the number of patients supported by a voluntary organization
R
W
L
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
64
Similar analysis also applies to non-business cases. Figure 3.4 shows the reasons why the number of patients who will need support from the charity in Figure 2.5 will grow to the scale feared by quarter 16. There is expected to be a strong increase in the rate of new patients seeking support, with slower growth in the numbers lost each quarter.
3.5 “Accounting” for Resources
Although we are not usually so explicit about the accumulation and depletion of resources as described above, it is of course how we account for cash. So if it important to control this critical resource by accounting for its changes in this way, why do we not do the same for other key resources? Many companies state, for example, that “people are our most important asset,” but include no accounting whatever for what has happened to those assets during the period in question. Table 3.1 shows how this would be done for the most common tangible resources.
Table 3.1: Accounting for resource movements
While this table shows how to report changes that have occurred to a business over a year-long reporting period, this time frame would not be adequate for management control. Just as most companies look at cash flows every month or week, it is equally important to look at other rapidly-changing resources, such as customers and staff, on a more frequent basis. Remember also that it will be necessary to examine such changes to subgroups of some resources, especially customer segments and key staff groups.
Cash Customers Staff Products Capacity
Opening balance
Cash at start of year
Customers at start of year
Staff at start of year
Products at start of year
Capacity at start of year
Added Cash in Customers won
Staff hired Products launched
Capacity added
Lost Cash out Customers lost Staff who left Product dropped
Capacity closed
Closing balance
Cash at end of year
Customers at end of year
Staff at end of year
Products at end of year
Capacity at end of year
“A
”
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
65
3.6 Control over the Building and Retaining of Resources
Some resource flows are determined directly by management—Ryanair simply decides how many aircraft to buy and sell, for example. Decisions to add other resources are not so guaranteed.
A law firm might want to hire ten lawyers this month, but its success depends on the availability of newly qualifying lawyers with the particular skills the firm is looking for, on rivals’ efforts to hire those same people, and on the firm’s reputation. It is not usually possible to be exactly certain of developing a target number of new products in a given period. Winning customers, of course, is rarely under accurate control at all. While management may want to add a certain number of customers in any year, the actual result depends on the decisions of customers themselves.
This range of controllability is shown in Figure 3.5. In each case, management wishes to add fifty units during the period. When buying vehicles, that is exactly
what they get. Staff hired may be more or less than the desired fifty, depending on the number of job offers accepted. For customers, marketing may build awareness, price changes may change customers’ perceptions of value, and sales effort may be sized to target a number of leads. But whether these factors will bring in ten, fifty, or one hundred new customers—or none at all—is rarely knowable. Note that the actual decisions available in each case (in red text) become increasingly remote from the vital flow rates as we move from direct to indirect control.
Management should be concerned about resource losses as well as gains, and the degree of control varies on these outflows, too. We can sell exactly the number of vehicles we do not need, but may not be able to simply lose staff that we no longer need. Customers, of course, are lost for all kinds of reasons, including the actions of competitors, and our control over their loss is limited at best.
Customers
50
Vehicles
50
Staff
50
Direct control
Indirect influence
new customers per month
perceived value
marketing spendprice
sales effort
vehicles bought per year new staffper month
jobs offered
100
fraction accepted
0.5 awareness
Figure 3.5
Figure 3.5: Varying management control over resource flows
Customers
50
Vehicles
50
Staff
50
Direct control
Indirect influence
new customers per month
perceived value
marketing spendprice
sales effort
vehicles bought per year new staffper month
jobs offered
100
fraction accepted
0.5 awareness
Figure 3.5
C
B
R
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
66
Note that if all resource flows were zero, then the firm’s resources would not change from period to period. So under constant external conditions, performance would be constant. The reason, then, that we are focusing so much on the flow rates of resources and management’s control over those flow rates is simple:
Management steers strategy and performance by influencing resource flow rates.
3.7 Generic Behaviors and Their Drivers
The least controllable resource flows are those which are affected by the choices of other people—employees and customers in particular. Therefore, it is helpful to clarify those mechanisms and understand the forces that cause these flows to change. Generically, there are three different things people choose to do that we want to influence (see Table 3.2):
1 join us (e.g., become a customer; accept a job)
2 leave us (e.g., stop being a customer; resign from a job)
3 do more or less (e.g., buy more from us; work less hard)
We have already shown these three mechanisms, and their links to performance in Figure 3.2.
Ryanair’s business growth provides a good illustration of how these three behaviors combine to determine sales. The volume of business (passenger journeys booked) depends on the current number of active customers and their average travel frequency, and the number of active customers depends on the history of customers won and lost. Note that the same three mechanisms apply in non- business cases, as well. People decide to become donors to a charity, to cease being a donor, and to change their rate of giving.
The first two choices (joining and leaving) are switching decisions, to move from one state to another—i.e., from not being a customer to being one (joining), or from being an employee to being an ex-employee (leaving). They are also quite significant but infrequent decisions. People do not keep changing their minds about their favored products or services, nor about starting and leaving a particular job. Even for the lowest-value products with the widest range of alternatives, consumer choice remains remarkably stable over long periods.
The third choice—to do more or less of some activity—is fundamentally different. This activity is done continually or rather frequently, so there is no change in the basic state of mind, merely an increase or decrease in strength of feeling.
G
B
T
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
67
Management actions and choices may influence any or all of the three behaviors. For example:
� A price increase may cut the customer win rate, increase the loss rate, cut the average purchase rate.
� A marketing promotion could be designed to capture new customers, rather than increase loyalty or the average purchase rate.
� Training service support staff may be aimed at cutting the customer loss rate.
� Offering higher starting salaries aims to increase the hiring rate.
� Efforts to increase staff productivity could lead to higher staff turnover.
� Bonus schemes aim to raise the efforts of staff.
� Increasing the numbers of police on the streets may deter people from taking to crime, encourage existing criminals to give up, and reduce the rate at which criminals commit crimes.
The groups of people we are concerned with (customers and staff) are also subject to other influences, of course, in particular those of competitors (or competing choices) and other external factors beyond our control.
Lastly, we need to complete the link to performance arising from these behaviors. How much total activity actually occurs depends on the number of people involved and the rate at which the average person is doing it That is:
Total sales = number of customers × purchases per customer Total work output = number of employees × output per employee Total crime rate = number of criminals × average crimes per criminal
We now need to go one stage further, and look at what might affect the three critical behaviors of joining, leaving, and doing more or less. There are many models of consumer choice in the marketing field, suited to different markets and varying in complexity. Some models go further and deal with how customers move between differing states—examining infrequent, regular, and loyal buying, for example—an issue we will look at in Chapter 6. One popular framework in strategy work is the value curve, discussed in Section 2.2.
Figure 3.6 shows how a mix factors may influence the whole structure of customer acquisition, retention, and purchase frequency, and thus explain how Ryanair’s sales may have developed over time. The items in red text are factors that Ryanair management can directly decide upon.
G
B
T
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
68
The figure includes some additional items influencing the three behaviors (dashed lines indicate that the causal relationship is not fully defined):
� The single largest factor winning new customers is simply opening operations at additional airports to reach new potential customers and offering routes from those new airports. The value-curve factors then determine how much of that potential is captured.
� The mix of factors has quite different effects on the behaviors of different customer segments. Ryanair’s offer is especially appealing to leisure travelers, but less well-suited to the needs of business professionals.
Fi gu
re 3
.6 : V
al ue
-c ur
ve fa
ct or
s d ri
vi ng
th e
th re
e cu
st om
er b
eh av
io rs
o f R
ya na
ir
So ur
ce : A
ut ho
r’s e
st im
at es
G
B
T
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
69
� The list of factors affecting behavior is not complete—quality of service will affect both customer retention and journey frequency, for example, and word-of-mouth recommendations will affect the win rate.
� Each factor has a distinct influence on each behavior. Choice of destinations may have a big impact on how frequently customers travel, for example, but may not seriously affect whether a customer is likely to switch airlines (leave) completely.
� In particular, certain factors cannot influence customer acquisition because people who are not yet customers have no experience of them—service quality, for example. Potential customers can only respond to what they hear about such issues, i.e., through reputation.
Important Detail: Dealing with consumables and durables
Finally in this Chapter, we should explain that the rule ‘customers drive sales’ operates differently for durable products, as compared with products and services that are repeatedly consumed. In the cases of a consumer brand and an airline, sales volume depends on the stock of current customers and their frequency of purchase. But that is only true because consumer brands and air travel are both consumable products. For durable products, such as televisions or cars, sales volume and revenue arise from winning customers, i.e., the inflow to the customer resource. The stock of active customers is then called an “installed base” of owners, who may not generate any continuing sales revenue at all until the next time they replace the product. Figure 3.7 (overleaf) shows how sales rates differ for a product serving the same potential market, depending on whether it is durable, semi-durable, or consumable. In contrast to very durable products, such as ovens, others are semi-durable, items that are replaced relatively frequently, but not repeatedly consumed, e.g., athletic shoes.
G
B
T
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
W
L
70
Fi gu
re 3
.7 H
ow d
ur ab
le , s
em i-
du ra
bl e,
a nd
c on
su m
ab le
p ro
du ct
sa le
s d ep
en d
on c
us to
m er
s a nd
th e
cu st
om er
w in
r at
e
O w
ne rs
‘0 00
To ta
l n ew
cu st
om er
s 00
0/ m
on th
0 36
m on
th s
20
S al
es v
ol um
e 00
0 un
its p
er m
on th
Fr ac
tio n
of o
w ne
rs re
pu rc
ha si
ng pe
r m
on th
0
6. 1
50 0
7. 0
42 8 U
ni t v
al ue
$1 ,0
00
50
6. 1
7. 0
S al
es r
ev en
ue $m
ill io
n/ m
on th
20 6.
1
7. 0
D U
R A
B LE
P R
O D
U C
T
O w
ne rs
‘0 00
To ta
l n ew
cu st
om er
s 00
0/ m
on th
0 36
m on
th s
20
S al
es v
ol um
e 00
0 un
its p
er m
on th
Fr ac
tio n
of o
w ne
rs re
pu rc
ha si
ng pe
r m
on th
0. 1
6. 1
50 0
7. 0
42 8 U
ni t v
al ue
$1 00
50
0. 61
49 .8
S al
es r
ev en
ue $m
ill io
n/ m
on th
10 6.
1 4.
98
S E
M I-D
U R
A B
LE P
R O
D U
C T
C us
to m
er s
‘0 00
To ta
l n ew
cu st
om er
s 00
0/ m
on th
0 36
m on
th s
20
S al
es v
ol um
e 00
0 un
its p
er m
on th
Fr ac
tio n
of a
ct iv
e cu
st om
er s
re pu
rc ha
si ng
pe r
m on
th 1.
0
6. 1
50 0
7. 0
42 8 U
ni t v
al ue
$1 0
5 00
43 5
S al
es r
ev en
ue $m
ill io
n/ m
on th
10
4. 35
C O
N S
U M
A B
LE P
R O
D U
C T
m on
th s
0 36
Fi gu
re 3
.7 I
D
:
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 4 INTERDEPENDENCE AND THE
STRATEGIC ARCHITECTURE
So far, we have shown (a) that there is a rigorous, mostly arithmetical relationship between performance and the tangible resources on which performance depends, and (b) that those resources fill and drain—again following rigorous arithmetical rules. We also started to explain what drives those resource gains and losses. Continuing this logic leads to a simple but complete model of how an organization actually functions and delivers performance. Three classes of factor drive the inflows and outflows of resources:
1 Management decisions
2 External factors, including competitors’ decisions
3 Existing levels of resources
Section 3.5 showed that management simply chooses how much of certain resources it wants to add or remove—by adding capacity, launching products, or by hiring or firing staff—while flows of customers and staff are less easy to influence. Chapter 8 will say more about how we might actually make those choices.
4.1 Competition and Other External Factors
Whether management’s choices of price, marketing spend, service level, and so on actually succeed in winning and retaining customers and persuading them to buy also depends, of course, on what competitors choose to do. For example, if all other factors are equal:
� A company spending more on marketing a particular brand than a competitor will likely win customers faster.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
72
� A company providing customers with a higher level of service support than competitors will likely lose customers at a slower rate.
� A company selling products at a lower price will likely capture a higher rate of sales to customers whom it shares with higher-priced rivals.
Chapter 7 will explain in detail how rivalry for resources operates. For now, we need only note that competition also impacts our ability to win and retain other resources, especially staff. This is a form of rivalry that primarily affects public services and the voluntary sector. A public service organization competes against other employers to hire school-leavers, while charities compete for donors, for example.
We can now be much more specific about how such factors, mentioned briefly in Section 2.1, actually work—by affecting both activity rates and the rates at which resources are won and lost. For example, rising income levels lead to increased numbers of people taking cruises for the first time and increase the frequency with which people take cruises. The 2008–2009 recession, on the other hand, slowed the flow of new car buyers dramatically. External factors affect staff flows too—rising unemployment can reduce the rate of staff losses, for example.
Demographic and social factors can also be powerful. The rate at which new viewers are won by a children’s television program reflects the number of children entering the target age group each year. A more bizarre example of social-factor impact concerns the increased entry into forensic science degrees, following the popularity of the television series CSI: Crime Scene Investigation. The increased enrolment later led to a surplus of potential staff for the profession as newly qualified specialists graduated and entered the job market.
4.2 Existing Resources Drive Gains and Losses
The third and most critical factor driving flows of resources is the quantities of resources that already exist. More sales people will cause us to win customers faster and slow the loss of existing customers. More research staff will speed development of new products, and more maintenance people will slow the rate at which failed equipment has to be replaced. Such linkages create the interdependencies that make an organization a functioning system. Some such relationships are clear and simple.
People are not the only resource that can drive the growth of other resources. A wider product range helps win customers more quickly, for example. Note that neither people nor product range are actually “used up” in helping other resources grow. Cash, though, is a special case—it is usually consumed when used to grow other resources—for example, buying more capacity, spending on marketing to
C
O
E
F
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
73
win customers, or raising salaries to slow staff losses. A faster inflow of customers, then, might require a higher outflow of cash.
Ryanair offers one especially clear example of how the existing level of one resource drives growth in another. Leisure travelers like to visit a variety of places, so the number of routes an airline offers has a big impact on its ability to win new customers. For example, if you want to travel from Paris to Barcelona, but a particular airline does not offer flights on that route, you will not become a customer, but if it does, you may not only start using them, but consider them for other journeys in future. The more routes in total that an airline offers, the more people like you can be won as customers, and the faster its customer win rate will grow. Figure 4.1 shows this relationship for Ryanair from 2004 to 2009 and provides plausible projections to 2014.
Quantities of resources also affect the rate at which other resources are lost, usually by failing in some way to support demands that those other resources create. If Ryanair has too few staff for the passenger journeys sold, for example, customers making those journeys will have a poor experience and may not use the airline again. Too few aircraft for the routes and flight schedule will result in delays and again increase the rate of customer losses (Figure 4.2).
Figure 4.2 summarizes the causal links between resource shortages and customer loss for simplicity. For example, it is the number of passenger journeys flown, rather than booked, that drives work for staff and hence service quality, and both delays and levels of service quality need to be properly specified. Nevertheless, the factors involved and the relationships between them can be rigorously defined and used to calculate these two key causes of customer losses.
Figure 4.1: How routes drive customer growth for Ryanair
Source: Company reports and author’s estimates. E
R
D
G
L
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
74
The next sections discuss two special cases of this kind of causal link—when resources affect their own gains and losses, and when potential resources limit growth.
4.3 How Resources Drive Their Own Growth and Loss
The simplest case of a resource level influencing flow rates is when a resource drives its own growth, leading to self-reinforcing feedback. An example of this is when word of mouth wins new customers. If one hundred existing customers win a company twenty new customers each month, then two hundred customers will win forty (providing nothing else changes). When this arithmetical relationship exists, we have truly exponential growth. Figure 4.3 demonstrates this causal chain for Ryanair’s customer growth. Keep in mind that for this to work, other factors must be satisfactory (e.g., maintaining low fares that are attractive to customers).
The mechanism does not arise in all cases, so you need evidence that current customers really do influence others to join. Rapid growth alone does not provide that evidence, since other mechanisms could be at work.
Such self-dependency between resources and their own growth rate can apply to other resources, too—for example, when people who enjoy their jobs recommend their friends to apply to the same employer. Nor need such mechanisms rely on active communication between existing participants and those who may be won.
Figure 4.2: How shortages of staff or aircraft drive Ryanair’s customer loss rate
Source: Company reports and author’s estimates.
H
R
D
T
O
G
L
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
75
Important Detail: How feedback drives and limits growth and decline
The fact that resources can drive their own growth and decline gives rise to “feedback” in business systems, which comes in two varieties. Reinforcing feedback is at work at the right of Figure 4.3—an increase in customers gives rise (everything else being equal) to more contacts per year, more customers won, and therefore more customers in the next period. If nothing were to stop this process, the customer base would grow exponentially. Of course, something will stop it—e.g., the decreasing number of potential customers.
So-called balancing feedback is at work at the left of Figure 4.3. Customers leaving the potential population reduce the number remaining, and therefore cut the number of contacts with active customers that can happen each year. As a result, the number of customers who will leave the potential population next period will be lower. Balancing feedback is also at work in Figure 4.2, where any loss of customers reduces the number who might then experience delays or service problems, and thus slows the loss of customers in future periods.
Tracing out these feedback structures is a popular way for teams to try and capture the big picture of the system they are trying to manage, so they can quickly identify opportunities to improve performance, typically by removing balancing limits or enabling reinforcing feedback. However, the method requires great care, because without supporting data, participants are very likely to assume causal relationships and feedback that, while perfectly plausible, do not in fact exist. A further problem is that the behavior of interacting feedback loops is virtually impossible to grasp intuitively from qualitative diagrams, risking the possibility of false insights and erroneous decisions.
Instead, it is strongly recommended that the causal logic be built up progressively, including data on how resources and other items are changing over time, as described in this book. Since it can be time- consuming to source all the required data (some of which may simply not be available), a short-cut is to develop the architecture with estimated sketches of how the team believes performance, resource levels, flow rates and other items are changing. Any critically important causal relationship can then be investigated by seeking confident supporting data.
H
R
D
T
O
G
L
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
76
Airport operators who welcome new services by Ryanair can readily see the passenger traffic the airline brings to other airports, with no need for any discussion with them.
Resources also drive their own loss rate. Figure 4.2 shows that an airline’s loss of customers depends on the number of staff and on the current quantity of customers themselves. It is the quantity of customers that drives the number of journeys flown, and hence the potential for poor service quality to arise if there are too few staff. In this case and others, the more customers exist, the more will be exposed to issues that may cause them to leave.
4.4 The Role of Potential Resources
Growth often hits limits because organizations run out of potential customers who are not already buying from them or their competitors. This fact has already been captured in Figure 4.3, where the number of contacts between actual and potential customers must fall when most of the potential has been captured— enthusiastic customers simply have no-one new to persuade. The same factor constrains the rate at which customers are won through other mechanisms. Marketing, for example, will win new customers more slowly as the remaining potential falls. This mechanism is well known as “diminishing returns” to advertising, but our approach allows the scale and rate of change of the mechanism to be quantified, and thus contributes to evidence-based strategic management.
Eventually, the remaining potential is so small that advertising may seem to be completely ineffective. But companies such as Coca-Cola or Heinz may still need to advertize mature products for the following reasons:
� Customers can lose interest, slipping out of the active customer stock, and will need to be recaptured.
Figure 4.3: How Ryanair’s customer win rate is increased by word of mouth
Source: Author’s estimates
T
R
P
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
77
� Demographic aging brings new potential customers each year who must be won.
� Customers may also be loyal to competitors, so marketing seeks to win them away from competitors.
� There may be potential to persuade active customers to purchase more of the product.
In Ryanair’s case, growth is so rapid because it regularly opens new routes, each of which brings with it new potential customers—the company is constantly refilling the left-hand stock of Figure 4.3. If it were to stop adding new routes, it would soon have captured just about every potential customer it can serve, and its growth would be seriously limited. The pool of potential customers is also being refilled by underlying growth in the number of customers wanting to fly and being able to afford to do so. In some cases, growth in this potential can be extremely rapid, even without any action by the business. Growth of cell phone sales in China from 2000 to 2010, for example, was fueled by a rapid increase in the number of potential subscribers, caused by rising incomes.
Limited numbers of potential staff can also cause difficulties. For example, a company that locates a customer call center in a town with high unemployment may easily be able to staff its operations until business growth requires more staff than it can hire in that locality.
4.5 The Strategic Architecture
Sections 4.2–4.4 have added the third element to the dynamic theory of performance:
1 Performance at any time depends on resources, management decisions, and external factors.
2 Resources are won and lost (accumulated and depleted) over time.
3 Resource win and loss rates also depend on existing resource levels, management decisions, and external factors.
These three elements describe the fundamental structure of how any organization functions and performs over time—its core “strategic architecture.” Chapters 5 through 10 will go on to add important extensions to these principles, but we can demonstrate the power of this core architecture alone with a simple example—a
Figure 4.3: How Ryanair’s customer win rate
T
S
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
78
consumer brand, such as a premium coffee1. We will not include production or distribution issues, but look only at the sale of products into stores and the capture of consumers. Three resources are involved (Figure 4.4):
a consumers, of whom 3 million are potentially available b retail stores, of which 5,000 are potentially available c the sales force, initially zero
Management has three controls over this situation (in red):
1 advertising spend to capture consumers’ interest
2 sales force hiring
3 wholesale price (i.e., the price charged to stores, who add a percentage mark-up when setting the price to consumers)
Here is the four-step process that produces the strategic architecture for this business, with the rigorous causal logic of each connection described. It is possible to trace out each sentence of the causal descriptions in the diagram.
1 Set out how the performance of concern is changing over time period of interest. We want to launch the product and build sales and profits over a four-year period. Profits will initially be negative, as we will have to invest in the costs of advertising and the sales force before sales and gross profit grow enough to cover those costs.
2 Lay out how performance depends on tangible resources. In this case:
Brand profit per month = gross profit − sales force cost − advertising cost Gross profit per month = sales revenue − product cost Sales revenue per month = sales volume × wholesale price Sales volume per month = consumers who want the product
× fraction who can buy it × volume bought per consumer
1 The challenge of building a consumer brand by steering the decisions in this strategic architecture can be explored or taught with the Brand Management Microworld business game (http://sdl.re/brands). Note, though, that the numerical values and relationships differ significantly from the model in Figure 4.4, so outcomes will also differ.
T
S
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
79
The fraction of consumers who can buy the product depends on the number of stores stocking it—the second resource.
Total costs are simply the sum of advertising spend (a management decision) and sales force cost, given by sales staff—the third resource— multiplied by the cost per employee.
3 Specify how resources depend, over time, on their flow rates. The number of consumers wanting the product at the start of each month is the number from the start of the previous month, plus any new consumers won, minus any consumers lost during the month. The number of stores stocking the product at the start of each month is the number from the start of the previous month, plus any new stores won, minus any stores lost during the month. The number of salespeople at the start of each month is the number from the start of the previous month, plus or minus any sales people added or lost during the month.
Consumers at start of this month = consumers at start of last month + consumers won during last month − consumers lost during last month
4 Identify how each resource flow rate depends at any time on existing resource levels (including the resource itself and any potential, undeveloped quantity of that resource that may be available), management decisions, and external factors. The number of consumers won each month equals the number won by advertising, plus the number won by seeing the product in stores and choosing to buy it. In this case, there is no word-of-mouth effect, so this portion of the consumer win rate does not depend on the current number of consumers.
The number of consumers won by advertising depends on (a) advertising spend (a management decision); (b) the fraction of potential consumers it reaches; (c) the number of those potential consumers, and (d) the fraction of those reached who become active consumers. The number won by seeing the product in stores and deciding to buy the product depends on the product’s availability, which in turn depends on the number of stores stocking the brand (a resource), and again on the remaining number of potential consumers. The number of consumers lost each month depends on the number who currently want the product, multiplied by the fraction who lose interest each month.
T
S
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
80
The number of stores won each month depends on the number of sales calls to new stores, multiplied by the fraction of those calls that are successful. The success rate depends on how many consumers want the product at that time (another resource). The number of sales calls each month to new stores is the total number of calls, minus the calls made on stores already stocking the product (a resource). The total number of calls per month equals the number of salespeople (a resource) multiplied by the number of calls each person makes in a month.
Sales calls always prioritize visits to existing stores, so there will be no loss rate of stores, unless there is a drop in sales force at any time.
N et
n ew
co ns
um er
s (in
th ou
sa nd
s pe
r m
on th
)
N ew
c on
su m
er s
fr om
a dv
er tis
in g
P ot
en tia
l co
ns um
er s
(in th
ou sa
nd s)
0 48
m on
th s
S al
es v
ol um
e (in
th ou
sa nd
s of
un its
p er
m on
th )
1, 00
0
W ho
le sa
le pr
ic e
$9 .0
0
A dv
er tis
in g
sp en
d ($
th ou
sa nd
s pe
r m
on th
) 50
0
Fr ac
tio n
of p
ot en
tia l
co ns
um er
s re
ac he
d 0.
61
3, 00
0
Fi na
lv al
ue m
on th
4 8
R ev
en ue
10 ,0
00
P ro
du ct
co st
10 ,0
00
R et
ai l
pr ic
e $1
1. 25
C on
su m
pt io
n pe
r pe
rs on
(in u
ni ts
p er
m on
th )
0. 8
P ro
du ct
c os
t pe
r un
it $7
S to
re s
w on
pe r
m on
th
P ot
en tia
l st
or es
0 24
m on
th s S
al es
pe op
le hi
re d
pe r
m on
th
S al
es pe
op le
ne ed
ed o
n ex
is tin
g st
or es
50 50
S al
es pe
op le
av ai
la bl
e to
w in
n ew
s to
re s
25 0
5
C on
su m
er s
fo rg
et tin
g
To ta
l s al
es a
nd m
ar ke
tin g
co st
1, 00
0 75
0 B
ra nd
p ro
fit ($
th ou
sa nd
s pe
r m
on th
) 1,
00 0
48 0
m on
th s
- 1 ,0
00
63 3
G ro
ss pr
of it
2, 05
0 1,
38 3
69 2
6, 22
4
5, 71
4
28 22
2 0
42
3, 57
9
S to
re s
st oc
ki ng
th e
br an
d
0 m
on th
s
5 00
0 ,
48
1, 42
1
P ro
du ct
av ai
la bi
lit y
(f ra
ct io
n) 1.
0 0.
60
20 0N
ew c
on su
m er
s fr
om p
ro du
ct av
ai la
bi lit
y 93
S al
es c
al l
su cc
es s
fr ac
tio n
1. 0
0. 55
65 14
5
20 0
20 0
20 0
13
In te
re st
ed co
ns um
er s
(in th
ou sa
nd s)
0 m
on th
s
3, 00
0
48
1, 44
8 1,
55 2
0 m
on th
s 48
10 00
81 7
S al
es fo
rc e
0 m
on th
s
50
48
50
S al
es fo
rc e
co st
25 0 (
A dv
er tis
in g
sp en
d )
50 0
S to
re m
ar k-
up 25
%
Fi gu
re 4
.4 Fi
gu re
4 .4
: T he
st ra
te gi
c ar
ch it
ec tu
re e
xp la
in in
g th
e sa
le s a
nd p
ro fi
ts o
f a n
ew c
of fe
e br
an d
N et
n ew
co ns
um er
s (in
th ou
sa nd
s pe
r m
on th
)
N ew
c on
su m
er s
fr om
a dv
er tis
in g
P ot
en tia
l co
ns um
er s
(in th
ou sa
nd s)
0 48
m on
th s
S al
es v
ol um
e (in
th ou
sa nd
s of
un its
p er
m on
th )
1, 00
0
W ho
le sa
le pr
ic e
$9 .0
0
A dv
er tis
in g
sp en
d ($
th ou
sa nd
s pe
r m
on th
) 50
0
Fr ac
tio n
of p
ot en
tia l
co ns
um er
s re
ac he
d 0.
61
3, 00
0
Fi na
lv al
ue m
on th
4 8
R ev
en ue
10 ,0
00
P ro
du ct
co st
10 ,0
00
R et
ai l
pr ic
e $1
1. 25
C on
su m
pt io
n pe
r pe
rs on
(in u
ni ts
p er
m on
th )
0. 8
P ro
du ct
c os
t pe
r un
it $7
S to
re s
w on
pe r
m on
th
P ot
en tia
l st
or es
0 24
m on
th s S
al es
pe op
le hi
re d
pe r
m on
th
S al
es pe
op le
ne ed
ed o
n ex
is tin
g st
or es
50 50
S al
es pe
op le
av ai
la bl
e to
w in
n ew
s to
re s
25 0
5
C on
su m
er s
fo rg
et tin
g
To ta
l s al
es a
nd m
ar ke
tin g
co st
1, 00
0 75
0 B
ra nd
p ro
fit ($
th ou
sa nd
s pe
r m
on th
) 1,
00 0
48 0
m on
th s
- 1 ,0
00
63 3
G ro
ss pr
of it
2, 05
0 1,
38 3
69 2
6, 22
4
5, 71
4
28 22
2 0
42
3, 57
9
S to
re s
st oc
ki ng
th e
br an
d
0 m
on th
s
5 00
0 ,
48
1, 42
1
P ro
du ct
av ai
la bi
lit y
(f ra
ct io
n) 1.
0 0.
60
20 0N
ew c
on su
m er
s fr
om p
ro du
ct av
ai la
bi lit
y 93
S al
es c
al l
su cc
es s
fr ac
tio n
1. 0
0. 55
65 14
5
20 0
20 0
20 0
13
In te
re st
ed co
ns um
er s
(in th
ou sa
nd s)
0 m
on th
s
3, 00
0
48
1, 44
8 1,
55 2
0 m
on th
s 48
10 00
81 7
S al
es fo
rc e
0 m
on th
s
50
48
50
S al
es fo
rc e
co st
25 0 (
A dv
er tis
in g
sp en
d )
50 0
S to
re m
ar k-
up 25
%
Fi gu
re 4
.4
T
S
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
81
The number of sales people added each month is simply a management decision.
Figure 4.4 may be a novel way of looking at a situation, but this too could readily be replicated in a spreadsheet. But pages of data in rows and columns can be quite impenetrable, compared to a graphic explanation, which shows how everything is connected. The time charts in this case clearly tell the story of how the launch strategy led to the performance results:
� Steady advertising won consumers at a reasonable but declining rate as the potential was used up. Spending more would have done so more quickly, but at a higher cost, which would have led to bigger losses in the early months. Reducing advertising spend in later months would have boosted profits, but at the risk of a slower consumer win rate or even a net loss of consumer.
� Stores were at first captured very slowly, both because of the small sales force and the low success rate, reflecting the limited consumer interest. After month 24, the larger sales force and increased consumer interest won stores quickly, but that rate then slowed as more sales calls were devoted to existing stores. A larger initial sales force would have captured more stores sooner, but at greater cost and greater early losses. However more stores would have increased product availability, accelerating the capture of consumers and increasing sales to those who already wanted the product but could not get it.
� The constant wholesale price determined both the average consumption rate and the profit margins made by stores and by the company. A higher wholesale price (implying a more premium product) may have either increased or decreased revenue, depending on its impact on average consumption.
The strategic architecture for Ryanair is too large to be readily displayed completely, but Figure 4.5 summarizes the core architecture resulting from following the same four-step process used in our coffee example. It is simplified for clarity, but nevertheless captures the critical features of the system driving revenues and profits. The management decisions that steer strategy from period to period are indicated in red. Note how many of those either concern flow rates directly (aircraft to buy, staff to hire), or are closely linked to a flow rate (fares affect customer win and loss rates).
Figures 4.4 and 4.5 are examples of strategic architectures for specific cases, but each is also an example of a generic architecture for many similar kinds of business. Any consumer brand will have consumers, stores, and sales force with interdependencies similar to those in Figure 4.4, and any airline will have
T
S
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
82
customers, staff, aircraft, routes, and airports organized in a structure like that in Figure 4.5. The wide variance of performance among businesses in various industries reflects vast differences among the values involved the strength of the causal relationships, and the choices made by management.
A working model of the strategic architecture for a start-up restaurant business is available in the Sysdea software help system at http://docs.sysdea.com – see Worksheet 5.
S ta
ff
A ir
po rt
s
R ou
te s
P ro
fit (€
m ill
io ns
pe r y
ea r)
S ta
ff co
st s
R ou
te co
st s
A ir
po rt
co st
s O
th er
co st
s (M
ar ke
tin g
) co
st s
A ir
cr af
t co
st s
To ta
l op
er at
in g
co st
s
To ta
l re
ve nu
e
A ve
ra ge
fa re
P as
se ng
er jo
ur ne
ys so
ld
A ir
cr af
t
C ho
ic e
of de
st in
at io
ns Fl
ig ht
s pe
r da
y
A ir
cr af
t ut
ili sa
tio n
D el
ay s
S er
vi ce
qu al
ity Jo ur
ne ys
pe r
pe rs
on pe
r ye
ar
C us
to m
er s
C us
to m
er s
w on
a nd
lo st
(S ta
ff)
(A irc
ra ft)
M ar
ke tin
g
Fu el
co st
s M
ill io
ns of
ki lo
m et
er s
flo w
n
H ir
in g
A dd
in g
ai rc
ra ft
O pe
ni ng
ro ut
es
A dd
in g
ai rp
or ts
M an
ag em
en t
de ci
si on
sa re
in di
ca te
d in
r ed
.
Fi gu
re 4
.5 Fi
gu re
4 .5
: T he
c or
e st
ra te
gi c
ar ch
it ec
tu re
o f R
ya na
ir
S ta
ff
A ir
po rt
s
R ou
te s
P ro
fit (€
m ill
io ns
pe r y
ea r)
S ta
ff co
st s
R ou
te co
st s
A ir
po rt
co st
s O
th er
co st
s (M
ar ke
tin g
) co
st s
A ir
cr af
t co
st s
To ta
l op
er at
in g
co st
s
To ta
l re
ve nu
e
A ve
ra ge
fa re
P as
se ng
er jo
ur ne
ys so
ld
A ir
cr af
t
C ho
ic e
of de
st in
at io
ns Fl
ig ht
s pe
r da
y
A ir
cr af
t ut
ili sa
tio n
D el
ay s
S er
vi ce
qu al
ity Jo ur
ne ys
pe r
pe rs
on pe
r ye
ar
C us
to m
er s
C us
to m
er s
w on
a nd
lo st
(S ta
ff)
(A irc
ra ft)
M ar
ke tin
g
Fu el
co st
s M
ill io
ns of
ki lo
m et
er s
flo w
n
H ir
in g
A dd
in g
ai rc
ra ft
O pe
ni ng
ro ut
es
A dd
in g
ai rp
or ts
M an
ag em
en t
de ci
si on
sa re
in di
ca te
d in
r ed
.
Fi gu
re 4
.5
T
S
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
83
4.6 Functional Issues and Other Objectives
This chapter has explained how the strategic architecture drives overall performance in terms of financial outcomes for an entire commercial organization. As explained in Chapter 1, this strategic architecture should focus on delivering a strong trajectory of future profits, or more strictly, cash flows.
The strategic architecture principles also apply to nonfinancial objectives and to functional issues with more localized aims. In Figure 4.6, a service department struggles to sustain good service quality for rising numbers of customers. This service quality should be expressed in terms that are relevant to the customers, such as the frequency of problems they experience. The demand being served in this case is calls per day, and is driven by the number of customers (a resource), multiplied by the frequency with which they need service. The capacity to deliver good service quality depends on the ability to deal with customer calls, which in turn depends on the number of staff (another resource), multiplied by the number of calls per day each can resolve.
Here, interdependency between the resources is not complicated. The customer win rate is not something the service department can influence, but the loss rate is affected by the level of service performance. Staff hiring is directly under their control, while staff losses depend on the pressure the service staff are under.
Exactly the same principles can be used to develop a strategic architecture for public services, voluntary groups, or not-for-profit organizations. The example in Figure 2.5 shows how the performance of a charity changes over time, as indicated by the fraction of required calls it can fulfill. Figure 3.4 shows how to add the inflows and outflows of patients who need the charity’s support, and, like the service department discussed above, it is a simple matter to add the flows of volunteers for this charity. This will allow the organization to anticipate performance changes and make appropriate changes to its strategy.
Section 2.8 pointed out that some objectives concern achieving a certain level of a resource by a certain time—capturing a target number of customers, or hiring a certain number of employees, for example. These cases are much simpler than those concerning performance outcomes, because analysis and activity need only focus on the flow rates for the single resource of concern—how quickly customers are being won and lost, or how quickly staff are being hired and leaving. Nevertheless, the principles in this section still apply. Those flow rates depend on management decisions (e.g., marketing spend or salaries offered), on external factors (e.g., competitors’ prices or salaries), and on existing resource levels.
F
I
O
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
84
Strategic plans versus strategic issues. Chapter 1 noted that strategic management often needs to address specific issues, as well as develop comprehensive plans for a whole organization. Figure 4.6 provides such an example. It is important to remember this point when considering the frameworks presented in Chapters 5 through 10; each can be valuable on its own or as part of a strategy for an entire organization.
Fi gu
re 4
.6 : T
he st
ra te
gi c
ar ch
it ec
tu re
o f a
fu nc
ti on
al is
su e
F
I
O
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
S
A
85
Important Detail: Tipping points and thresholds
Situations where modest changes trigger sudden rapid growth are popularly known as tipping points, but we need to be clear about what exactly causes such phenomena.
Reinforcing feedback itself does not indicate a tipping point, and if, for example, customer numbers grow from 10 to 50 to 200 to 700 in successive quarters, the growth rate is actually falling. A true tipping point happens when a modest growth rate suddenly escalates. At least four mechanisms can bring this about:
1 Segmentation. A market consisting of a large, low-growth segment and an initially smaller, high-growth segment will take off when the faster-growing segment overtakes the slower one.
2 Role models or opinion leaders. A market may grow quite slowly until certain key individuals buy the product, making members of the general population suddenly want to buy the product, as well. For example, doctors use procedures they know to be reliable until they see experts endorsing a new treatment, which leads to a sharp increase in the use of the new treatment. The same clearly happens in fashion businesses, which go to great lengths to gain celebrity endorsement for products.
3 When win rates overtake losses. If a customer base of 100 is decreasing by a rate of 20 per quarter, but is also growing at 20 per quarter, growth is zero. A small increase in the win-= rate—which could be due to word of mouth or other factors—results in rapid positive growth.
4 Thresholds. If a product is not especially appealing to potential customers, growth may be slow, and remain so for some time, even though it is being steadily improved. Flat-screen televisions, for example, did not sell well for many years. When the products’ ratings crossed a threshold of acceptability, though, the rate at which potential customers bought them (and became actual customers) jumped sharply.
All these mechanisms can be identified in real cases, and quantified, enabling important strategic choices to be made by management—whether they should focus on developing promising market segments, capturing opinion leaders, reducing customer losses, or improving the product.
F
I
O
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
86
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 5 RESOURCE QUALITY
Chapters 1 to 4 assumed for simplicity that resources were uniform. Customers all purchased at the same rate, staff had equal skills, products were equally appealing. In reality, some customers buy more than others, staff have varying skill levels, and products differ in their appeal. A sound explanation of performance therefore requires these characteristics to be included in the analysis and planning of strategy. Most tangible resources have one or more such qualities—referred to as “attributes”—and these affect their impact on the business system and its performance.
The way an attribute changes over time can be understood by extending the bathtub analogy from Section 3.1. The resource itself is the quantity of water in the bath, and the attribute of interest is its temperature. If you want a warmer bath, you can add hot water. The initial temperature is the initial quantity of heat divided by the amount of water that it warms, and the final temperature is the new, greater quantity of heat divided by the now larger quantity of water.
An important difference between business resources and the bathtub example is that, for businesses, the different qualities of resource remain separate. If a firm with small customers wins larger ones, it ends up with some small and some large, not a uniform population of mid-sized customers. Nevertheless, in many cases it is possible to understand and improve performance by working with the average quality of a resource. Figure 5.1 shows how the math works—a company’s average customer size increases as it wins customers three times larger than it started with:
� At the start, the company has 240 customers, each bringing sales of 35 units per month, so total sales = 8,400 units per month.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
88
� After one month it has 245 customers, and total sales have been increased by 5 × 105 = 525 units per month. Therefore, total sales = 8,925 units per month, and the average customer size is 8,925 ÷ 245 = 36.4 units per month.
This mechanism is known as the “co-flow” structure, because the attribute flows along with the resource that it describes. The same logic applies to working out the impact of losing customers that are larger or smaller than average.
Instead of winning larger customers, the business in Figure 5.1 could improve the quality of its customer base by closing smaller customers, say those with average sales of only 10 units per month. It need not do so steadily over time, of course. It could cease doing business with them all at one time, perhaps by passing their accounts on to distributors. If it identified 100 such customers, then their loss would cut direct sales by 1,000 units per month. After the change, it would have 140 customers delivering total sales of 7,400 units per month, so the average remaining customer would be generating 7,400 ÷ 140 = 52.8 units per month. There is a third way to improve customer quality, of course, and that is to sell more to the existing customers, either by capturing sales they would otherwise have given to competitors or by helping them grow.
Improving a customer base then, as measured by average sales, has three generic solutions:
1 Win bigger customers 2 Lose smaller customers 3 Increase the size of existing customers
Businesses of all kinds face questions about how they want their business to develop. It is very common, for example, for companies to have too many small
5
New customers per month
Customers
525 New sales
each month units per month
Total sales units per month
105 New sales per new customer
units per month 58.33
25000
400
21000
360
8400
240
60
35
Month0 24
Initial value
Final value
Month0 24 Average sales per customer
units per month
Figure 4.6 Figure 5.1: How winning larger customers increases average customer size
5
New customers per month
Customers
525 New sales
each month units per month
Total sales units per month
105 New sales per new customer
units per month 58.33
25000
400
21000
360
8400
240
60
35
Month0 24
Initial value
Final value
Month0 24 Average sales per customer
units per month
Figure 4.6
H
R
D
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
89
customers who are too costly to serve, while others focus on the largest customers only. Some companies, on the other hand, recognize that they cannot compete for the largest customers, and develop ways of doing business that are well-suited to the needs of mid-size to smaller customers.
5.1 Size Is Not the Same as Quality
Larger customers are not always more profitable than smaller ones. They may demand lower prices or be more costly to serve, so it is important to understand the relationship between size and profitability. This is best done with a device called the “quality curve,” which lays out the quality profile of individual customers. In the top half of Figure 5.2, the revenue from a company’s largest customer is shown at the left, and to this is added the revenue from the second largest, the third largest, and so on. At the far right is the very small revenue contributed from the company’s smallest customer. The average revenue per
customer is the total revenue, A, divided by the total number of customers, B.
The lower part of Figure 5.2 shows the differing customer profile in terms of quality, rather than sheer size. The same curve is built, but this time using the profit contribution from each customer, rather than their revenue. With this information, a management team can make explicit choices about how they want their business to develop, such as:
Lowest-revenue customer
Cumulative revenue
A
B
etc.
Cumulative customers
Cumulative profit
contribution
B Most profitable customer
Cumulative customers
Most unprofitable
customer
Total profit contribution from all customers
Total profit contribution from all profitable customers
D
E
C
Figure 5.2
Lowest-revenue customer
Cumulative revenue
A
B
etc.
Cumulative customers
Cumulative profit
contribution
B Most profitable customer
Cumulative customers
Most unprofitable
customer
Total profit contribution from all customers
Total profit contribution from all profitable customers
D
E
C
Figure 5.2
Figure 5.2: The resource quality curve—customers’ contribution to total revenue and profits
S
I N
S
Q
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
90
� Just grow in total, by trying to win more customers across the entire profile.
� Sort out the unprofitable tail, by shifting customers from right to left (e.g., by raising prices or cutting support costs), or closing them where this cannot be done.
� Focus on winning only larger new customers.
� Develop a lower-cost business model to improve the profitability of mid-range customers.
The message is to be deliberate about the choice of strategy, quantify its impact, assess the time scale over which the impact of strategy will occur, and track its progress using a combination of the methods used in Figures 5.1 and 5.2.
5.2 Attributes of Other Resources
Customers are a common example of resources whose quality is important to understand and manage. Quality curves can be useful for other resources, as well, such as the profit contribution from each product in a product range1. In one confectionery market, for example, a competitor of Mars Inc. had a range of over thirty products, but was much less successful, which offered a much smaller range in the same market. For every dollar this company spent to advertise one of its brands, Mars could spend double, leading to persistently stronger sales. The
1 See for example: Nikhil Bahadur, Edward Landry, and Steven Treppo, How to slim down a brand portfolio, Strategy+Business, 44, 14–16. Also available at http://www.strategy-business.com/article/06315 (registration required).
Important Detail: Take care with cost allocations
The analysis in Figure 5.2 can be sensitive to exactly how costs are allocated. Note that the vertical axis is “profit contribution,” not simply “profit,” to recognize that some costs are not driven by specific customers but are incurred across the customer base (e.g., the cost of the sales vice-president’s salary or the customer information systems). Closing customers between D and B will not remove the need for such costs. It is therefore important to know by how much total costs will actually be cut if such a rationalization is to be carried out. Similar caution is needed if the curve is built for the products in a product range. Many shared costs will not disappear if the product range is reduced.
A
O
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
91
company with the wider range also found that each new product it launched only captured sales by cannibalizing existing products.
A contrasting case concerns a commercial law firm whose services require specific skills and experience. The firm responds to many client requests on a case-by-case basis, although many of these requests are quite similar. These similar cases could be codified as standard services, making each client contract more efficient, reliable, and profitable. Furthermore, the defined service could be marketed more explicitly and used to capture many new clients.
You can also adapt Figures 5.1 and 5.2 to understand and improve staff team performance. This can be valuable at all levels, from front-line operatives to senior executives1. In Figure 5.3, a call center manager has 80 staff, each of whom needs certain skills to deal adequately with customers’ enquiries. Ideally, each employee should be able to respond to the full range of twenty issues, but that ideal is hard to achieve because of staff turnover, which is running at five per month. Initially, the average employee knows how to respond to twelve of the twenty issues. The management decisions are shown in red—how many staff to hire and how much training to do.
New staff have none of the necessary skills (we can only add cold water to this particular bathtub!), so training gives new staff some basic skills, and increases the skills of existing staff. Staff also forget some skills they were trained in, because issues arise too seldom to be reinforced through practice. With no staff turnover (dashed lines) and no hiring, training only builds skills among existing staff. Improvement slows because learning is overtaken by the rate of forgetting (the hotter the water in the bathtub becomes, the faster it cools down). Staff turnover (solid line and bold text), raises the rate at which skills are lost (warm water is lost, to be replaced by the cold water of unskilled staff).
One technical point to be aware of is that the resource attribute (the lower stock) tracks the total skills of the team, rather than the average. Its units are “person skills” which may seem an odd concept, but it is the only easy way to do the necessary calculations. A closely related issue concerns staff experience, measures of which may either be the number of years’ employment in an industry, or more specifically, the time spent in a particular role or with a particular employer.
1 Linda Gratton. People processes as a source of competitive advantage, in Strategic Human Resource Management, Linda Gratton, Veronica Hope Hailey, Philip Stiles, and Catherine Truss, eds. (Oxford: Oxford University Press, 1999), 170–198. Laura Tovey, Competency assessment: A strategic approach—part II, Executive Development, 7(1), 16–19.
A
O
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
92
Experience is a critical resource quality in a wide variety of cases. Professional athletes commonly improve their skills with each season of play1. Staff also carry experience with them as they move up through levels of seniority, a mechanism we’ll examine in Chapter 6.
Staff skill is one example of a resource attribute declining if not constantly reinforced. Another such case concerns equipment reliability, a concern for organizations as diverse as water companies, whose pumps and other equipment wear out, hotels and restaurants whose furnishings and fittings
1 The Football League Challenge is a fun and insightful class game in which the dynamics of player experience features strongly—see http://sdl.re/football.
Fi gu
re 5
.3 : S
ki ll
bu ild
in g
an d
lo ss
es in
a c
al l c
en te
r w
it h
an d
w it
ho ut
st af
f t ur
no ve
r
A w
or ki
ng m
od el
d em
on st
ra tin
g th
e re
la tio
ns hi
p be
tw ee
n ch
an gi
ng st
af f n
um be
rs a
nd ex
pe ri
en ce
(r el
at ed
to , b
ut d
iff er
en t f
ro m
sk ill
s) c
an b
e ex
pl or
ed a
th tt
p: //
sd l.r
e/ m
hg s.
T hi
s i s a
n ex
am pl
e of
w or
ki ng
m od
el s i
nc lu
de d
in th
e st
ra te
gy d
yn am
ic s c
ou rs
e.
A
O
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
93
become damaged, and public infrastructure, such as roads and bridges. Some firms depend so heavily on such assets that their upkeep is the dominant strategic issue, where choice of maintenance, repair, and replacement are all critical. In such cases, the tangible resource is the population of assets, and the co-flow attribute is some measure of their state of repair, such as failure rate.
Similar analysis is useful in public service and voluntary sectors. Charitable organizations receive varying contributions from individual donors and make explicit choices about whether to target fewer, bigger donors or larger numbers of smaller donors. The voluntary organization described in Figure 2.5 found that the activity of their volunteers was highly skewed. A small fraction of volunteers were very active, supporting many patients, while most were making very few visits. Training was costly, so the organization became more selective in accepting volunteers and provided more intense encouragement and support to these smaller numbers. The result was more care, of a consistently higher quality, delivered by fewer volunteers, at much lower cost.
5.3 When Resources Bring Access to Others
An important category of attributes arises when one resource brings access to another. Each of Starbucks’ new stores, for example, gives access to potential customers, and the challenges described in Section 1.4 arose from misjudging this issue. The company simply opened too many new stores with too few new customers.
In Figure 5.4, an initially successful retailer is expanding its store network over ten years1. Its first stores each reach 20,000 consumers who spend $500 per year on its goods, so management believes much growth is possible. For the first five years, plans go well, with sales and profits growing strongly. Unfortunately, these top-line indicators hide a sharp fall in the true number of new consumers won with each new store opening, and new stores increasingly succeed only by taking sales away from established stores. Furthermore, the newer consumers turn out to spend less at the stores than those captured around the initial locations. (This spending rate is an attribute of the consumers, omitted here for simplicity). The costs of operating these newer stores, despite being lower than the costs of the earlier stores, are not covered by the incremental revenues and gross profit from the sharply declining rate of new consumers, and profits go into decline.
1 The challenge of building a retail business while the quality of potential locations changes can be explored or taught as one of the features in the Beefeater Restaurants Microworld business game, see http://sdl.re/beefeater
A
O
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
94
Had this company stopped expanding when the extra consumers won with each new store dropped sharply ( as shown in gray text about half-way through its expansion in year 6), it would have captured most of the potential market, kept profits at $32 million per year, and spent much less capital.
This is not to say that management should give up at the first sign of having “used up” a business opportunity. For example, some retailers develop slimmed-down units, offering limited product ranges with much lower operating costs precisely so they can reach smaller local markets.
Fi gu
re 5
.4 : S
al es
a nd
p ro
fi ts
o f a
r et
ai le
r ex
ce ed
in g
th e
vi ab
le p
ot en
ti al
fo r
st or
e nu
m be
rs
10 0
P ot
en tia
l co
ns um
er s
no t y
et w
on (th
ou sa
nd s)S
to re
s
P ot
en tia
l n ew
co ns
um er
s pe
r ne
w s
to re
‘0 00
Y ea
r 0
10
N ew
st or
es pe
r ye
ar 50
0
20
To ta
ln ew
p ot
en tia
l co
ns um
er s
‘0 00
/y ea
r
Y ea
r 0
10
C on
su m
er s
w on
‘0 00
/y ea
r
10 00
20 00
A ct
iv e
co ns
um er
s (th
ou sa
nd s)
Y ea
r 0
10
50 00
S al
es p
er co
ns um
er $/
ye ar
To ta
l s al
es $m
/y ea
r
Fr ac
tio n
of n
ew po
te nt
ia l c
on su
m er
s w
on im
m ed
ia te
ly ..
0. 5
Fr ac
tio n
of r
em ai
ni ng
po te
nt ia
l c on
su m
er s
w on
p er
y ea
r. .0
.3
20 00
[ A pp
ea l a
nd v
al ue
of p
ro du
ct s
of fe
re d
]
G ro
ss m
ar gi
n fr
ac tio
n ..
0. 20
G ro
ss p
ro fit
$m /y
ea r
40 0
O pe
ra tin
g pr
of it
$m /y
ea r
50 O
pe ra
tin g
co st
s $m
/y ea
r 40
0
41 85
14 00
10 00
45 0
50 0
33 5
28 0
9 27
1
32 47
6
20
0. 8
26 1
26 90
80
Fi gu
re 5
.4
A
O
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
95
Ryanair’s airline business, which we have been following as we develop the strategy dynamics frameworks, features a very similar mechanism—the opening of new airports and routes brings access to new potential customers1. When the airline starts services to a particular city, it gains access to the people in the region of that city who might want to travel. Each new route it offers to and from that airport attracts some more of that region’s potential travelers, as well as appealing to a few more travelers from each of the regions at the other end of those routes. In addition, closing routes removes access to some customers—an advisable decision if a route has less potential than expected. Figure 5.5 shows a simplified summary of these relationships, ignoring (a) the distinction between expanding the number of airports it serves and adding new routes, and (b) the conversion of potential travelers to actual customers.
5.4 Using the Quality Curve to Beat Competitors
Competitive strategy is rarely done well, in spite of all the articles and books on the topic. Mostly, firms try to capture little bits of market share by being generally a bit better than rivals. This has several disadvantages:
1 This mechanism is included in the LoFare Airline Microworld busines�s game mentioned in Chapter 4, see http://sdl.re/lofare.
Figure 5.5: How adding and closing routes adds and loses customers for Ryanair
Source: Company reports and author’s estimates.
A
O
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
96
� Effort spread across the whole will have less impact than if focused on specific parts of the market or against specific competitors.
� Trying to under-price, out-sell, or out-service all competitors will incur considerable cost.
� Industry-wide competitive efforts will be visible and will attract retaliation from all competitors.
� Such generalized efforts exacerbate the very competitive conditions that damage profitability, for example by triggering price wars, or escalating advertising commitments.
Selecting specific competitors to attack is more advisable than making indiscriminate efforts. It is even possible, and often desirable, to seek the total elimination of a competitor, but this requires a deep understanding of individual rivals. A more rigorous approach to competitive strategy builds on two principles:
1 A close competitor will likely have both a similar set of resources and a similar architecture to one’s own business.
2 Each competitor will, like one’s own business, have a range of attributes (quality) for each resource—larger and smaller customers, more and less popular products, stronger and weaker salespeople.
Investigation of these two elements makes it possible to evaluate any competitor’s performance and to use that insight to inflict damage at little risk to ourselves.
One mid-market restaurant chain, growing fast in a promising market, was number-two to a long-established market leader. Although operating only 120 restaurants compared with the leader’s more than 300, it was generating nearly as much profit, due to a recent history of finding great locations and developing better products. With a good understanding of the profit contribution profile of its own restaurants, it was able to estimate the same profile for the competitor. Revenue could be determined by counting the customers visiting each of the rival’s restaurants, and costs could be estimated by comparison with the company’s own units. The resulting estimation was not exact, of course, but close enough to know roughly how much profit was coming from each of the competitor’s units.
Selecting the point of attack was relatively simple (Figure 5.6). Trying to damage their most profitable restaurants would be difficult—they were popular with their local consumers, well-run, and received regular attention from senior management. Any attack on these would have been noticed and vigorously defended.
U
Q
C
B
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
97
Attacking unprofitable restaurants was pointless, as the competitor would not be concerned or damaged by their loss. The best targets were the restaurants contributing profits in the mid-range. A random selection of these were chosen, geographically dispersed and supervised by different regional managers, so that attacks would not be noticed, provided that the tactics were subtle.
The tactics were not led by price cuts, since this would have hit profit margins hard and been highly visible to the competitor. Instead, marketing promotions were aimed at specific neighborhoods from which each competing restaurant drew its customers. Local tactics also included overstaffing their own restaurants, deploying the best unit managers, and providing the best possible customer e n v i r o n m e n t — d é c o r , lighting, and music.
These tactics took a large fraction of the revenue from the competitor’s targeted units, at which point the competitor’s own policies started to act against them. Management tried to sustain profits by cutting costs, especially staffing costs, which damaged consumers’ experiences. With the targeted restaurants becoming rather quiet, they lost still more consumers. Eventually, the targeted units moved into losses or marginal profits, following which they were neglected by management until they were closed.
Repeating these tactics across a selection of mid-profit outlets inflicted disproportionate damage to the competitor’s overall profits, cutting sharply the right-hand end of their profit curve (see Figure 5.6). The pressure to sustain profits drove them into system-wide policies that did further damage, such as price discounting and cuts in staffing, marketing budgets, product development, and maintenance, all of which undermined critical resources in their strategic architecture. Their management started to lose motivation and commitment, and many left for better opportunities—often with the attacker! The competitor’s corporate owners turned down requests for investment, making it impossible to match the high-quality new units that were being added to the attacker’s business.
Cumulative profit
contribution $millions
Cumulative restaurants 100 200 300
50
100
Competitor’s business under attack
losing 11 profitable restaurants (3%) hits competitor’s
profits by $11m (12%)
$92m
$81m
Figure 5.6 Figure 5.6: Damaging the profit contribution curve for a mid-market restaurant competitor
Cumulative profit
contribution $millions
Cumulative restaurants 100 200 300
50
100
Competitor’s business under attack
losing 11 profitable restaurants (3%) hits competitor’s
profits by $11m (12%)
$92m
$81m
Figure 5.6
U
Q
C
B
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
R
Q
98
The competitor left the industry after just three years of the competitive strategy being implemented, selling most of its remaining viable restaurants to the one-time number-two, who was now in a dominant position.
This is just one illustration of how to use the quality curve to undermine competitors by damaging the quality of its resources. In other cases, such tactics have been deployed to hit the profit contribution of competitors’ mid-range customers or products, or to capture key staff. In this case the chosen target was the market leader, but often that is not feasible or is too risky, so in other cases it may be preferable to target competitors who are large enough to be worth the effort, but not so large as to threaten serious retaliation.
5.5 Other Uses for the Quality Curve and Resource Attributes
This chapter has outlined some basic uses for the analysis of change to resource attributes. Other uses include:
� assessing the development of product features; for example, to avoid feature fatigue when consumers become overwhelmed with features they can neither understand nor use
� strategic turn-round to rescue an organization in difficulty, commonly due to having over-built low-quality resources, such as unprofitable customers and products. (This was a widespread feature of troubled companies following both the 2001 and 2009 recessions.)
These and other issues are developed extensively in Chapter 5 of Strategic Management Dynamics.
O
U
Q
C
R
A
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 6 DEVELOPING RESOURCES
As with the attribute analysis in Chapter 5, the frameworks in this chapter can be used on their own to improve business performance or as part of a complete strategy. So far, we have assumed that resources are simply won or lost, in or out of the business system. A more realistic picture of this process shows that resources can develop through a series of states.
6.1 Developing Staff
Figure 6.1 shows how a law firm hires lawyers and promotes some to partner through a simple two-stage chain. The problem here is the rising attrition of lawyers at the bottom of the diagram, caused by the slow rate of promotion.
The same rigorous math we saw in Chapter 3 for how resources accumulate applies equally to each state in this chain. The number of 180 lawyers is entirely explained by the firm’s history of all lawyers who were ever hired, promoted, and lost. Similarly, the number of twenty-eight partners at quarter 12 is precisely explained by the sum of all lawyers ever promoted, minus all partners who ever left. There could also be a third flow affecting partner numbers—experienced lawyers hired directly into the partner grade from outside—but we will not be including this possibility in this example.
Note once again that the flow rates highlighted in yellow totally explain the number of lawyers and partners at each point in time. The red items indicate the critical decisions of the firm’s leadership that have led to the current situation and could lead to the plausible situation by quarter 20.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
100
This stage-wise development of resources has important implications, since it extends the timescale over which strategic choices operate. In Figure 6.1, no more than four percent of lawyers had ever been promoted in any quarter (and then, only in quarter 3, when four out of 107 were promoted, or about one in twenty-seven). A newly hired lawyer would therefore to have to wait on average twenty-seven quarters, or nearly seven years, before being promoted. If the firm had hired no new staff six years ago, then there would be no new partners now with seven years’ experience. A resulting shortage of partners could cut the firm’s ability to win clients, so the resulting drop in performance would have been caused by the firm’s hiring failure six years previously!
Such diagrams provide a powerful basis for assessing strategy, since they include both management choices—in this example, rates of hiring and promotion (in red)—and their consequences. Here, a rate of promotion that is too slow causes lawyers to leave the firm too quickly. This picture can be readily integrated with other parts of the business system, such as the winning of clients and delivery of client work. It can also be linked to the resulting staff costs in order to work out the resulting financial implications.
Note also that resources carry with them from stage to stage those same attributes we looked at in Chapter 5. A lawyer with seven years’ experience in the first level
Figure 6.1: Two-level staff chain in a law firm 1
1 Strategic management of staff development in a consultancy business can be explored or taught with the Professional Services Microworld business game, see http://sdl.re/profserv.
D
S
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
101
here will add those years of experience to the partner group when she is promoted, and a partner with twenty years’ experience will take that with her when she leaves. If both events coincide, the average experience of the partner group will fall, but will of course increase again each year as staff age.
6.2 The Customer Choice Pipeline
Customers are rarely switched from “potential” to “active” in the simple way described in Section 4.4. Most often, they must be moved through a number of stages. A simple model of customer development, often referred to as “AIDA,” has four stages: gaining customer attention, attracting their interest, stimulating desire, and finally motivating them to act (i.e., to buy the product).
These pre-purchase states are important. First, marketing expenses may be quite high in order to bring customers through those stages before you get any sales from them. Second, interactions among not-yet-active customers can have important effects on customer acquisition. Many people who admire BMW cars, despite never having owned one, influence other potential owners positively. Conversely, people with negative attitudes towards fuel-inefficient SUVs, again without having owned one themselves, can have a negative impact on the attitudes of other drivers.
Figure 6.2 shows a customer-development chain for the 36-month product launch of a consumer product, such as the coffee brand discussed in Chapter 4. The marketing challenge is to move consumers through a series of stages until they become buyers of the product. Marketing tries to influence their choice toward the brand—hence the framework’s name, the “choice pipeline.”1
There are fifty million potential consumers who need to be won by marketing and promotional spending. This particular chain combines the interest and desire stages of the AIDA framework into a single population of interested consumers. However, it divides active customers into two groups: “loyal” consumers, who always purchase the brand when they need this kind of product, and “disloyal” consumers, who divide their purchases between this brand and its competitors.
Management has a total marketing budget of $20 million per month to split across four categories of spending: (1) advertising to build awareness; (2) advertising to communicate the brand’s “values” and win consumers’ interest; (3) promotions
1 Lars Finskud. Developing Winning Brand Strategies. (Williston, VT: Business Expert Press, 2009).
T
C
C
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
102
to persuade interested consumers to add the brand to those they purchase; and (4) loyalty promotions to persuade consumers to purchase only this brand.
In the base case (dashed lines of graphs), the budget is allocated evenly, so $5 million is spent on each of the four activities in every month. It takes a long time for the early advertising at the front end of the chain to bring consumers within reach of the promotion spending that persuades them to actually buy the product. In this case, sales are slow to take off, the brand does not break even until month 20, and by month 36, the brand has failed to pay back the spend on its own marketing—cumulative profits are still negative.
Fi gu
re 6
.2 : P
ha si
ng m
ar ke
ti ng
sp en
d pr
io ri
ti es
to a
cc el
er at
e pr
of it
g ro
w th
fo r
a co
ns um
er br
an d. A
w or
ki ng
m od
el o
f a c
on su
m er
b ra
nd la
un ch
b as
ed o
n th
is st
ru ct
ur e
is a
th tt
p: //
sd l.r
e/ m
kc w
. T
hi s i
s a n
ex am
pl e
of w
or ki
ng m
od el
s i nc
lu de
d in
th e
st ra
te gy
d yn
am ic
s c ou
rs e.
T
C
C
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
103
In the better case (solid lines of graphs), awareness advertising is prioritized first, followed by values advertising, in the first year. This pumps consumers more quickly into the aware and interested states, so when spending switches to promotion from month 12, it is much more effective. Spending at the front of the chain can then be cut, since few people remain to be captured, and can be increasingly focused on activating customers and winning their loyalty. Total active consumers (disloyal plus loyal) climb to roughly the same number as in the base case, 37 million, but a larger fraction are now loyal. Also, the total between months 18 and 30 is much greater than in the base case, growing profits faster during the same period and easily paying back the investment by month 36.
This framework is not limited to consumer markets, or even to commercial cases. New business-to-business (B2B) products and services must also pull potential customers through stages, and have their own mechanisms for achieving these flows, such as trade shows and direct sales calls. Similarly, charitable organizations work hard to win attention and participation from potential donors.
The framework is so useful in a wide variety of cases because it captures how people move between states of mind, a phenomenon that arises in numerous situations, including politics and organizational change. A particularly important case is the slow adoption of actions to reduce carbon dioxide emissions. Contrary to popular belief, most emissions can be eliminated by existing known technology that is simply not being deployed. Nor is it a costly endeavor—some twenty-five percent of global emissions can be eliminated at a profit (i.e., savings exceed costs), and well over half of all emissions can be cut at no net cost to the economy1. The problem is neither technology nor cost, but adoption—people and organizations are unaware and uninformed about what can be done, which suggests that governments wishing to make faster progress in reducing greenhouse gas emissions should make advertising of information about the issue a priority.
Figure 6.2 simplifies greatly how marketing really works:
� It is unlikely that roughly comparable spending on four types of marketing would be advisable in any real case.
� Each type of marketing will impact on more than a single flow of customers—e.g., values advertising will help build both awareness and interest.
1 For more information, see http://www.mckinsey.com/ Article : Greenhouse gas abatement cost curves
T
C
C
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
104
� Multiple consumer segments may exist, each a different size, purchase rate, and so on, and each will respond uniquely to marketing efforts. This variability can be dealt with simply by replicating the consumer chain with the relevant data for each segment and adding up the resulting sales and profits.
� There may be word-of-mouth effects, where already-won customers at the top of the pipeline pull others up through earlier stages, causing faster growth than marketing alone can deliver.
� Potential customers may be divided among “early adopters,” who will try anything new, others who start buying the product after they see its benefits, and late adopters, who only buy long after the majority have done so. This is again dealt with by segmenting these groups, and capturing any word-of-mouth effects both within each group and among groups.
� The chain does not show all possible movements—e.g., the possibility of consumers moving directly into upper levels of the chain with spontaneous purchases.
� There may be different or additional states into which customers may move, especially in terms of “rejectors,” who positively dislike the product, and try to stop others liking and purchasing it.
Even so, this model illustrates important general issues.
The difference between good and ineffective strategy is massive. In the example from Figure 6.2, the brand does not just deliver a few percentage points of extra margin in the better case, but hugely positive cash flows.
It may be best to change decisions substantially over time. A constant marketing spend over many periods could never be the best answer. It is worse still to decide spending amounts as a fraction of sales revenue—the company would spend nothing at first, and overspend later on.
The missed potential may never be known. Had management followed the first strategy (base case) for the brand, they may have been happy enough that it eventually became profitable, never realizing that they could have made over $260 million more using a better strategy. Such massive strategic underachievement is widespread.
Strategic management is all about the flow rates! Chapter 2 explained how performance depends on resources; therefore (other things being equal) if resource levels don’t change, performance will stay the same. Changing performance must, then, mean managing flow rates.
T
C
C
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
105
Some minimum input is needed for the system to perform at all. Had the company invested only, say, $5 million per month in the brand in total, rather than $20 million, it would not have achieved one-quarter of the sales and profits. It would have achieved very low sales and continued losing money indefinitely. This is a common phenomenon in strategy—a perfectly attractive and feasible opportunity fails to be taken because management will not commit what is needed.
There is some maximum performance that the system can deliver. There are finite resources here, and increasingly powerful mechanisms preventing indefinite growth. The brand might just generate $500 million per month in profit, but not $1 billion or $2 billion. Only by adding to the resource potential could this limit be raised. Lack of awareness regarding the finite nature of resources is a common strategy error, as shown by the over-expansion of stores by Starbucks discussed in Chapter 1.
6.3 Product Development
Product development is another resource-development process, the typical stages being:
1 idea generation, making use of basic R&D, competitors’ products, customer focus groups or direct requests, employee suggestions, and so on.
2 initial screening, where the basic technological feasibility and market potential are assessed.
3 technical development, which includes the product’s initial specification and testing.
4 commercial evaluation, including market research, assessment of likely sales, prices, and revenues, and estimation of production costs and the capital investment needed.
5 final development, when the product takes the form that customers will actually see, and the details of the production process are specified.
6 product launch, when marketing and sales activity start, sales are generated and the product is shipped.
The exact stages involved and the activities in each vary considerably from industry to industry. It is also common for companies to run stages in parallel in order to collapse the time between initial idea creation and product launch—for
P
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
106
example, production engineering may run in parallel with final product design and market testing1.
6.4 Deteriorating Resources
Staff and products are cases where we usually expect quality to increase as an item (each person or product) moves from stage to stage. Some assets, though, degrade over time, and it is not usually adequate to treat the asset base as a single population, as we implied in Chapter 5. Instead, we need to distinguish units that are in different stages of their life.
Figure 6.3 shows such a picture for a utility company’s assets. New units spend some time bedding in, then spend many years in a highly reliable state. They then start to deteriorate, when failures become more frequent, until finally they become very unreliable indeed. The company makes revenue for each unit of power it transmits (assumed to remain stable over time), but the price it receives is penalized by an industry regulator for each power failure that occurs.
The business starts with a poor asset base that continues to deteriorate for the first five years. Maintenance spending on reliable assets is not sufficient to keep the asset base at its initial level. Maintenance and refurbishment of degenerating units are not enough to prevent a rapid flow of units into the unreliable state. (“Refurbishment” means taking a unit out of service, replacing worn-out parts and restoring it to an as-new state.) Spend on new equipment is too slow to prevent the number of unreliable units growing quickly, so total failures are growing, which hits the company’s revenue. Falling revenue and rising operating costs are causing a rapid decline in the company’s net cash flow.
The strategic question here is how spending rates should change over time, in order to re-establish and sustain a lower failure rate and still produce a healthy, sustainable cash flow. The strategy depicted in Figure 6.3 is as follows:
� In years 5 through 10, a big investment program replaces many of the unreliable units and refurbishes most of the degenerating units. This reduces the total failure rate and cuts drastically the pool of degenerating units that will later become unreliable. So the number of unreliable units continues to decline, even after the investment program is cut back from year 10.
1 Karl T. Ulrich and Steven D. Eppinger. Product Design and Development, 4 ed. (New York: McGraw-Hill/Irwin, 2007).
D
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
107
� Maintenance of reliable units is increased. The rate that units start to degenerate does rise, simply because there are more reliable units in place over all, but the degeneration rate does not rise enough for the number of degenerating units to increase.
� Over later years, the continuing decrease in the number of unreliable units reduces failure rates still further, though this is partly countered by the small increase in degenerating units.
� After the replacement and refurbishment programs are ended, and maintenance is reallocated to reliable units, falling failure rates reduce
Fi gu
re 6
.3 : R
ec ov
er y
st ra
te gy
fo r
a po
w er
c om
pa ny
’s d
et er
io ra
ti ng
a ss
et s
C ap
ita l c
os t p
er :
- n ew
u ni
t$ 50
,0 00
$2 0,
00 0
- r ef
ur bi
sh ed
u ni
t To
ta l c
ap ita
l sp
en d
($ m
ill io
ns p
er y
ea r)
N et
c as
h flo
w ($
m ill
io ns
p er
y ea
r) 15
0 0
-1 00
0 Y
ea rs
25
40 0
0 Y
ea rs
25
R el
ia bl
e un
its (in
th ou
sa nd
s)
10 0
60
st ar
tin g
to de
ge ne
ra te
(th ou
sa nd
s pe
r y ea
r) D
eg en
er at
in g
un its
(in th
ou sa
nd s)
50 10
30
be co
m in
g un
re lia
bl e
(th ou
sa nd
s pe
r y ea
r) U
nr el
ia bl
e un
its (in
th ou
sa nd
s)
0 Y
ea rs
25
20 10
5
M ai
nt en
an ce
sp en
d ($
m ill
io ns
p er
y ea
r)10 0
on r
el ia
bl e
un its
on de
ge ne
ra tin
g un
its
To ta
l f ai
lu re
s pe
r ye
ar
Fa ilu
re s
pe r
ye ar
p er
1 00
: - b
ed di
ng -in
u ni
ts 1.
0 - r
el ia
bl e
un its
0. 5
- d eg
en er
at in
g un
its 1.
0 -u
nr el
ia bl
e un
its 2.
0
1, 50
0 0 Y
ea rs
25
C os
t p er
fa ilu
re $2
0, 00
0 To
ta l c
os t f
ix in
g fa
ilu re
s ($
m ill
io ns
p er
y ea
r)
To ta
l op
er at
in g
ex pe
nd itu
re ($
m ill
io ns
p er
y ea
r) 20
0
50
U nr
el ia
bl e
un its
re pl
ac ed
pe r
ye ar
U ni
ts re
fu rb
is he
d pe
r ye
ar
5
5 U
ni ts
be dd
in g-
in (in
th ou
sa nd
s)
0 Y
ea rs
25
be co
m in
g re
lia bl
e (th
ou sa
nd s
pe r y
ea r)
20
10 5
To ta
l n or
ew as
-n ew
u ni
ts pe
r ye
ar
0 Y
ea rs
25
10
R ev
en ue
($ m
ill io
ns p
er y
ea r)
P ri
ce , a
ft er
p fo
r en
al ty
fa ilu
re s
50 0
0 Y
ea rs
25
Fi gu
re 6
.3
D
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
108
the price penalty, allowing net cash flow to recover to a healthy and sustainable rate.
(Note that this illustration does not include the cost of capital, but the result with this cost included is still preferable to the original performance.)
6.5 How Resources Develop in Noncommercial Cases
Resources also “develop” in noncommercial situations. Staff chains work exactly as those in business cases, but demand drivers also develop—for example when political parties try to move voters from apathy to interest and then to support. Not all developments are desirable, though, and some must be deterred rather than encouraged. Today’s criminals previously moved through increasingly serious levels of antisocial and illegal activity before becoming known to police.
One public policy case concerns type 2 (adult-onset) diabetes, the prevalence of which is rising strongly in many countries. The disease does not just strike overnight, but arises after individuals have spent some time in a pre-diabetes state,
Important Detail: Developing resources must be “MECE”
Care is needed to lay out these resource chains accurately to ensure the math works out correctly. The resource development chains in this chapter each encompass the entire population of the resources involved. The law firm chain includes all lawyers in the firm (partners and others); the product chain covers all products undergoing development; and the utility company model includes all of the company’s units of equipment. This illustrates an important principle that is essential to making sure the performance outcomes are numerically accurate—the stages in the chain must be “MECE,” mutually exclusive and collectively exhaustive; that is, each unit of resource must be in one, and only one, of the chain’s stages. For example:
� The law firm’s staff are either regular lawyers or partners, but not both.
� The utility company’s assets are either bedding-in or reliable or degenerating or unreliable.
In the choice pipeline, any individual consumer must be in one, and only one, of the stocks, so careful definition of each stage is needed. The second stage is “aware but not interested,” for example, because “aware” alone would describe consumers in all but the first stage.
H
R
D
N
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
109
where blood glucose levels exceed normal levels but do not climb high enough to trigger a diagnosis. Genetic factors play a role in determining who develops pre-diabetes, but obesity is also a risk factor. With weight loss, it is possible to recover from pre-diabetes and move back to a situation where glucose levels are again in the normal range. Excess weight can also speed the progression from pre-diabetes to the full disease.
A diagnosis of full diabetes marks the point at which the progression of the disease becomes irreversible, and damage to small blood vessels starts to occur. Diabetes can, though, be effectively managed with proper medication and monitoring, and with improved diet and exercise. At some point it may become necessary to start injecting insulin. Many of those with diabetes also develop secondary complications, such as heart disease, kidney failure, blindness, or amputation.
Policy options include1:
� Improved treatment for those with complicated diabetes, resulting in the immediate benefits of improved health and productivity, fewer hospitalizations, and a net reduction in costs.
� Increased efforts to detect and manage patients with uncomplicated diabetes. This implies immediate cost but no immediate benefit; expenditure must be viewed as an investment with the pay-off coming in later years, through reduced disease progression and thus lower medical and productivity costs.
� Increased efforts to detect and manage pre-diabetes, when simpler and cheaper interventions emphasizing diet and exercise may prevent the onset of diabetes and its associated costs. Again, this effort involves immediate costs but later savings.
� Increased efforts to reduce the prevalence of obesity in society, and thereby reduce the onset of both pre-diabetes and diabetes.
6.6 Boundaries of the Firm
Section 6.2 showed one particular resource—customers—being developed through stages before actually becoming part of the business system. The same process may apply to other resources. For example, potential staff need to be made
1 Andrew P. Jones et al., “Understanding diabetes population dynamics through simulation modeling and experimentation,” American Journal of Public Health, 96(3), 488–494.
H
R
D
N
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
D
R
110
aware of a potential employer and then be made aware of the benefits of working for a particular firm before they decide to apply for a job.
Some resources continue to be relevant even after they have left the system. Former customers may no longer generate sales, but they may continue to recommend a company to others—wedding services and property agents are excellent examples of this. Former staff can also inform potential employees about what it’s like to work for a given company. Both cases, of course, have a negative alternative—unhappy former customers and staff can continue to damage your business, even though they are no longer directly involved.
Figure 6.4 brings together the key resource-development chains, showing the boundaries where resources are an active part of the system. Note that management can often influence and make use of, those resources that are outside of the system.
Figure 6.4: The multiple resource chains of the business system, and the firm’s boundaries
Boundary of the organization
Ideas in screening
In commercial evaluationIn technical
development
Current products
Products launched
discontinued
On order
Reliable assets
Cash
Capacity
Cash
initiation coming
on-stream
income spending
degrading disposal
Deteriorating assets
Unaware
Informed Aware
Junior Senior
Staff
hiring promotion
leaving
Former employees
Unaware
Informed Aware
Disloyal Loyal
Customers
Former customers
Potential resources
Active resources
Former resources
won developed
lost
Figure 6.4
B
F
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 7 COMPETITIVE RIVALRY
Chapters 1 through 6 have explained the need to constantly build and retain resources over time in order to deliver continually improving performance, but we must remember that competitors are trying to do the same. All rivalry situations can be captured by just three dynamic structures. Each can apply not only to customers, but also to other resources that may be scarce, such as staff and sources of supply. The three dynamic structures include:
a. Racing to capture potential resources: winning first-time buyers or hiring newly qualified staff
b. Trying to steal resources away from competitors and prevent the reverse: hiring rivals’ staff or keeping customers from switching to a competitor’s product
c. Struggling for share of activity from resources shared with competitors: consumer goods firms winning the largest share of retailers’ shelf space or voluntary groups capturing a larger share of donors’ total giving.
These three mechanisms often operate together. In the case of fast-moving consumer goods, competitors rush to win new consumers with a new type of product, and strive to replace their rivals’ products in stores. Since neither consumers nor stores buy exclusively any single product, companies must then try to capture a larger share of purchases than their rivals.
Competition for some resources involves no interaction—we might term these cases of “type-zero” rivalry. For example, my efforts to develop a new product do not hinder your efforts to do the same. We may compete for some related resources, such as the skilled scientists we both need, but the product development race itself is not contestable.
These mechanisms of rivalry are most easily understood with a simple example, so consider two coffee shops starting up in a town with 5,000 potential customers.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
112
7.1 Type-1 Rivalry
Figure 7.1 shows the first year of each shop’s growth, when potential customers are being captured either by our shop or by the competitor. Customers may move back into the “potential” pool again if offered poor value by one or the other of the coffee shops. Both shops start by charging $2.95 for their typical coffee product and have captured about 2,000 customers each after six months. Then, our competitor raises their price to $3.15, and we cut ours to $2.80:
� We start to win customers more quickly, while the competitor’s win rate drops. Their customers also start going back into the potential pool, so their customer base starts falling.
� Our customers visit more often, and spend more per visit, whereas their customers visit less frequently, and spend less on each occasion.
� Our sales jump slightly at this point, with the higher number of customer visits and spend, and then continue to rise as we continue winning new customers. Their sales drop slightly, with less frequent customer visits and lower spend, and start to decline.
Note that this result is specific to this case and is not a general rule about the impact and advisability of price changes!
This case is an example of a “first-mover advantage,” since our successful capture of a new customer removes that customer from the competitor’s potential pool. Unless we give that customer cause to leave us and once again become a potential customer, we deny our rival the chance of winning that same new customer at any future time.
When competition builds a market. This illustration assumes that potential customers exist—i.e., people who know what a coffee shop is and why they might like to visit one—and need only be captured. But as the choice pipeline in Figure 6.2 showed, this is not always the case. The marketing communications of all competitors can therefore speed the development of the potential market itself. For example, many companies advertising high-definition televisions (HDTVs) in 2005 brought potential customers into stores to investigate HDTVs generally, as well as to look at a particular model.
Such cases have further important implications. First, you can access potential customers who have been alerted by competitors’ marketing. Conversely, you can spend money to stimulate potential customers, only to see them be captured by rivals! Therefore, great care is required in choosing what marketing and promotion activities to deploy, at what rate and at what times, and what prices
T
- 1
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
113
to set. It is especially hard to be sure of the best choices when there are significant product differences between companies.
Some industries repeatedly replace one generation of product with another. In the case of cell phones, for example, early analogue services captured a small fraction of the potential market but were then replaced by the first digital services, which in turn were replaced by 3G services. For every generation after the first, the existing population of customers or users becomes a major part of the potential customer base for the next generation. Also, if the new product generation offers significant benefits over the previous one, it can accelerate the development and capture of new potential customers (Figure 7.2).
Fi gu
re 7
.1 : H
ow c
om pe
ti ng
c of
fe e
sh op
s c ap
tu re
p ot
en ti
al c
us to
m er
s i n
a pr
ev io
us ly
u ns
er ve
d to
w n
Fi gu
re 7
.1
V is
its p
er c
us to
m er
pe r
w ee
k
O ur
p ri
ce $2
.9 5
$2 .8
0 à
O ur
q ua
lit y
1. 0
P er
ce iv
ed v
al ue
of fe
re d
by ou
r st
or e
1. 0
1. 11
à
S pe
nd p
er v
is it
V is
its p
er c
us to
m er
pe r
w ee
k
S pe
nd p
er v
is it
P er
ce iv
ed v
al ue
of fe
re d
by ri
va l’s
s to
re 1.
0 0.
9 à
R iv
al ’s
qu al
ity 1.
0
R iv
al ’s
p ri
ce $2
.9 5
$3 .1
5 à
O ur
n et
cu st
om er
s w
on p
er w
ee k
P ot
en tia
l cu
st om
er s
20 0
0 w
ee k
52
5, 00
0
O ur
cu st
om er
s 3,
00 0
R iv
al ’s
cu st
om er
s 3,
00 0
50
S to
re s
al es
($ th
ou sa
nd s
pe r w
ee k)
R iv
al ’s
n et
cu st
om er
s w
on p
er w
ee k
70 7
21
-1 4
2, 65
4
16 39
29 .5
14 .8
-1 00 20
0
-1 00
S to
re s
ta ff
an d
ov er
he ad
0 w
ee k
52
10
S to
re p
ro fit
($ th
ou sa
nd s
pe r w
ee k)
0. 7
4. 2
0
-1 0
E xt
er na
l f ac
to rs
O ur
d ec
is io
ns O
ur r
es ul
ts R
iv al
’s d
ec is
io ns
/r es
ul ts
T
- 1
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
114
Products that evolve through several generations bring new opportunities and challenges. It may be easiest, for example, to move all of your existing first-generation customers to your s e c o n d - g e n e r a t i o n offer. However, this risks cannibalizing sales to those existing customers, so you might focus first on stealing competitors’ first-generation customers with your superior second-generation offer, or target potential customers who did not find the first-generation sufficiently attractive to become active customers.
7.2 Type-2 Rivalry
Let’s take the coffee shop example forward in time to a point where all potential customers have been captured, but the two shops are still charging the same price of $2.95 per coffee, and each has half of the 5,000 available customers. Halfway through the year, the two shops make the same decisions as above: we cut our price to $2.80, and our competitor raises its price to $3.15. Figure 7.3 shows the impact of these price changes on customers’ switching. Our sales immediately jump a little, as customers’ visit frequency and average spend increases, but this does not increase our profits because of the lower price. But customers then start switching to our store at the rate of nearly 100 per week. Our customer base grows while our competitor’s falls, leading to a rise in our weekly sales and profits and a fall in theirs.
A key assumption here is that a constant fraction of the competitor’s customers switches each week, which means that some customers tolerate the higher price of the competitor for a longer time than others. This scenario implies customers are very patient. After twenty-six weeks, less than half the rival’s customers have switched, in spite of the $0.35 lower price at our shop. Different behavioral assumptions would result in different dynamics. At the extreme, if all customers respond to a price gap immediately, a pulse of customers would switch, rather than the continuing flow shown in the center of Figure 7.3.
Figure 7.2: The development and capture of customers with new-generation products
Potential customersfor 1st generation
productsEver-likely customers
not yet reached
0 10years
0 10years
Our 2nd generation customers
0 10years
Rival’s 2nd generation customers
Potential customer base for second-generation products
Functionality and value of 2nd generation products
vs. 1st generation
Our 1st generation customers
0 10years
Rival’s 1st generation customers
Figure 7.2 T
-
2 R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
115
Type-1 and type-2 rivalry can of course occur together. Figure 7.4 shows the two shops opening at the same time and capturing customers both from the undeveloped potential and from each other. In this case, we start with a higher price than the competitor, so win potential customers more slowly. We also start losing to the competitor some of those customers we managed to win. In the second half of the year, we drop our price to try and recapture lost ground, and the competitor tries (unsuccessfully) to make more profit by raising their price.
T
- 2
R
Fi gu
re 7
.3 : T
yp e-
2 ri
va lr
y: c
us to
m er
sw it
ch in
g be
tw ee
n ri
va l c
of fe
e sh
op s
Fi gu
re 7
.3
O ur
cu st
om er
s 4,
00 0
R iv
al ’s
cu st
om er
s50 S
to re
s al
es (th
ou sa
nd s
pe r w
ee k)
S to
re s
ta ff
an d
ov er
he ad
S to
re p
ro fit
(th ou
sa nd
s pe
r w ee
k)
3, 52
2
1, 47
8
45 .5
8. 2
C us
to m
er s
sw itc
hi ng
to o
ur st
or e
pe r
w ee
k
Fr ac
tio n
of r
iv al
’s cu
st om
er s
w ho
sw itc
h pe
r w
ee k
10 0
4, 00
0
2, 50
0
2, 50
0
25 O
ur r
el at
iv e
va lu
e fo
r m
on ey
1. 0
1. 24
à
0 w
ee k
52
10
-2 .7
8. 1
0
-1 0
3. 3
V is
its p
er c
us to
m er
pe r
w ee
k
O ur
p ri
ce $2
.9 5
$2 .8
0 à
O ur
q ua
lit y
1. 0
P er
ce iv
ed v
al ue
of fe
re d
by ou
r st
or e
1. 0
1. 11
à
S pe
nd p
er v
is it
V is
its p
er c
us to
m er
pe r
w ee
k
S pe
nd p
er v
is it
P er
ce iv
ed v
al ue
of fe
re d
by ri
va l’s
s to
re 1.
0 0.
9 à
R iv
al ’s
q ua
lit y
1. 0
R iv
al ’s
p ri
ce $2
.9 5
$3 .1
5 à
E xt
er na
l f ac
to rs
O ur
d ec
is io
ns O
ur r
es ul
ts R
iv al
’s d
ec is
io ns
/r es
ul ts
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
116
7.3 Type-3 Rivalry
Not all coffee shop customers use a single store exclusively—many use two or more interchangeably. These customers constitute a fourth population in addition to the three discussed thus far: disloyal customers. These customers may be in that state from the beginning. Alternatively, they may be so satisfied with the first shop they use that they remain loyal to it for some months, only to become disloyal at a later time. Disloyal customers present a challenge, because their choice must be competed for on every purchase occasion.
Fi gu
re 7
.4 : T
he im
pa ct
o f c
ha ng
in g
pr ic
e le
ve ls
fo r
ty pe
-1 a
nd ty
pe -2
r iv
al ry
b
et w
ee n
co ff
ee sh
op s
O ur
n et
cu st
om er
s w
on p
er w
ee k
P ot
en tia
l cu
st om
er s
20 0
0 w
ee k
52
5, 00
0
O ur
cu st
om er
s 4,
00 0 R
iv al
’s cu
st om
er s
4, 00
0
O ur
c us
to m
er s’
vi si
ts p
er w
ee k
an d
sp en
d pe
r vi
si t
50S to
re s
al es
($ th
ou sa
nd s
pe r w
ee k)
R iv
al ’s
n et
cu st
om er
s w
on p
er w
ee k
S to
re s
ta ff
an d
ov er
he ad
0 w
ee k
52
10
S to
re p
ro fit
($ th
ou sa
nd s
pe r w
ee k)
53 9
18 -2
3, 63
8
82 3
-3 .1
6. 8
40 7
0
-1 0
-1 00 20
0
-1 00
$3 .5
0
$3 .0
0
$2 .5
0 $2
.8 0
$3 .1
5
$2 .8
0
$2 .9
5 O
ur s
R iv
al ’s
P ri
ce s
1. 5
1. 0
0. 5
O ur
s R
iv al
’s
P er
ce iv
ed va
lu e
1. 11
0. 9
1. 11
1. 0
R iv
al ’s
c us
to m
er s’
vi si
ts p
er w
ee k
an d
sp en
d pe
r vi
si t
C us
to m
er s
sw itc
hi ng
to o
ur s
to re
pe r
w ee
k
31
10 2
-2 9
2, 61
8
1, 40
9
E xt
er na
l f ac
to rs
O ur
d ec
is io
ns O
ur r
es ul
ts R
iv al
’s d
ec is
io ns
/r es
ul ts
Fi gu
re 7
.4
T
- 2
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
117
In Figure 7.5, the two shops again follow the pricing policies in Figure 7.4, but customers only become disloyal and do so at a rate that reflects the best value offered by either of the two stores. For the first six months, the competitor offers better value, so our price of $2.95 wins us only one-third of customer visits and a low spend from each. Our rival enjoys two-thirds of customer visits and their lower price of $2.80 drives sales of $5.27 per visit, so they break into profit after just a few months.
The situation reverses in the second half of the year as our competitor raises its price to $3.15 and we drop ours to $2.80, and we get a larger share of customer visits and higher spend. Note that if we had gone along with the competitor from
Fi gu
re 7
.5 : T
yp e-
3 ri
va lr
y fo
r vi
si ts
a nd
sp en
d by
d is
lo ya
l c of
fe e
sh op
c us
to m
er s
N et
n ew
d is
lo ya
l cu
st om
er s
w on
p er
w ee
k P
ot en
tia l
cu st
om er
s 20
0
0 w
ee k
52
5, 00
0
D is
lo ya
l cu
st om
er s50
S to
re s
al es
($ th
ou sa
nd s
pe r w
ee k)
S to
re s
ta ff
an d
ov er
he ad
0 w
ee k
52
10
S to
re p
ro fit
($ th
ou sa
nd s
pe r w
ee k)
85 6
29 -3
.36. 3
37 7
0
-1 0
-1 00
$3 .5
0
$3 .0
0
$2 .5
0 $2
.8 0
$3 .1
5
$2 .8
0
$2 .9
5 O
ur s
R iv
al ’s
P ri
ce s
1. 5
1. 0
0. 5
O ur
s R
iv al
’s
P er
ce iv
ed va
lu e
1. 11
0. 9
1. 11
1. 0
0 w
ee k
52
5, 00
0 4,
14 4
1. 0
S ha
re o
f d is
lo ya
l cu
st om
er s’
v is
its 0.
81 0.
67
0. 19
0. 33
$5 .5
0
$5 $4 .5
0
5. 27
5. 27
4. 74
5. 00
C us
to m
er s’
s pe
nd pe
r vi
si t
E xt
er na
l f ac
to rs
O ur
d ec
is io
ns O
ur r
es ul
ts R
iv al
’s d
ec is
io ns
/r es
ul ts
Fi gu
re 7
.5
T
- 3
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
118
week 26 and also charged $3.15, customer development would have stopped completely, and any higher price would have seen many customers leave the market altogether, flowing back into the potential pool.
7.4 Further Issues with the Three Types of Rivalry
Some further issues of competitive dynamics will need to be assessed in practice:
� Switching costs and adoption costs. Our examples have assumed that customers change behavior when they perceive even the smallest reason to change, but that is rarely the case. There can be costs, both financial and nonfinancial, in starting to buy a product—taking on your first mortgage for example. And there may be costs or effort in switching between rival suppliers. Changing your car insurance provider incurs time and effort, and switching mortgage provider costs money. Customers must feel that any better value you offer exceeds this cost threshold if they are to switch.
Important Detail: The value curve and blue ocean strategy
For simplicity, these examples assumed that all aspects of what the two shops offered—products, service, and so on—were equal, so competition concerned only one factor—price. In competitive situations, assessing rivals’ products on the multiple dimensions of a value curve (Sections 2.2 and 3.6) can give clearer answers to why customers and sales change over time. Many people would prefer to operate in an environment in which competition is limited or nonexistent. This is the essence of “blue ocean” strategy—the discovery and exploitation of some novel proposition that is a leap in value for customers, compared with existing alternatives.
Blue ocean strategies have always occurred, from the 19 century introduction of mail-order retailing to Apple’s iPhone. Such transformational strategies create a substantial new potential market, along with a value proposition so compelling that it both exploits that potential quickly and establishes a lead that competitors can only pursue after many years of effort. Note, though, that finding blue ocean opportunities is not always possible, and most businesses must continue competing in situations where only less clear-cut differentiation is possible.
F
I
T
T
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
119
� Market segmentation again needs to be evaluated where significantly different customer groups might exist. One competitor may lose out in the largest part of a market, but totally dominate another segment.
� Limited rationality. We have also assumed customers are rational and change behavior whenever it would make objective sense to do so. This is rarely the case, and you may not know when, or to what extent they do otherwise. People are not always honest, either with themselves or interviewers, claiming their choice of car, for example, reflected a careful appraisal of its performance and comfort, when in fact its lovely shade of blue was just irresistible! Nevertheless, in most cases people behave rationally enough for the kind of analysis described here to be useful.
� Delays may arise between changes that should trigger behavior and customers actually acting on those changes. For example, you may need to hear from several friends before you actually decide to switch to the new iPhone, even though you hear nothing new from each successive person you talk to. This reflects the build-up of an intangible state of mind, an issue covered in Chapter 9.
� Such factors can lead to path dependency, where outcomes reflect the order of prior events. For example, you may buy the first model of a new product, even though you know that better ones will be available later, resulting in the first supplier gaining a market lead, even if it never offers a better product than later competitors.
7.5 Competing with Intermediaries
Intermediaries are businesses that sit between you and the end user of your product, and they can create powerful type-3 rivalry, both in consumer and business markets. Often, suppliers fight to get intermediaries to promote their product more strongly than those of rivals. For example:
� A consumer-goods company wants retailers to allocate a larger share of shelf space to its product than to rival products in the same category.
� Business-supply companies want their products to have more pages in distributors’ catalogues than competitors’ products.
� Insurance companies want brokers to spend more time selling their policies to customers than those of competitors.
� Intel would like PC manufacturers to feature Intel processors in more of the models in their product ranges, rather than AMD processors.
C
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
120
� Suppliers use Google Adwords to get advertising prominence for their products over their competitors.
In Figure 7.6, we and a single competitor have equally appealing fast-moving consumer products for a market of 5,000 entirely disloyal customers in the locality surrounding a store. We and our rival charge a wholesale price of $1.50 (bottom left), to which the store adds a 25% mark-up, resulting in a retail price of $1.88. The store has twenty feet (six meters) of shelf space for the product category, and this is initially allocated equally between the two products. Total sales of 30,000 units per month give the store a gross profit of $11,250 per month, or $562.50 per foot of shelf space. (There are small rounding differences in Figure 7.6.)
Fi gu
re 7
.6 : A
w ho
le sa
le p
ri ce
r ed
uc ti
on c
ap tu
re s r
et ai
le r
sh el
f s pa
ce fr
om r
iv al
p ro
du ct
s
0 m
on th
24
S he
lf sp
ac e
fo r
ou r
pr od
uc t
(in fe
et )
O ur
s R
iv al
’s
W ho
le sa
le p
ri ce
s
0 m
on th
24
15 .0
C on
su m
er s
in st
or e
lo ca
lit y
G ro
ss p
ro fit
($ th
ou sa
nd s
pe r m
on th
)
5, 00
0
P ro
du ct
c os
t ($
p er
u ni
t) 1.
00
S he
lf sp
ac e
fo r
ri va
lp ro
du ct
(in fe
et )
20
20
G ro
ss p
ro fit
($ p
er u
ni t)
R et
ai l p
ri ce
s ($
p er
u ni
t)
S to
re p
ro fit
o n
ea ch
p ro
du ct
($ p
er u
ni t)
O ur
s ha
re o
f sh
el f s
pa ce
50
P ro
du ct
s al
es (th
ou sa
nd s
of u
ni ts
p er
m on
th )
N or
m al
pu rc
ha se
s pe
r co
ns um
er (u
ni ts
p er
m on
th )
6
O ur
s
R iv
al ’s
1, 00
0S to
re p
ro fit
o n
ea ch
p ro
du ct
($ p
er s
he lf-
fo ot
) O ur
s
R iv
al ’s
5
S he
lf fe
et m
ov ed
ea ch
m on
th to
o ur
p ro
du ct
S to
re m
ar k-
up on
w ho
le sa
le pr
ic e
0. 25
$1 .5
0 $1
.3 0
34 2
87 3
3. 0
15
40 .3
4. 6
15
7. 5
12 .1
2. 3
O ur
s
R iv
al ’s
0 m
on th
24
$2 $1
.6 0
$1 .4
0
1. 30
1. 50
1. 88
1. 63
10 5.
0
10
56 3
-5
Im pa
ct o
f p ri
ce s
on c
on su
m er
pu rc
ha se
s
1, 07
5
0 E xt
er na
l f ac
to rs
O ur
d ec
is io
ns O
ur r
es ul
ts R
iv al
’s d
ec is
io ns
/r es
ul ts
Fi gu
re 7
.6
C
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
121
To boost our sales and profitability, we cut the wholesale price to $1.30 from month 6.
The store keeps the same percentage margin, so reduces the retail price by the same fraction. The store thus makes less cash margin per unit. The lower retail price immediately wins our product a larger share of consumers’ purchases. If this were all that changed, the store would make less cash margin on the same total sales, but the lower retail price also increases consumers’ total purchases.
With higher total sales, and a big increase in our product’s share, the store makes more profit per foot on our product than before, and rather less on the rival product (graph in top left of Figure 7.6). The store therefore allocates three feet more shelf space to our product, taking it away from the rival product.
The next month, the store still makes more cash per foot on our product, and moves still more space from the competitor to us. We increase our sales volume and profit, despite the lower price. In each subsequent month our product wins still more shelf space, until it has gained five feet from the rival product. Although the store could make still more profit with further reallocations, it is reluctant to do so because it does not want to be dependent on a single supplier’s product.
The competitor could well respond with price reductions of its own, which could raise total product sales again and cause the store to reallocate shelf space away from our product. If rivals continued this tit-for-tat price cutting, there would be a progressive reduction in the wholesale price toward the minimum that we and our competitor could tolerate and a considerable loss of total supplier gross profit.
7.6 Competing for Other Resources
Customers are the most obvious resource that competitors fight over, but rivalry for staff is also widespread. This may even cross between different types of employer, such as retail staff moving to work for an airline.
Figure 7.7 shows type-1 and type-2 rivalry for employees when two call centers start up in a single town, with a total of 500 potential staff. We need to hire 250 people within twelve months, and offer a pay rate of $10 per hour. This encourages people to consider working for us, first becoming potential staff, then joining us so our staff numbers start to rise.
After three months, a competitor starts a similar call center, also wanting 250 staff by the year-end. They need to grow faster and so offer $11 per hour. This has three effects:
C
O
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
122
1 People start considering this employment opportunity at a faster rate (top left graph).
2 In a type-1 process, the competitor starts hiring potential employees who were already considering working for us but now have a better offer (graph of Rival’s new staff per month)
3 In a type-2 process, the competitor starts taking people who already work for us (graph of Staff switching per month).
If this were to continue (dashed lines), the competitor would nearly reach their goal, capturing 215 staff by the end of the year. Of these, about thirty will have been stolen directly from us and the rest both stimulated and captured from the ever-likely population. Halfway through the year, therefore, we raise our pay rates to $12 (solid lines). This stimulates still-potential staff (the third peak of the “New potential staff per month” graph) and nearly doubles the rate at which we can hire. It also steals staff from our competitor, with the net flow of staff switching each month(right) reversing in our favor.
Naturally, other factors feature in people’s choice of job, such as working conditions and the recommendations of friends, but these other factors will also operate by changing the flows of staff around this structure. Poor working conditions in the rival’s call center, for example, will cause staff to switch more quickly to our call center, and employees who tell their friends how much they enjoy working for our call center will increase the rate at which we win new staff each month. Some people may also leave these organizations, returning to the potential population.
Where staff develop through stages, competition may focus on particular groups. There could be particular competitive effort on developing and retaining supervisors, for example. In extreme cases, a “war for talent” can develop as competitors fight to hold on to critical skilled staff1. In 2005, the Royal Dutch/Shell oil company wanted to hire 1,000 experienced petroleum engineers. Following years of under-recruitment across the industry, and young people choosing other careers, all oil companies faced a shortage of such staff and an aging workforce, so the company was unable to find so many experienced staff.
Other cases feature rivalry for certain industry-specific resources. Our coffee shops from earlier, for example, compete for the best store locations. Low-fare
1 Ed Michaels, Helen Handfield-Jones, and Beth Axelrod. The War for Talent. (New York: McKinsey & Co., 2001).
C
O
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
123
Fi gu
re 7
.7 : R
iv al
ry w
it h
pa y
ra te
s f or
n ew
c al
l c en
te r
st af
f
airlines, such as Ryanair and Southwest, have grown aggressively by starting operations on previously unserved routes. They may also make a rather different choice—to start services on routes already being served by direct rivals.
7.7 Rivalry in Noncommercial Cases
Competition is widespread in public services and voluntary organizations and is sometimes similar to the rivalry for business customers. Voluntary organizations need to be seen to serve beneficiaries of their services if they are to win donors, so may compete for such “customers.” Other customer-like rivalry occurs with political parties competing for voters, and churches competing for followers.
R
N
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
124
Rivalry for staff is ubiquitous among public service organizations, where the framework in Figure 7.7 is directly applicable. Competition for staff can become international in scope, such as when nursing staff are attracted from low-paying economies to work in the health services of richer nations. Even terrorism is a choice that those participating in it make, in preference to benign alternatives. Recognizing that the groups concerned use tactics such as intimidation and the kidnapping of children to feed their organizational structure allows for the selection of counter-measures likely to be most effective. In Sierra Leone during the 1990s, for example, children displaced by violence were specifically sought out and protected to prevent them falling into the hands of insurgents.
Voluntary organizations clearly compete for donors and for the giving they bring in through the three standard rivalry mechanisms —capturing new donors, winning donors from other organizations, and capturing more share of the giving from donors who support more than one organization. As with businesses competing for customers and sales, voluntary organizations need to understand which form of competition they are engaged in, and where their efforts should be focused.
7.8 Dealing with Multiple Competitors
Few companies are so fortunate as to have only one competitor to deal with, so it is often necessary to extend the principles discussed in this chapter to handle multiple rivals. It is relatively straightforward to adapt the three standard rivalry structures to a case of multiple competitors—we simply add an extra customer stock for the second rival and the appropriate additional flows. However, this approach soon gets unwieldy for larger numbers of competitors, so other solutions are needed:
� To avoid the diagrams becoming too complex, several rivals’ customers can be shown in a single resource, using differing line styles or colors for each, but this is only a presentational benefit and does not simplify the analysis.
� It may be adequate to show only the key competitors to and from whom you expect to see significant customer movement. Mercedes might worry most about rivalry with BMW and Lexus, for example, rather than other car makers.
R
N
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
R
125
� Treat large numbers of similar competitors as a “strategic group.” For example, Mercedes might treat several luxury car makers as one group1.
It will still be necessary to evaluate carefully the factors motivating customers to switch between each pair of competitors or groups of competitors, but these simplifications significantly ease the workload and the complexity of the strategy debate the management team must have.
1 Discussion and examples of strategic groups can be found in: Robert M. Grant. Contemporary Strategy Analysis, 5 ed. (Oxford: Blackwell Publishing, 2005), Chapter 4.
D
M
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
126
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 8 STEERING STRATEGY AND
PERFORMANCE
Now that we have a rigorous picture of how an organization functions, performs, and competes, we can use this understanding to make better decisions on several questions:
1 Whether to take part. Following through the concept of the business life cycle (Figure 1.2), the first question is whether it is a good idea to get involved in a particular market at all.
2 Choosing a strategy for taking part. The decision to compete in a market can only be made in the context of how that business might actually work and how it might perform. The high failure rate among new ventures—both independent and in large companies—suggests there is much scope for a more effective evaluation of business ideas.
3 Designing a likely path to success. Beyond describing how a possible business could operate, there is a big challenge to define a realistic path by which that business could be started, developed, and grown to create one’s ultimate vision.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
128
4 Steering strategy through time. Having built a picture of how a business could develop, we are then concerned with directing the enterprise as it actually progresses. Competitive conditions are rarely stable, and external factors also change. Strategic management must therefore encompass the full range of decisions that significantly affect the organization’s internal development and interactions with outside forces.
5 Whether to extend or revise the strategy. This is in essence a repeat of the first two questions—whether to take part, and if so with what strategy—except that we are now looking to add to an existing enterprise. Should we enter a new market, extend our product range, and/or develop a service offer alongside our product sales? This issue covers all kinds of possible changes to the current business model, and therefore any changes to the three principal elements of strategy position—who to serve, with what products and services, and how.
The contribution of strategy dynamics to the first three questions lies in its ability to lay out on paper a crystal-clear picture of how the prospective business should work, including quantification of how all key numerical values of the business should develop over time. The same method is just as appropriate when considering whether to extend the strategy; simply show how doing so will add to the resource levels likely to develop in the already-active strategy, and the resulting impact on revenues, costs, and profits.
8.1 The Difference between Good and Poor Strategies
We have noted before that, even when a good strategic position has been chosen (who to serve, with what products, and how), the implementation decisions that steer strategy can have a powerful impact on outcomes, and therefore on whether it’s a good idea to pursue the plan. Those implementation decisions basically involve choosing what to do, when, and how much, across all significant decision items.
In the consumer brand example from Chapter 4 (Figure 4.4), we must decide each month how much to spend on advertising, how much sales effort to deploy, and what price to charge. The base case in Figure 8.1 (dashed lines) makes simple, static choices on each decision—advertising of $0.5 million per month, a sales effort of fifty salespeople, and a wholesale price of $9.00. But why is the outcome so poor, and can it be improved with a better strategy?
D
G
P
S
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
129
1 First, the advertising spend of $0.5 million per month reaches only a small fraction of potential consumers, where “reach” implies that people are not just exposed to the advertising, but notice it. Second, salespeople will struggle to win stores until consumers are interested in the brand. Last, since stores add a mark-up to the wholesale price, a relatively higher price reduces consumption, while a lower price increases it. Given these considerations, perhaps a better launch strategy might look something like Table 8.1, which includes the following steps:
Fi gu
re 8
.1 : A
n im
pr ov
ed st
ra te
gy fo
r a
br an
d la
un ch
D
G
P
S
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
130
2 Start by spending heavily to win consumer interest; delay selling to stores until there is enough interest for stores to take the product; and price low to grow volume and stores’ commitment. (In practice, a lower launch price might reflect an introductory discount, rather than a low-ticket price.)
3 Next, keep advertising heavily to capture most of the potential market as quickly as possible, and sell heavily to grow distribution.
4 Last, raise the price and cut sales and marketing to extract profitability.
Table 8.1: An improved launch strategy for a consumer brand
Figure 8.1 compares the result of this strategy (solid lines) with the simple, static policies of the base case (dashed lines). This revised strategy is not just somewhat better than the base case—it is considerably better, achieving much larger and more rapid uptake by consumers and stores, breaking into profit after just twelve months, and driving high profits. This dramatic result would not, however, be discovered from any discussion of the strategic positioning of the brand, nor from qualitative debate among the management team. Even if the general shape of the strategy, such as “advertise hard early on, then cut back,” was agreed on in principle, the actual numbers and timing would still need to be worked out.
Of course, these outcomes can only be predicted for certain with perfect information on all of the relationships in the strategic architecture of the situation, which is never entirely possible. Nevertheless, much of the benefit can be obtained, even with partial information:
� A similar episode has probably happened before. Either the company itself or others will have launched similar products in a similar market, so examine the data from similar cases.
� Even where the initiative is quite novel, well-reasoned estimates of the performance trajectory can be made by adjusting from experience in other cases.
Month Advertising Salespeople Wholesale price 1–6 $1.2m 0 $8.00
7–12 $1.2m 100 $8.00
13–18 $1.2m 100 $10.00
19–48 $0.5m 50 $10.00
D
G
P
S
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
131
� Remember that evaluation does not stop on launch date! As soon as the initiative starts, it generates information on how the situation is actually developing, so management need not make an all-or-nothing choice immediately. Instead, they can start the initiative, then either revise the implementation as data comes in, or kill it if it is clearly not working.
In spite of the clear benefits, few firms carry out such fact-based analysis and steering of strategic initiatives, with serious consequences:
� Many initiatives are begun that could never deliver what management hope for, because there is simply no plausible set of relationships between the numerical values in the strategy’s architecture that could ever lead to the desired result.
� Management persists with unfeasible strategies because there is no understanding of why progress is disappointing or the likelihood that it will continue to disappoint.
� Conversely, organizations also under-invest in strategic initiatives, and remain unaware of how much stronger their performance could have been had they chosen otherwise. Firms often kill perfectly attractive initiatives before they reach the point at which they will deliver good results. In Figure 8.1, management can be confident of the strategy, even before strong sales and profits arise, because key flow rates are increasing quickly—consumers are becoming interested, and stores are adopting the product.
8.2 Steering Strategy and Performance
What we have done in Figure 8.1—planning out in advance the decisions we think will be best—is not enough. Although we have improved confidence in the likely progress of the strategy, things will certainly turn out differently than expected. This may seem to invalidate the task of planning, but does not in fact do so. The strategy in Table 2.1 would still be vastly superior to the base case, even if it took twice as long as expected to capture consumers. This slower progress would, however, imply the need for adjustments to the strategy, such as persisting longer with the high advertising rate or delaying somewhat the large sales force commitment.
What we need, then, is a way to make such adjustments continually in response to emerging information. Such means must include clear and well-chosen indicators to be tracked, and equally clear procedures for using that information
S
S
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
132
to adjust each decision. The strategic architecture of interdependent resources that we have developed effectively specifies the business “machine” and how it functions and performs. Like any machine, this one too needs a control system if it is to perform well.
The machine analogy is made more complex when human behaviors are involved, because unlike the physics of mechanical systems, human responses are not entirely predictable. However, that is no reason not to exploit the limited predictability that exists. History shows an increasing penetration of organized control systems into more and more parts of the business system. For example, we gave up long ago any attempt to control oil refineries by looking at meter readings and using judgment to make decisions. Logistics systems, once managed by the intuition of experienced managers, are now largely computer-controlled. Systematic decision making is even encroaching on issues with substantial behavioral dimensions, such as the lending decisions of banks that were once made on the basis of individual managers’ personal assessment of each loan applicant. Such applications now rely almost entirely on credit-scoring systems. The subprime lending crisis of 2007 shows just how badly things can go wrong when such disciplines are sidelined! Loans were granted to customers who would never come close to reaching reasonable credit scores, and who came to be known as “ninjas”: no income, no job, no assets. The risk that a larger fraction would default on their loans was supposed to be covered by charging much higher interest than normal, but the default risk became so high that no reasonable interest rate would be enough to cover that risk, and the whole scheme collapsed.
Systematic decision making systems do not magically tell management the best decision to make in specific circumstances. Rather, managers follow long- established procedures for working out the best decision under a variety of conditions. Even when such procedures are not made explicit, the idea implies some kind of “formula,” into which information about the situation is entered and out of which a decision emerges. Various kinds of rule are common.
Fixed decision rules. Keeping the price fixed, regardless of the situation, is hardly a decision rule at all, but although it may seem unlikely, such rules are sometimes adopted. Some organizations operate a headcount limit for certain departments regardless of circumstances. Financial versions of such rules are quite common; a budget is effectively a rule defining how much can be spent on a given activity in a given period of time. The brand example, however, shows how suboptimal such fixed decisions can be.
Fixed fractional decision rules. This is another naïve, but common type of decision rule—i.e., setting a fixed percentage of turnover on activities such as
S
S
P
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
133
marketing, training, and so on. The problem here is obvious. If you have no revenue, you spend nothing, and therefore do nothing to improve disappointing results, such as a low customer win rate or staff skill levels. Some organizations use a version of this rule to control labor cost—e.g., managers requiring labor cost to not exceed a specific percentage of weekly sales. Therefore, if sales fall for some reason, staffing would be cut, which would then damage service, which would in turn drive more customers away! It seems likely that such damaging policies have become widespread because of the ill-advised performance focus discussed in Chapter 1: the focus on profitability, return on sales (ROS), or on invested capital (ROIC), rather than growth in cash flows.
Last year +/− x% rule. This is a common variation on the fixed decision rule and suffers from the same problems. It assumes last year’s spend was about right in the first place, and that no significant change to that spending is necessary. This is an example of a widely observed psychological phenomenon known as “anchoring and adjustment.”1 To make an estimate, people start with an implied reference point (the “anchor”) and adjust from there to reach their estimate. People asked if they think the population of Shanghai is more or less than ten million, for example, will on average make higher estimates than those asked if it is more or less than five million. It is easy to see how this thought process gets into decision making. Needing to work out the best decision for something today, people reasonably anchor their thinking with recent decision values.
8.3 Policy to Guide Decisions
Although the rules just described are seriously inadequate, they illustrate the key elements involved in making decisions to control strategy:2
� Data about some measure or measures of the current state of the system—e.g., profits
� A target for that measure or measures—e.g., target profits
� A rule or “policy” for adjusting a decision, in order to close the gap between the measure and the target—e.g., raising the price by two percent because profits are five percent below target.
1 Amos Tversky and Daniel Kahneman, Judgment under Uncertainty: Heuristics and Biases, Science, 185, 4157.
2 More can be found on formulating policy for guiding decisions in: John Sterman. Business Dynamics: Systems Thinking and Modeling for a Complex World. (New York: McGraw-Hill/Irwin, 2000), Chapter 13.
P
G
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
134
The decision that emerges from the policy is expected to work by affecting other items in the system. Here, the higher price is expected to increase profit margins by more than may be lost through lower sales. Every policy is therefore an integral part of the strategic architecture itself. Figure 8.1, for example, already shows in red where policies are located in our brand launch, although it does not show the measures that feed into each policy and does not specify the rule for each. So what data should inform a policy?
Basing decisions on performance outcomes. Perhaps our policy for the advertising spend for the brand from Figure 8.1 should be based on profit. None of the simplistic policies above takes any account of profit, so they are not likely to be especially effective.
One possible policy to adjust advertising spend in response to profit might be:
� If a previous increase in advertising led to higher profits, repeat the increase, and keep doing so until profits plateau. Increased advertising might also seem best if a previous cut in advertising led to a fall in profits.
� If, on the other hand, previous increases in advertising led to lower profits, or a decrease in advertising led to higher profits, reduce advertising spend.
The reduction in advertising spend today raises profit immediately, simply by reducing the amount of advertising cost in the income statement. But this policy is ultimately ineffective because with a continued reduction in advertising spend, we will never capture many consumers, and the brand will not take off.
Maybe we need to give advertising enough time to bring in new consumers and sales before deciding whether the spending was useful, but how long should we wait? If we look at the profit impact of higher spend after three months, it has still not captured enough consumers to justify its cost, so the policy again results in continued cuts, and sales and profits disappoint once again. If we adjust the policy to allow six months of higher advertising spend, the results are quite different—we gradually spend more, and eventually spend enough to bring in customers, sales, and profits. This is an improvement but is still way off the much better policy we worked out in Table 8.1.
Basing decisions on immediate effects. It is possible to improve policies substantially by recalling a key principle of how a business system actually works. It is resource flows that determine how performance changes, and those flows depend, either directly or indirectly, on management decisions (as well as on
P
G
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
135
existing resources and other factors). It therefore makes sense to use information on flow rates to inform the decision that influences them.
In this case, advertising spend could sensibly be decided upon based on its success in winning potential consumers. Therefore, it would make sense to link the decision to the consumer win rate itself. If more advertising wins customers faster, increase advertising;, otherwise, decrease it. This promises to be a strong and reliable policy, and illustrates a useful principle:
Policies should be informed by data on the factors, especially resource flows, which they directly affect.
This principle can rarely be applied in isolation. Higher spending affects current profits, for example, and may simply be unaffordable. Nevertheless, this principle should certainly be a key part of any policy. It may also be necessary to adjust a policy or the data used to account for other factors that interfere with the policy. In the case of the brand launch that we have been discussing, some active consumers lose interest each month and flow back into the potential population, while others are won simply by seeing the product in the stores.
Figure 8.2: Results of a policy to adjust advertising spend based on changes in the consumer win rate
20000
Potential stores
Interested consumers
‘000 Net new
consumers 000/month
0 months
Potential consumers
‘000
0 48months Sales volume
000 units per month
Advertising $000/month
3000 3000
Brand profit $000/month
1000
48
480 months - 1000
Stores stocking the brand
2375
1532
913
Sales force
50
Wholesale price $9.00
Did consumers won change in the same direction as
advertising spend?
DECISION: Spend more or less
on advertising
200
2000
9302
2000 1900
500
Figure 8.2
P
G
D
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
136
A benefit of linking decisions to the resource flows they control is that we don’t have to wait for their effects to work through the system, as we had to do when using profit to inform the advertising decision. As soon as we know that more advertising raises the consumer win rate, we can decide whether to increase spending again.
Also, as the policy becomes more accurate in its predictions, we can adjust our decisions—for example, increase our advertising spend by greater amounts. Figure 8.2 shows our company starting with an advertising spend of $0.5 million per month, and increasing or decreasing it by $100,000 each month depending on whether it results in a faster or slower consumer win rate. Although this is the net consumer win rate, including gains from product visibility and the backflow of consumer losses, we want advertising to ensure a positive win rate overall, so the policy should still work. And indeed, it does, with advertising spend increasing rapidly to a rate at which the potential market is captured quickly.
What this policy has not done, however, is ensure that the cost of advertising is justified by the additional gross profit made from sales to the additional consumers the increased advertising has won us. The simplest way to do this is to set a threshold for the number of consumers who need to be won for the higher advertising spend to pay for itself. In this case, such an adjustment results in advertising being reduced after the initial potential is captured, and profits increase progressively to a sustainable rate of over $1.3 million per month.
Finally, note that similar policies could be devised for setting the size of the sales force, based on the store win rate. The pricing policy would be more complex, since lower pricing works indirectly through increasing the consumers’ purchase rate, store profitability, and hence the store win rate.
8.4 Controlling Indirect Decisions and Interference
Some critical decisions are so disconnected from tangible performance outcomes that tracking their immediate impact is especially important.
� Where the resource concerned goes through development stages. Marketing activities along the choice pipeline (Section 6.2) may result in significant changes to the number of people who are aware and informed about a product, long before they become active customers. For example, The global drinks group Diageo PLC was about to kill a product launch after spending tens of millions of dollars, because it had seen few sales. Checking the research, though, showed that advertising had made large numbers of consumers both aware and interested in the product, so its sales were on the point of taking off.
C
I
D
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
137
� Where the resource concerned is intangible. Training is the classic example of a factor linked to an intangible resource. Spending on training today builds staff skills, that in due course impact on tangible resource flows (e.g., winning customers), which eventually add up to enough change that performance is improved. You may have to wait months or years for the training benefit to show up, by which time you may have decided to drop it, unless you are tracking its immediate impact—e.g., on staff skills or effectiveness.
� Where inflows and outflows interact. As always, it’s the flow rates you need to check, but the impact of any decision or policy can easily be lost if other flows are affecting the same resource. You might respond to a service quality problem by hiring more staff, but if work conditions are so poor that staff leave again, you will see no business benefit from what is otherwise a good decision.
� When there are multiple drivers. Chapter 3 explained how multiple factors—both within and beyond our control—affect many resource- flows, especially where people are concerned (e.g., staff and customers). You may therefore have a perfectly good policy toward one factor (e.g., salary or price levels), but still see unfavorable flow rates in other areas (e.g., staff losses or decreased customer win rate). Where this happens, you need to be still more localized in tracking the effect of your decision, for example by researching why people are not making the choices you would like. Doing so might lead you to discover that there is no problem with the salaries you offer or with your price, but that staff are leaving due to lack of promotion prospects or customers are not being won because they prefer rivals’ products.
8.5 Conflicting Objectives
Management must often, of course, pursue a balance of two or more aims; growth and profitability are a common pair. Consider an electricity supply company trying to resist attack by a new competitor who enters the market by offering lower prices. We will need a composite policy that both minimizes customer loss while still protecting prices and margins.
Our rival initially has a simpler aim: they know they will not be profitable at first, so they set prices sufficiently below ours to keep winning customers. This cannot continue indefinitely, however, and at some point they must try and become profitable, too. One possible outcome is shown in Figure 8.3.
C
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
138
Both companies cut their price by $5 whenever they lose too many customers, or fail to win enough. But in addition, we raise prices (or cut them by less than we would otherwise have done) if our profit margin drops below ten percent, and the lower the margin, the more we raise prices. Our rival starts with no objective for profitability, but from quarter 6 they, too, seek to get their net margin up to ten percent by raising prices. The scenario in Figure 8.3 showing both companies ending up with losses reflects the particular policy rules for each—other rules would lead to quite different results.
8.6 Goals and Policy in Noncommercial Cases
The principles of developing policy to steer strategy are readily applicable to voluntary and public service cases. A voluntary organization wanting to increase fundraising, for example, would learn little to guide its efforts from simply monitoring its flow of cash. It would be well advised to track instead the inflow of new donors, the donation rate of donors, and the rate at which previously loyal donors were lapsing. Another example concerns many cities’ efforts to deal with homelessness. Voluntary organizations often responded to the issue by offering meals to the homeless on the street. The unintended consequence is to discourage the homeless from seeking help to escape from their situation, so the numbers sleeping on the streets escalates rather than falling. Policy in many cities now focuses instead on slowing the rate at which people become homeless, and
Figure 8.3: Composite policies of rivals in electricity supply
Our customers
000
Rival’s customers
000
1000
0 quarters12
0 quarters
12
331
200
-200
100
-100
0
-200
-77
-1
Operating profit
$m/quarter
Ours
Rival’s
Rival’s 400
375
Ours 391 392
669 Prices
$ per quarter
... for customer switching
... to maintain
profitability
... to maintain
profitability
... for customer switching
Our policy ...
Rival’s policy ...
Figure 8.3
Our decisions Our results Rival’s decisions/results
C
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
S
S
P
139
enabling people already in that situation to access services that move them out of it, sustainably.
Non-business cases can, though, bring additional complications:
� Multiple agencies. Homelessness is affected by the wide range of policies of a number of different agencies, such as social services, the police, and charities, of which many will exist, offering a variety of services. It is hard enough for a single management team to design coherent business policies, so it is hardly surprising that public-policy cases are so complex.
� Zero-goals. A further feature of noncommercial cases, introduced in Chapter 1, is that the goal is often to move an indicator toward zero—e.g., no crime, no diabetes, no homelessness, and so on. Such zero-targets can quite readily be built into the standard policy frameworks discussed in this chapter.
� Conflicting goals. Public services and voluntary organizations will also face the challenge of conflicting goals. Many countries want to stop growth in the number of people with type 2 diabetes, for example, but there is clearly a conflict between this goal and the need to limit the costs involved. In addition to this conflict is the conflict between seeking the greatest quality of life for people with the disease and the wish to minimize the number of sufferers. Success in the first objective results in failure on the second, due to increased life expectancy.
G
P
N
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
140
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 9 INTANGIBLE RESOURCES
Intangible, or soft, factors clearly have a big impact on performance—a damaged reputation can destroy a business, strong staff motivation can drive powerful growth, proprietary knowledge can generate market-leading products, and so on1. But making practical use of these factors to steer strategy is difficult, because terminology is often abstract, ambiguous, and inconsistent.2
1 See for example: Charles. J. Fombrun. Reputation: Realizing Value from the Corporate Image. (Boston: Harvard Business School Press, 1996).
2 See for example: Jay B. Barney. Gaining and Sustaining Competitive Advantage, 3 ed. (Upper Saddle River, NJ: Prentice Hall, 2007), Chapter 5; Robert M. Grant. Contemporary Strategy Analysis, 5 ed. (Oxford: Blackwell Publishing, 2005), Chapter 5.
All classes of asset-stockTangible resources
Attributes of tangible resources
Intangible resources
Tangible Intangible
Capabilities
including skills
Informational
Psychological
Quality-related
Customers Staff
Products Capacity
Cash Other: context-
specific plus operational
items, e.g. inventory, orders
... consisting of routines or procedures
+ skills and
knowledge
plus non-resource quality items
Figure 9.1Figure 9.1: A classification of resources and capabilities
All classes of asset-stockTangible resources
Attributes of tangible resources
Intangible resources
Tangible Intangible
Capabilities
including skills
Informational
Psychological
Quality-related
Customers Staff
Products Capacity
Cash Other: context-
specific plus operational
items, e.g. inventory, orders
... consisting of routines or procedures
+ skills and
knowledge
plus non-resource quality items
Figure 9.1
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
142
Resolving these problems leads to a classification of “assets”—resources (both tangible and intangible), and capabilities—that has proved reliable and practical (Figure 9.1).
To expand on the table above:
� The tangible resources listed at left have dominated our discussion to this point, and constitute the core of an organization’s strategic architecture. Not all of these resources arise in every case, and some cases may require additional resources, such as “projects” in contract- based firms.
� Other asset stocks exist but have little effect on strategic performance. Inventory levels, for example, are rarely significant to overall performance, even though they observe the bathtub behavior of all asset stocks.
� The next category of asset stocks is the attributes of those tangible resources, discussed in Chapter 5. Some of these attributes are actually intangible, such as staff skills or product appeal.
� The three main categories of intangible resources are: (1) psychological; (2) information-based; and (3) quality-related. Some quality items do not accumulate, however.
� This leaves capabilities, which combine the skills of individuals and teams with knowledge that may have been captured, as well as routines and procedures. Capabilities will be discussed in greater depth in Chapter 10.
Once we have clearly defined intangible resources, their levels can be measured and their impact on the rest of the system assessed. It is then possible to design actions to improve matters and put in place procedures for ensuring they stay that way.
Not “intangible assets.” The term “assets” is used indiscriminately in strategy to define all types of resource and capability. All such factors do indeed fill and drain over time, a process technically termed asset stock accumulation. But there are two reasons for not using “assets” when assessing resources and capabilities. First, a term encompassing such a wide variety of factors is confusing when also used to define any particular category. Second, the term “assets” has a specific, accepted meaning relating to items in a company’s accounting statements. The term “intangible assets,” too, has a specific meaning in that context, referring to the value of such items as patents and brands. Our concern, though, is with the intangible elements themselves, not with their financial value, so to avoid
A C
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
143
confusion with the use of the term in financial accounting, we will therefore refer to intangible resources or intangibles, rather than “intangible assets” or “invisible assets.”
The three most common categories of true intangible resources are:
1 Psychological factors, concerning the state of mind of key groups, especially customers and staff, but also investors and other stakeholders.
2 Information-based resources, such as data, technology, and knowledge.
3 Certain quality factors that must be built up and sustained over time, such as the reliability of new vehicle models or electronic equipment.
9.1 State-of-Mind Intangibles
The principle behind state-of-mind intangibles is the core concept of cognitive psychology—that state of mind drives behavior—so detecting, measuring, and managing state of mind is key. The case of the service department, introduced in Section 4.6, includes two such factors that substantially worsen the company’s problems. The department is the major unit of an IT service company, serving medium-sized businesses. It recommends choice of hardware and software, installs these for clients, and provides maintenance, support, upgrades, and training.
At the beginning, the company served approximately ninety clients who required about seventy-five hours per month of service. Its founder and CEO was signing up just under two new clients per month, and none were leaving. Each new client also needed about 200 hours of initial work to get them set up on a sound basis. The company was delivering good service to its clients, and had a solid reputation that helped the CEO sign up more new clients.
The company employed seventy technical staff, of whom fifteen were relatively new. It was taking on two or three new people per month to cope with growing demand, and losing only about one person per month. Staff had 120 hours per month to serve clients, after allowing for administration, holidays, and so on. New staff were not so effective, taking about three times as long as experienced staff to do typical tasks. In addition, each new employee needed about ten hours per month of supervisory time from an experienced employee. It took three months for new staff to become fully productive. Staff morale was high, reflecting the company’s stable situation and the interesting work. Staff were busy, but not overloaded, working at about ninety-five percent capacity to ensure good client service.
S
-
-M
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
144
The CEO found himself with too little time to continue his selling efforts, so brought in an experienced business development executive. With the company’s strong reputation, this new executive was soon bringing in more than double the rate of new clients per month. All seemed well until about nine months later. Staff were increasingly busy—payroll information suggested they were putting in fifteen percent more time than normal—and there had also been an increase in complaints from clients about service quality.
After a further three months, things really started to fall apart. Staff pressure got still worse, even though staff numbers had grown to eighty-seven, and complaints from clients, now numbering 143, escalated sharply. For the first time, some clients left the business to find support from other providers, and success in winning new clients declined. Still worse, staff started to leave more quickly than ever before, apparently due to the pressure they were under.
Figure 4.6 provided a simplified explanation of the problem, but it does not adequately address the following questions:
� Staff were overloaded from month 9, so why was it not until month 15 that customers started complaining?
� Why was it a further few months before clients actually left?
� Why did staff losses only pick up some three months after the overload started, and grow only slowly over the next few months?
� If the worst point reached was an overload of about twenty percent, how come the firm subsequently lost nearly fifty percent of its clients?
Figure 9.2 pulls apart one of the mechanisms at work. Starting from the bottom, we see that the staff are just able to sustain good service quality until the pressure gets too high—the ratio between the amount of work to be done and the capacity to do the work. Service quality then falls. This is not easily measured, as it consists of various issues, from delays in answering the phone, to incorrect advice, to faulty installations. But one measurable consequence is the number of problems reported by customers. This is not a nice, smooth data series, but nevertheless, we can see that the problem rate clearly increased during the time that staff pressure escalated.
The escalating service problems built up a level of dissatisfaction among clients (the key “state-of-mind” intangible, and a negative one), but clients tolerated this for a few months. It was only when dissatisfaction reached a threshold limit some time later that it triggered actual client losses.
S
-
-M
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
145Similar reasoning explains changes in staff morale and attrition1. The client losses poisoned the positive intangible—the company’s reputation—that had enabled more clients to be won in the first place.
How might this firm have avoided these difficulties? Simply by increasing staff numbers and continuing to hire ahead of increasing demand. How did it dig itself out of the problem? This was not so simple and involved the rather shocking tactic of terminating still more client contracts. This is not a general recommendation on how to resolve service-quality problems, but was necessary in this case because the difficulties were so serious. Only by radically cutting the demands on the system could pressure be relieved and normal service resumed. The cut had to be handled carefully, but had the additional benefit of improving the quality of the firm’s client base (see Chapter 5).
Situations involving state-of-mind intangibles raise other common observations:
1 Strategic management of reputation and morale can be explored or taught with the Professional Services Microworld business game, see http://sdl.re/profserv. The Football League Challenge—an exercise for larger classes—also explores the intangibles of team morale and cohesion; http://sdl.re/football.
[ Service quality ]
Clients
0 months
200
24
79
143
New clients per month
10
4.5 2.2
C lostlients per month
25 19
3
Forgiveness rate
Increased annoyance per month
0.04
Fraction of clients lost
0.25 0.17
Problems per client per month
2.0
1.0
0
Client annoyance
0 months 24
1.0
0.5 0.7
tolerance limit
Pressure on staff1.5
1
0.5
0 months 24
0 months 24
0
1.0
Figure 9.2 Figure 9.2: How staff pressure negatively affects service quality, client satisfaction, and staff attrition
[ Service quality ]
Clients
0 months
200
24
79
143
New clients per month
10
4.5 2.2
C lostlients per month
25 19
3
Forgiveness rate
Increased annoyance per month
0.04
Fraction of clients lost
0.25 0.17
Problems per client per month
2.0
1.0
0
Client annoyance
0 months 24
1.0
0.5 0.7
tolerance limit
Pressure on staff1.5
1
0.5
0 months 24
0 months 24
0
1.0
Figure 9.2
S
-
-M
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
146
� Intangibles can have very large-scale impacts on the tangible system and performance and so should not be ignored.
� Negative issues are different from a lack of positives. Client dissatisfaction in Figure 9.3 is not the same as low satisfaction, and this difference matters.
� Intangibles are both influenced by and influence the tangible system.
� The client dissatisfaction and staff morale illustrated in Figure 9.2 were caused by changes to the tangible resources: clients and staff.
� Intangibles are manageable. Sometimes they can be influenced directly. UK bank Barclays PLC, for example, identifies “miserable moments” that its customers experience, and always follows up such incidents with a personal apology and symbolic gesture to the offended customer.
Current customersNew customers
per month Customers lost
per month
Change in reputation per month
Reputation
Current quality
(Factors affecting quality)
Time needed for reputation
to change
Figure 9.3: A common relationship between quality, reputation, and customer movements
(Other factors causing loss of customers)
(Other factors causing new
customers to be won)
indicates incomplete causal links, simplified
for clarity
© Kim Warren, 2011
Important Detail: Quality and reputation: A common structure
The connection between quality and reputation is very common. The principle is that quality affects current customers, while reputation affects potential customers (Figure 9.3). Since we are considering two different groups, our concern differs in each case—we want potential customers to join us, and we want active customers to stay with us and buy more from us. The factors that influence each group will also be different. Only current customers can have real experience of product or service quality, and therefore be motivated to leave if these are not good enough. Potential customers, on the other hand, can only respond to what they hear about product or service quality (i.e., the company’s reputation).
Figure 9.3: A common relationship between quality, reputation, and customer movements
Current customersNew customers
per month Customers lost
per month
Change in reputation per month
Reputation
Current quality
(Factors affecting quality)
Time needed for reputation
to change
Figure 9.3: A common relationship between quality, reputation, and customer movements
(Other factors causing loss of customers)
(Other factors causing new
customers to be won)
indicates incomplete causal links, simplified
for clarity
© Kim Warren, 2011
S
-
-M
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
147
� Problems can fix themselves, though not always in an ideal way. Losing customers relieves staff pressure, for example, but is not a desired outcome.
� Intangibles can lead to persistent under-performance. This is a common and serious failure of strategy, resulting from many organizations’ obsession with tight cost control. They never have quite enough staff to get things done well, so are widely infected by poor intangibles. They still push their employees to perform, but keep hitting limits caused by damaged intangibles.
� Sporadic events are common drivers of intangibles. The failures that the clients of the firm we have been discussing in this chapter experienced were occasional incidents of varying seriousness. This makes it important to capture incidents, not just continuous measures, as the bank did with its miserable moments.
� Thresholds drive tipping points. People cannot possibly react to every small change in how they feel about things, so it is extremely common to see no significant reaction from a particular group until feelings, whether positive or negative, have built up to some threshold level. This makes it vital to track those feelings, if only by rough checking and estimation, in order to anticipate discontinuities that may occur at some future time.
� Success on balanced scorecards can actually cause failure. All looked good for the service firm we have been discussing in terms of productivity, profitability, and business growth for many months, during which time the seeds of its failure were taking root.
9.2 Information-based Intangible Resources
The simplest informational factor is data: specific information about something important. This can be illustrated by extending the example from Chapter 5 concerning the performance of a call center.
In Figure 9.4, the call center started with 200,000 customers making on average one call per month. The eighty staff could each take 125 calls per day, or 2,500 per month if fully trained, and so could just handle all the calls received. But to answer customers’ enquiries, the staff need information about customers and their activity—that is, data. Some data is long-lived, such as name, address, and other personal details, while other data is transitory, such as recent transactions or enquiries.
I
-
I
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
148
Here, twenty items of data are needed if every possible enquiry is to be answered, but at first only sixty percent of that data is actually available, enough to deal with ninety percent of customer enquiries. Missing data also increases the time needed for staff to spend on each enquiry, so instead of being able to respond to 125 calls per month, each employee can only handle ninety-nine. The call center therefore has too few staff, simply because of the lack of easily accessible data. Over a quarter of calls go unanswered, causing a loss of about five percent of customers each month. (This example does not include customer dissatisfaction, discussed above.)
Each month, more data becomes obsolete: old transactions become irrelevant, customers’ details change without being updated, and so on. By month 12, the
Fi gu
re 9
.4 : A
dd in
g sy
st em
s t o
ca pt
ur e
da ta
a llo
w s g
ro w
th in
c al
l- ce
nt er
a ct
iv it
y
I
-
I
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
149
system comes into balance, but only because customer losses bring pressure back to a level that staff can deal with. This is helped by the rising staff numbers, so although each person can only respond to ninety-two calls per day, there are now enough staff to manage all the incoming calls.
Realizing from the start that lack of data is damaging staff productivity and customer service, the call center invests in improved systems to collect data. This takes the first twelve months to complete, but when those systems come online, the flow of up-to-date data increases rapidly, and the database starts to fill. Staff productivity rises strongly over months 12 to 18, so the center’s capacity is easily able to handle the larger number of calls from the increasing customer base. Note, incidentally, that the total quantity of data to be captured is itself rising, due to the growing number of customers.
A working model of a similar situation in which data supports customers service is at http://sdl.re/mpdg. This is an example of working models included in the strategy dynamics course.
This is a simple case of an investment in information systems (IS) that clearly makes sense. It is not always so easy to define or portray the information itself, its impact on the rest of the system, or the impact of actions taken to manage it. Further issues include:
� Interdependence between systems. Managers find it hard to assess the value of IS investments, because the combined benefit of multiple investments is not a simple addition of the individual benefit of each investment. The value of improved call-handling, for example, will be reduced if not accompanied by improved data quality. This makes it meaningless to try, as most organizations do, to work out the financial case for each IS initiative. It is the program of investments that improves performance, not the individual systems.
� The contribution of IS degrades. Here, there is a problem with the obsolescence of data. But decay arises for other reasons, as well. The business processes that IS supports will change, so the system becomes increasingly ineffective at supporting those processes. The organization’s needs may change in scope and complexity, so an unchanging set of systems again becomes less effective.
� IS investments are not just about processing data. IS investments often change the business processes themselves and affect the skills of the people using them. A bank’s credit-scoring system relieves executives
I
-
I
R
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
150
of the need to be skilled at assessing a customer’s risk, while sophisticated support systems enable service staff to handle enquiries they would previously have needed to refer to others.
“Knowledge” is a less well-defined form of information-related resource, since it can refer to knowing what as well as knowing how. Knowledge management is especially relevant in consultancies and other professional firms, but also features in other organizations1. Collecting and organizing such knowledge is of course costly, but doing so leads to a number of potential benefits. As Figure 9.1 indicates, knowledge frequently operates by contributing to an organization’s capability at getting things done, an issue we will deal with in Chapter 10.
9.3 Quality-based Intangibles
The focus on quality over recent decades has been of much more importance than merely making business operations cost-efficient and reliable. As the case of the computer service company introduced in Section 4.6 and discussed above shows, quality issues are of critical importance to strategy—quality has a significant impact on the medium- to long-term development of performance2.
Certain quality measures can move immediately, in response to the factors on which they depend. If for example a retail store finds that five of its usual twenty staff do not show up for work one day, its service quality will instantly suffer. If those same employees show up again on the following day, service quality will jump back to its normal level. A law firm winning a major engagement that pushes workloads well above the capacity of its staff will deliver poor work, either to this project or others, but quality will recover when that engagement ends
Quality is key to strategic performance, so should feature in the business architecture if its certainty is not assured.
Many firms manage quality so well that it has become utterly reliable, and so can be safely ignored when assessing its strategic prospects. However, quality is not always assured, and there are various elements of quality that need to be tracked:
� Quantification is important. There is a big difference for the customers of the retail store if its staff shortage leads to queues of five minutes
1 Georg von Krogh, Kazuo Ichijo, and Ikujiro Nonaka. Enabling Knowledge Creation: How to Unlock the Mystery of Tacit Knowledge and Release the Power of Innovation. (Oxford: Oxford University Press, 2000).
2 Barrie G. Dale. Managing Quality, 4 ed. (Oxford: Blackwell Publishing, 2003), 51–65.
Q
-
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
151
or fifteen minutes; the law firm’s overload will be more serious if it leads to clients losing cases, rather than just experiencing delays.
� Track the correct quality indicator. Consider an office furniture supplier who is experiencing delivery problems after expanding its product range. Delivery lead-time, measured in days, is an obvious measure of service quality, but is not actually the most important factor, since office refits are mostly planned well in advance. On-time delivery—delivering on the day and time promised—is more important to customers and therefore a better measure of service quality. Delivery completeness is also important, since missing items cause inconvenience.
� Track the quality drivers, not just the quality itself. The computer service case we have been discussing shows how situations can progress toward a crisis, while exhibiting no actual problems until the factors driving it have moved into critical territory. Tracking workload and staff capacity would have shown the quality problem approaching.
� Focus on the customer. The office furniture example illustrates a case where several quality issues arise, but some are more important than others. This is why quality initiatives place heavy emphasis on the “voice of the customer” (VoC)1.
� Internal quality matters, too. Many activities are carried out by one department or unit in an organization for the benefit of other departments or units (“internal customers”), so poor quality can undermine the ability of other units to perform well.
Although many quality measures respond more or less instantly to the factors that drive them, others clearly accumulate. Product development efforts, for example, progressively increase product performance or functionality. In essence, we are assessing how much of the users’ requirement the product fulfills, assuming it is produced entirely to specification.
Other positive quality indicators that accumulate include various forms of efficiency, for example the fuel-efficiency of motor vehicles or aircraft, or the fraction of incident energy captured by solar cells. Some important indicators are not so much quality as “lack of quality.” Examples include the manufactured
1 Michael L. George, John Maxey, David Rowlands, and Malcolm Upton. The Lean Six Sigma Pocket Toolbook: A Quick Reference Guide to 70 Tools for Improving Quality and Speed. (New York: McGraw-Hill, 2004), 55–68.
Q
-
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
152
quality of electronic components, which shows up in the reject fraction; the quality of chemicals, measured in terms of purity; and the quality of manufactured products that shows up in the fraction of subsequent warranty claims.
In Figure 9.5, a piece of software developed for 1,000 users is released with 100 unknown bugs. As users start using the software, they find the bugs and report them. The bugs are gradually fixed by a small team, but users continue to experience bugs until the fix is made and deployed.
The initially high discovery rate of unknown bugs falls sharply because so many users are working with the software and therefore discovering the bugs. However, not all bugs are quickly discovered because, even after many months, there is a low probability that any particular feature is used in exactly the way that will result in a bug event.
In the first case (dashed lines and light text), a team of five people have failed to fix all known bugs, even after twelve months. In the second case (solid lines and bold text), a larger team fixes the known bugs more quickly. Surprisingly, though, this quick elimination of known bugs is not enough to make much reduction in the total number of bug events users experience over the entire two-year episode. The majority of bug events were experienced by users in the early months, filling the stock of known bugs not yet fixed to a high level. Even with a larger team, it takes some time to fix those bugs—too late to undo the problems users experienced early on.
Figure 9.5: Fixing bugs discovered in a software application
Q
-
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
I
R
153
This example illustrates features that arise in other cases of fault-based quality:
� The problem experienced by the customer or user is different in nature from the fault itself. For example, the experienced problem could be a numerical error, a messy screen format, or a complete system crash. A problem with the physical manufacture of a car could show up as excess noise, high fuel consumption, or a road-side breakdown.
� There is a trade-off between the desire to release a product as soon as possible, and the aim to release it with the fewest number of problems.
� Many problems persist, unknown either to the producer or the customer, simply because the occurrence of the combination of events needed for the problem to arise is rare.
� Discovered problems require time and effort to be fixed, so a balance has to be made between the cost of resources needed to fix all discovered problems as soon as possible, and the costs associated with allowing problems to persist.
Q
-
I
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
154
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CHAPTER 10 CAPABILITIES
Capabilities have received considerable attention in strategy research and popular writing1. However, making use of capabilities to design and deliver performance faces the same difficulty that arises with intangible resources: terminology that is inconsistent and abstract. We start with a firm distinction:
Resources are things we have (or can access); capabilities are activities we are good at doing.
In simple grammatical terms, since a capability is about doing something, it can be expressed as the present participle of a verb—marketing, hiring, serving customers, developing products—or the noun describing such a process— product development, recruitment, and so on.
� Capabilities, like resources, are asset stocks. Capabilities accumulate and deplete, and exhibit the characteristics of asset stocks established in Chapter 3.
� Capabilities differ from skills. Skills are attributes of individuals, and move with the people who hold them, but teams need more than the sum of those skills if they are to be capable.
� Capabilities are composite factors. Capabilities consist of people’s skills, knowledge, and procedures for getting things done (see Figure 9.1). Nevertheless, it is possible and helpful to formulate a capability as a single factor.
1 See for example: George Stalk, Philip Evans, and Lawrence Shulman, “Competing on Capabilities,” Harvard Business Review, 70(2), 57–69.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
156
The terms “capability” and “competence” are interchangeable.
� Core competences are different from capabilities1. The catchy phrase “core competence” is used widely and indiscriminately in articles, books, and discussions among executives. The phrase originally described the powerful underlying technologies in multi-business firms. Honda’s four-stroke engine technology enabled them to compete in motor-cycles, cars, and snowmobiles, and Canon’s laser technology lay at the heart of its scanners and printers. But a core competence alone is not enough if we lack basic capabilities in other functions. Honda, for example, struggled in the car industry during the 1990s, because of weaknesses in design, marketing, and production engineering.
� Capability building is best done deliberately. Well managed organizations do not merely hope that capabilities will emerge by trial and error—they know what they need to be good at doing, and deliberately work out how best to do those things. They then continue relentlessly to keep improving what they do.
10.1 Dimensions of Capability
To specify capabilities requires that we be clear what being “good at” something actually means. There are three elements to consider: (1)is the activity done quickly; (2) is it done well; (3) and is it done at low cost?
1 C. K. Prahalad and Gary Hamel, The Core Competence of the Corporation, Harvard Business Review, 68(3), 79–91.
Important Detail: Keep it simple with Ockham’s Razor
Ockham’s razor is a principle attributed to the 14th-century friar William of Ockham. It states that the simplest explanation for any phenomenon is likely to be the best. This implies that, in the search for scientific theories, it is advisable to seek explanations that are “parsimonious” (requiring the minimum number of factors), and do not involve ambiguous or abstract factors and mechanisms. This principle is worth bearing in mind when considering the likely usefulness and reliability of strategy theories and frameworks.
D
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
157
Figure 10.1 looks again at the retailer discussed in Section 5.3. As the retailer gains confidence, it wants to open stores quickly in order to capture the market before competitors have a chance to move in. As the retailer uses up the potential market, it slows its store-opening rate. At this point, the fastest it could possibly open any new store, after deciding to do so, is over three quarters (dashed lines and light text). However, with a low capability of finding locations and then acquiring and developing stores, the best it can do is open each new store after a process taking six quarters.
Opening stores quickly is not enough to ensure that sales and profits grow strongly, however. The retailer wants high-quality stores, which means finding locations that can attract as many potential customers as possible in each area. It also wants to acquire each store for a low cost. Figure 10.2 adds these two capabilities. The business spends twenty-five percent more on each store than it has to, so its cost capability is 1 ÷ 1.25, which is 0.8. The average store’s location reaches only about seventy percent of the potential customers that the best locations could access, so its quality capability for opening stores is 0.7.
In Figure 10.2, other capabilities are assumed to be adequate. For example, the retailer offers appealing products, hires and deploys enough staff to meet demand and keeps them well-trained. It is remarkable, then, to see the dramatic impact of a single strong capability concerning the opening of new stores. The difference would have been still greater with a higher capability at merchandising: the choice of products to offer and their positioning in stores. If this led to the capture of the nearly one million potential customers still remaining at quarter 20, the
Figure 10.1: Impact of store-opening lead time capability for a retailer
D
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
158
company’s sales would have been nearly $300 million per quarter and operating profits would have exceeded $8 million per quarter. It is not surprising, then, to find that organizations with even small capability advantages across the many key activities—staffing, product development, marketing, sales and so on—deliver orders-of-magnitude more performance than less capable rivals.
Note too that the measures of capability are neither abstract nor impractical. It is perfectly possible for a retailer to track how long it takes to get a new store opened, to compare its costs with the best that others can achieve, and to assess each store’s reach into its local market. It is equally possible for most firms to evaluate their capability of doing certain key tasks quickly, well, and economically, and, armed with this information, set plans and targets for improving each capability.
Fi gu
re 1
0. 2:
Im pa
ct o
f m ul
ti pl
e st
or e-
op en
in g
ca pa
bi lit
ie s f
or a
r et
ai le
r
D
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
159
Figure 10.2 also illustrates a further common observation:
Most capabilities are concerned with building and retaining resources.
There are exceptions, but most important capabilities are clearly connected to the flow rates in the strategic architecture—winning customers, developing products, retaining staff, and so on.
Capabilities in low-fare airlines. Ryanair and similar low-fare airlines demonstrate many of the principles regarding capabilities, following the principle that capabilities will be found in the major resource flows of the strategic architecture, as well as with certain non-resource performance measures.
The opening of operations at new airports is a particularly close match for the site-acquisition capability of the retail store chain. Ryanair has by now developed a strong capability to identify and add new travel routes and to market those services in order to build customer numbers and capture the journeys they wish to make.
Capabilities and business processes. Strong capabilities depend on having effective business processes, so it is helpful to understand the connection between these concepts1. In simple terms, business process design identifies the set of activities required to make something happen in the most efficient way. Since our strategic focus is on resource flow rates, these are the processes that are of most concern. Figure 10.3 expands the store-opening resource flow into key stages of the process. If these are optimally designed, and if the staff involved are skilled and have the information they need to do their tasks well, then the retailer be able to find and open quality stores, quickly and inexpensively.
1 An accessible explanation of business process design can be found in: Daniel V. Hunt. Process Mapping: How to Reengineer your Business Processes. (New York: John Wiley & Sons, Inc., 1996).
Figure 10.3: The processes of store opening for a retailer
Receive opportunity
Meets our needs?
YES
NO
Rejected
Negotiate opportunity Success?
YES
NO
Lost
Acquire site
Develop site
Fit out store Open store
BUSINESS PROCESS STEPS
Stores opened per quarter Stores
operating
Figure 10.3
D
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
160
10.2 Learning Develops Capability
At the founding of the retail business above, management would likely know that finding locations for new stores would be an important capability and therefore hire someone with relevant experience for the task. Since capabilities, like resources, are asset stocks, they accumulate and deplete, which in this context means learning and forgetting. To complete the picture of how capabilities arise and help grow resources, then, we need a model for how learning occurs.
The retailer might reasonably hope that the employees who find and open stores will gradually learn to do the job better across three dimensions; opening better stores, more quickly, and at lower cost. The following events describe this capability development for UK restaurant and hotel group Whitbread PLC over approximately five years:
� Their first expert had some initial success in finding good locations for new restaurants, but was soon very busy, traveling long distances to visit promising locations, assess them, reject many as unsuitable, negotiate to acquire good opportunities (many of which failed), and follow through with purchases.
� The company’s ambitions required faster site acquisition, so the company hired more experts, but its capability was still no more than the sum of these experts’ skills. By this time, the group was getting information on which factors were most important to store performance—visibility, passing traffic, ease of access, and so on. So they wrote out these specification and issued them to real-estate agents in their search areas. They started to receive fewer opportunities, but much better ones. They also discussed how successful they were with each purchase negotiation, and shared with each other tips for achieving the lowest possible price.
� The analysis required for new opportunities was now better understood, so the team hired an analyst. They had computer systems set up to automate parts of the assessment, such as the socio- demographic profile of a locality’s population. This allowed a quick desktop appraisal of each new opportunity, cutting the number of locations that had to be visited still further.
� Meanwhile, more staff were added, but now the retailer could bring in people with less experience, quickly train them in their processes, and support them with the information and systems that had been developed.
L
D
C
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
161
As a result of these events, the company was soon acquiring much better locations, more quickly, and at lower prices than any competitor in its sector—an advantage that remains with the business to this day, some 20 years on.
Taking a high-level view of this story offers a simple architecture for capturing how a capability develops. The small initial capability (acquiring effective locations for their restaurant) drives an inflow of the target resource (the locations),and that flow itself drives an increase in capability. The next period’s higher capability drives a faster, better, less expensive inflow of further resource, leading to the generic architecture for learning mechanisms described in Figure 10.4. This powerful self-reinforcing feedback between resource acquisition and the growing capability can be observed in many companies who emerge rapidly to dominate their markets, such as Starbucks, Amazon, and Google.
If even modestly effective learning is added to the initially poor site-finding capability of the retail business shown by the solid lines of Figure 10.2, performance can improve sufficiently quickly for the organization to achieve very similar results to the best achievable outcome, shown by the dashed lines of Figure 10.2.
10.3 Capabilities Not Linked to Resource-building
Although most strategically critical capabilities are linked to the acquisition, development, and retention of an organization’s tangible resources, certain capabilities operate on factors other than resources.
One example concerns the teams of sales people that some newspapers and magazines employ to sell advertising space by phone to known advertisers. Other sales staff are responsible for winning those advertisers in the first place. This is
Figure 10.4: Generic structure for the dynamics of learning to build capability
C
N
L
R
-
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
162
a common practice in other organizations, too, where winning and retaining customers is the main responsibility of sales people who call on customers, leaving continuing sales capture to telephone-based sales groups.
Chapter 3 showed how sales performance in many situations depends on winning and keeping customers, and on growing sales to existing customers. We therefore commonly find a capability associated with each of these three elements (Figure 10.5).
Other examples of capabilities not directly related to resource flows include distribution capabilities to ensure on-time delivery to customers, engineers’ capability to repair failed equipment, and the capability to select a product range likely to maximize customer purchase rates.
Measuring capabilities. If capabilities are to be usable for strategy analysis and planning, it is important to give them measures that are accurately expressed and unambiguous. As for intangible resources, it is often helpful to start with the extremes of a zero-to-one scale:
� Zero capability implies that, no matter how much of other useful resources a team is given, it would not succeed in building the resource for which it is responsible. If the team’s task is to retain a resource against loss, then the outflow of this resource continues at the rapid rate that would occur if the team did not exist.
� A capability of 1.0 is the maximum performance that can be imagined, or that is possible, given absolute limits.
This zero-to-one range often leads to capability levels that are nearer the 1.0 end of the scale, since organizations with especially low capabilities on important processes will likely not survive. Three common reference points lend some precision to these measures:
� A maximum rate of resource-build customer awareness if they had our product and marketing budgets to work with.
Figure 10.5: Distinct capabilities linked to winning, retaining, and selling to customers
Customer acquisition
capability CustomersCustomer
win rate Customer loss rate
Customer retention capability
Sales per customer per period
Ongoing sales
capability
Total sales per period
Figure 10.5: Distinct capabilities linked to winning, retaining, and selling to customers
© Kim Warren, 2011
C
N
L
R
-
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
163
A team’s capability is then defined as the ratio between the rate at which they are actually building the resource, and the best rate that can be imagined, given one of the benchmarks above.
10.4 The Balanced Scorecard
� The balanced scorecard has transformed the way in which organizations track and steer their performance and is now a popular performance assessment tool among large companies1. The method recognizes that management needs to track a range of measures if it is to sustain strong performance and so includes measures in four categories:
� financial: e.g., revenue growth, margins, profitability, return on capital
� customers: e.g., satisfaction, retention, market share and share of business
� internal processes: e.g., delivery systems, service response and new product introductions
� learning and growth: e.g., employee expertise and staff development
The balanced scorecard process recognizes the interconnectedness between the different activities of the business and the importance of measuring and managing soft factors, such as staff skills and quality. Increased training of support staff about a company’s products, for example, will improve sales effectiveness, which will, in turn, improve sales and margins. However, a particular challenge arises in choosing exactly which factors are important in each of the four domains and which measures to adopt to monitor and control them. A strategic architecture can help identify metrics for the four domains.
Figure 10.6 shows a strategic architecture for a consulting firm. It makes one big simplification in treating all professional staff as a single resource, rather than the several distinct levels of experience and seniority that exist in reality. It also adds an intangible factor—staff expertise. Populating this architecture with time chart data provides the rigorous, integrated numerical measures that a sound balanced scorecard requires. Measures for the financial domain can be extracted from the revenue, cost, and profit region of the architecture, plus others not shown in this
1 Robert S. Kaplan and David P. Norton. The Balanced Scorecard. (Boston: Harvard Business School Press, 1996); Robert S. Kaplan and David P. Norton. Strategy Maps. (Boston: Harvard Business School Press, 2004). See also http://www.balancedscorecard.org.
T
B
S
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
164
limited picture, such as the salary rates of different groups of staff and the costs of information systems. Data and movements in customer-related measures can be extracted from the customer region at the top of the architecture, and we can be safe in the knowledge that these measures are entirely coherent in their causal relationships.
A full strategic architecture would make clear exactly which internal processes are key and enable relevant measures for each to be identified. Finally, learning and growth measures should be tracking the organization’s capabilities rigorously, as defined in this chapter.
10.5 Capabilities in Public Sector and Voluntary Organizations
The frameworks linking capabilities to resource development are directly applicable to noncommercial organizations. Voluntary organization wanting to win new donors, retain existing donors, and raise more money from current donors will likely get better at each of these tasks as it grows in experience. As for the retailer’s location-finding team, this will not simply involve increasing the skills of the individuals involved, but also the development of better procedures and the capture of useful information, such as the socio-demographic profile of donors who are most easily won or who give more often.
The charity supporting patients with terminal illnesses, described in Sections 2.6 and 3.3, also shows how problems can arise as a result of ineffective learning. The
Figure 10.6: Extracting balanced scorecard measures for a consulting firm
ClientsClient win rate
Professional staff
Hiring
Total professional
capacity
Projects per year
Services offeredAdd/cease
services
Client loss rate
Workload per project
Project workload
Workload to build
and develop services
Sales effort
Fee rate charged
Fee incomeTotal
workload
Professional staff cost
[ Other staff, premises and other costs ]
Staff loss rate
Total costs Pressure on staff
Operating profit
$m/year
Quality of work
Staff expertise
Training and
coaching
Financial
Learning and
growth
Customer
Internal processes
Figure 10.5
C
P
S
V
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
C
165
low frequency of calls made by volunteers to the patients they support undermines important mechanisms. Not only do individual volunteers have little opportunity to improve their own skills, but they can also make little contribution to developing the procedures and information required to help the whole team improve.
C
P
S
V
O
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
166
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
CONCLUSIONS AND FURTHER STEPS
This book provides only a brief explanation of how the strategy dynamics approach works and contributes to radically improved strategic management in all kinds of organization. A more extensive reference is Strategic Management Dynamics by Kim Warren (John Wiley & Sons., Inc., 2008).
Further resources
� An online course on strategy dynamics is available at http://sdl.re/sdcourse. The full course consists of 10 classes each corresponding to a chapter in this book, each with 3-5 video segments. The videos include embedded quizzes to help check your understanding. Being quite substantial, this course can be taken in 3 parts. Alternatively, an introductory course can be chosen, consisting only of the summary video segments for each class.
� The easy-to-use Sysdea online software for building and sharing working dynamic business models is at http://.sysdea.com. Numerous examples feature in its Help system at http://docs.sysdea.com.
� For education and training purposes, some “serious games”, built on the principles in this book are available at http://sdl.re/microworlds. These games each provide several strategic challenges, plus simple tools for instructors to manage and assess learners’ use of the games.
Discussion and questions concerning the method can be raised through the Strategy Dynamics Network on LinkedIn.
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
168
APPENDIX 1: TECHNICAL SPECIFICATION
The relationships making up the strategic architecture of an organization and driving its performance over time, developed in Chapters 1 through 4, can be formalized as follows:
Performance P at time t is a function of the quantity of resources R₁ to Rn, discretionary management choices, M, and exogenous factors, E, at that time.
P(t) = f [R₁(t), …Rn(t), M(t), E(t)] (1)
The current quantity of each resource Ri at time t is its level at time t−1 plus or minus any resource flows that have occurred between t-1 and t.
Ri(t) = Ri (t−1) +/− ΔRi(t−1…t) (2)
The change in quantity of Ri between time t-1 and time t is a function of the quantity of resources R₁ to Rn at time t-1, including that of resource Ri itself, on management choices M and on exogenous factors E at that time.
ΔRi(t−1…t) = f [R₁(t−1),…Rn(t−1), M(t−1), E(t−1)] (3)
For these equations to be accurate, the time period must be short enough for the change ΔRi(t-1. . t) to be small relative to the scale of resource Ri. It is equally true that resource quantities tomorrow will be equal to the quantities today plus or minus the rate at which they are currently changing, i.e., Ri(t+1) = Ri (t) +/−ΔRi(t…t+1).
No additional equations are needed to capture the frameworks discussed in Chapters 5 through 9. The only extension required is to distinguish resources held by different competitors, as discussed in Chapter 7. Intangible resources
A
1
T
S
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
169
(Chapter 9) and their relationship with the strategic architecture are already captured by equations (1) to (3).
Capabilities (Chapter 10) can also be encompassed by equations (1) to (3). However, as explained in the text, it is helpful to distinguish capabilities from resources, which results in equation 3 being extended as follows:
The change in quantity of resource Ri between time t–1 and time t is a function of the quantity of resources R₁ to Rn at time t–1, including that of resource Ri itself, on management choices M on exogenous factors E at that time, and on the related capability Ci.
ΔRi(t−1…t) = f [R₁(t−1),…Rn(t−1), M(t−1) , E(t−1), Ci (t−1)] (3a)
The current quantity of capability Ci at time t is its level at time t–1 plus or minus any flows into or out of that capability that have occurred between t–1 and t.
Ci(t) = Ci (t−1) +/− ΔCi(t−1…t) (4)
The change in quantity of capability Ci between time t–1 and time t is a function of the change that occurs to the quantity of the associated resource Ri during that same period.
ΔCi(t−1… t) = f[ΔRi(t−1…t)] (5) T
S
A
1
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
170
APPENDIX 2: PROBLEMS WITH CORRELATION
A big and important consequence arising from the behavior of accumulating stocks is that it messes up our ability to learn much from statistical analysis – especially the simple correlation methods often used to find “explanations” for business performance. The reason for this is that the math of accumulation prevents any possibility of a linear relationship between cause and effect in any situation where an accumulating resource sits between the two.
To illustrate this problem and how it arises, consider a simple manufacturing company that wants to understand whether it should spend more or less money on marketing. Management looks at the company’s recent history and sees the patterns for marketing and operating profit shown below.
Marketing spend and operating profit history for a company
The charts reflect the following history:
� The company was originally dissatisfied with its low and stagnant profits up to month 6.
� A new head of marketing arrived, who persuaded the company to increase its marketing spend from month 7. This increase in expenditure immediately led to lower profits, as the increase in sales was insufficient to cover the extra cost, although profits started to recover over the following nine months.
P
C
A
2
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
171
� In month 15, the head of finance lost patience with the situation, pointing out that the company had seen a total loss of profits over those nine months of more than $500,000, compared with what they could have expected from just pursuing the original low rate of marketing. He even doubted that the previous rate of marketing was necessary.
� The CEO agreed, so the company cut its marketing spend sharply from month 16. Sure enough, profits jumped as the savings in marketing spend were far greater than the value of lost sales. The head of finance was clearly correct and pointed out to his colleagues that he had made the company over $1.5 million of additional profit over the twelve months since his recommendation, compared with the original low profit rate.
The head of marketing was feeling somewhat dejected about this. Convinced that she was right to increase marketing spend, she commissioned some industry research. The company was one of fifty near-identical firms, and luckily, information on monthly marketing spend and profits was available for all of its competitors.
The next figure shows two results from comparing operating profit with marketing spend for this large sample of companies.
The left-hand graph shows how operating profits in any month compare with marketing spend in that month. The head of marketing was not at all happy with this finding, which suggested that profits were negatively correlated with marketing spend. On reflection, she was not too surprised, since marketing would surely take some time to have its effect, and its immediate impact would of course be an increase in cost.
Perhaps operating profits would increase some months after is the increase in marketing spend? The head of marketing then looked at how competitors’ marketing spend compared with operating profits three months later (right-hand chart, above). Disappointingly, there still seemed to be no positive correlation
P
C
A
2
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
172
P
C
A
2
between marketing spend and profits—but at least the negative relationship had disappeared.
In spite of these apparently approximate and perplexing relationships between marketing spend and profit, the business model at work here is totally deterministic, with no uncertainty whatsoever:
operating profit = revenue – production costs – marketing spend – overhead
sales revenue = customers * sales per customer * unit price
production cost = sales in units * variable cost per unit + fixed production cost
customers today = customers last month + customers won – customers lost
sales per customer = base sales per customer + marketing spend * sales increase per marketing dollar
customers lost = five per month
customers won per month = marketing spend * customers won per marketing dollar
Moreover, the cost of acquiring a customer is $50,000. On average, each customer stays for 20 months before being lost to competitors. During that time, the customer generates 2,000 units of sales per month, on which the gross profit is $40, making a total profit contribution of $80,000. There is no ambiguity whatsoever that every marketing dollar generates $1.60 of value in less than two years.
The next figure portrays these relationships diagrammatically. T
h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
173
P
C
A
2
If the business structure is very simple, why could the head of marketing not discover any correlation between marketing spend and profits? The problem lies at the flow-to-stock boundary. There is no obvious relationship between the number of customers in any month and the win rate of customers in that same month—nor should we be particularly surprised at the lack of such a relationship, since today’s number of customers reflects the entire history of customer gains and losses.
This has profound implications for any effort to explain performance outcomes:
� It is unsafe to seek correlation between any possible causal factor and performance outcomes if any accumulating stock exists between the cause and the outcome.
� It is meaningless to seek correlation between asset stocks and the flow rates that determine them—their relationship is precisely defined by the math of integration.
Now consider that a real business consists of many accumulating resources, that those resources move through multiple stages, and that the rate at which any is changing reflects other resource quantities, which themselves are subject to the same difficulty. No wonder correlation analysis rarely discovers anything useful about strategy!
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
Strategy Dynamics Essentials
Kim Warren
Strategy D ynam
ics Essentials - Kim
W arren
While many businesses may be well-managed, poor strategy choices and implementation lead to a perpetual, grinding under-achievement of potential, ill-advised initiatives, or avoidable failures. In other fields of human endeavor, we reduce the risk of serious failure by building models of what we want to do before trying it for real, and codifying how things are supposed to work. Learning from what we do, we revise the models and update those codified processes to improve performance further. But for the most important function of all—figuring out what the enterprise could achieve and how that might be done—most organizations still rely on qualitative judgement and superficial analysis. This is unacceptable.
The strategy dynamics method— in essence, the application of engineering control theory principles to enterprise systems—solves this problem. Strategy Dynamics Essentials explains the frameworks that accomplish this aim and shows how they are applied. The method can be used to build working, quantified models of any enterprise, or any part thereof, of any scale, in any sector—or of any issue that an enterprise may face. The book is written for executives responsible for any aspect of business performance, consultants and other advisors, business students at all levels, and business teachers. No advanced technical skills are needed—just the will and ability to think quantitatively about whatever enterprise or function you are concerned about. Even the earliest principles can be applied right from the start, and this fast ROI can be repeated because key structures can be usefully deployed to tackle strategic challenges in marketing, staffing, and other functions—which is less difficult and costly than trying to assemble coherent strategy and plans from current alternative approaches.
The book is supported by a full, Masters' level on-line course, “serious games” for training and education, and the powerful and easy-to-use Sysdea software for modelling strategy and performance.
In clu
de s F
ig ur
es
in C
ol or
T h is
c o
p y
p ro
vi d e d t
o s
tu d e n t
a t
C S
U C
h ic
o .
N o t
fo r
re d is
tr ib
u tio
n .
© K
im W
a rr
e n 2
0 1 0
- Title Page
- Copyright
- Contents
- About the Author
- Preface
- Chapter 1 - Building Performance Over Time
- 1.1 A Life-Cycle Example: Blockbuster® Inc.
- 1.2 Strategic Management: Positioning versus Delivery
- 1.3 What “Performance” Do We Want To Improve?
- 1.4 Building Future Performance
- 1.5 Nonfinancial Performance Objectives
- 1.6 Levels of Strategy
- 1.7 Example: Low-fare Airline Ryanair
- Chapter 2 - How Resources Drive Performance
- 2.1 Strategy Methods Focusing on External Factors
- 2.2 Strategy Methods Focusing on Firm-specific Factors
- 2.3 Limitations of Common Strategy Approaches
- 2.4 Tangible Resources and Profits
- 2.5 From Performance to Resources
- 2.6 Resources and Nonfinancial Performance
- 2.7 “Stocks” of Resources
- 2.8 When Resources Themselves Are the Objective
- 2.9 Specifying and Quantifying Resources
- Chapter 3 - Resources Won and Lost
- 3.1 Quantifying the “Bathtub Behavior” of Resources
- 3.2 Accumulation over Time
- 3.3 Consequences of Resource Accumulation
- 3.4 Resources Won and Lost by Ryanair
- 3.5 “Accounting” for Resources
- 3.6 Control over the Building and Retaining of Resources
- 3.7 Generic Behaviors and Their Drivers
- Chapter 4 - Interdependence and the Strategic Architecture
- 4.1 Competition and Other External Factors
- 4.2 Existing Resources Drive Gains and Losses
- 4.3 How Resources Drive Their Own Growth and Loss
- 4.4 The Role of Potential Resources
- 4.5 The Strategic Architecture
- 4.6 Functional Issues and Other Objectives
- Chapter 5 - Resource Quality
- 5.1 Size Is Not the Same as Quality
- 5.2 Attributes of Other Resources
- 5.3 When Resources Bring Access to Others
- 5.4 Using the Quality Curve to Beat Competitors
- 5.5 Other Uses for the Quality Curve and Resource Attributes
- Chapter 6 - Developing Resources
- 6.1 Developing Staff
- 6.2 The Customer Choice Pipeline
- 6.3 Product Development
- 6.4 Deteriorating Resources
- 6.5 How Resources Develop in Noncommercial Cases
- 6.6 Boundaries of the Firm
- Chapter 7 - Competitive Rivalry
- 7.1 Type-1 Rivalry
- 7.2 Type-2 Rivalry
- 7.3 Type-3 Rivalry
- 7.4 Further Issues with the Three Types of Rivalry
- 7.5 Competing with Intermediaries
- 7.6 Competing for Other Resources
- 7.7 Rivalry in Noncommercial Cases
- 7.8 Dealing with Multiple Competitors
- Chapter 8 - Steering Strategy and Performance
- 8.1 The Difference between Good and Poor Strategies
- 8.2 Steering Strategy and Performance
- 8.3 Policy to Guide Decisions
- 8.4 Controlling Indirect Decisions and Interference
- 8.5 Conflicting Objectives
- 8.6 Goals and Policy in Noncommercial Cases
- Chapter 9 - Intangible Resources
- 9.1 State-of-Mind Intangibles
- 9.2 Information-based Intangible Resources
- 9.3 Quality-based Intangibles
- Chapter 10 - Capabilities
- 10.1 Dimensions of Capability
- 10.2 Learning Develops Capability
- 10.3 Capabilities Not Linked to Resource-building
- 10.4 The Balanced Scorecard
- 10.5 Capabilities in Public Sector and Voluntary Organizations
- Conclusions and Further Steps
- Appendix 1: Technical Specification
- Appendix 2: Problems with Correlation
- LicName: This copy provided to John Juzbasich, not for redistribution. © Kim Warren 2010