Case study analyze

70 Harvard Business Review | April 2008 | hbr.org

AS THE WORLD’S MAJOR ECONOMIES

have matured, they have become dom-

inated by service-focused businesses.

But many of the management tools

and techniques that service managers

use were designed to tackle the chal-

lenges of product companies. Are these

suffi cient, or do we need new ones? Ja so

n L

e e

by Frances X. Frei

Let me submit that some new tools

are necessary. When a business takes a

product to market, whether it’s a basic

commodity like corn or a highly engi-

neered offering like a digital camera,

the company must make the product

itself compelling and also fi eld a work-

force capable of producing it at an

The Four Things a Service Business Must Get Right

Extensive study of the world’s best service companies reveals the principles on which they’re built.

1084 Frei.indd 701084 Frei.indd 70 3/4/08 10:12:24 PM3/4/08 10:12:24 PM

hbr.org | April 2008 | Harvard Business Review 71

1084 Frei.indd 711084 Frei.indd 71 3/4/08 10:12:31 PM3/4/08 10:12:31 PM

The Four Things a Service Business Must Get Right

72 Harvard Business Review | April 2008 | hbr.org

attractive price. To be sure, neither job is easy to do well;

enormous amounts of management attention and academic

research have been devoted to these challenges. But deliver-

ing a service entails something else as well: the management

of customers, who are not simply consumers of the service

but can also be integral to its production. And because cus-

tomers’ involvement as producers can wreak havoc on costs,

service companies must also develop creative ways to fund

their distinctive advantages.

Any of these four elements – the offering or its funding

mechanism, the employee management system or the cus-

tomer management system – can be the undoing of a ser-

vice business. This is amply demonstrated by my analysis

of service companies that have struggled over the past de-

cade. What is just as clear, however, is that there is no “right”

way to combine the elements. The ap-

propriate design of any one of them

depends upon the other three. When

we look at service businesses that have

grown and prospered – companies like

Wal-Mart in retail, Commerce Bank in

banking, and the Cleveland Clinic in

health care – it is their effective inte-

gration of the elements that stands out

more than the cleverness of any ele-

ment in isolation.

This article outlines an approach for crafting a profi table

service business based on these four critical elements (col-

lectively called the “service model”). Developed as a core

teaching module at Harvard Business School, this approach

recognizes the differences between service businesses and

product businesses. Students in my course learn to think

about those differences and their implications for manage-

ment practice. Above all, they learn that to build a great

service business, managers must get the core elements

of service design pulling together or else risk pulling the

business apart.

1 The Offering The challenge of service-

business management be-

gins with design. As with

product companies, a ser-

vice business can’t last long

if the offering itself is fatally

fl awed. It must effectively

meet the needs and desires of an attractive group of cus-

tomers. In thinking about the design of a service, however,

managers must undergo an important shift in perspective:

Whereas product designers focus on the characteristics buy-

ers will value, service designers do better to focus on the

experiences customers want to have. For example, customers

may attribute convenience or friendly interaction to your

service brand. They may compare your offering favorably

with competitors’ because of extended hours, closer proxim-

ity, greater scope, or lower prices. Your management team

must be absolutely clear about which attributes of service

the business will compete on.

Strategy is often defi ned as what a business chooses not

to do. Similarly, service excellence can be defi ned as what a

business chooses not to do well. If this sounds odd, it should.

Rarely do we advise that the path to excellence is through in-

ferior performance. But since service businesses usually don’t

have the luxury of simply failing to deliver some aspects of

their service – every physical store must have employees

on-site, for example, even if they’re not particularly skilled

or plentiful – most successful companies choose to deliver a

subset of that package poorly. They don’t make this choice

casually. Instead, my research has shown, they perform badly

at some things in order to excel at others. This can be consid-

ered a hard-coded trade-off. Think about the company that

can afford to stay open for longer hours because it charges

more than the competition. This business is excelling on con-

venience and has relatively inferior performance on price.

The price dimension fuels the service dimension.

To create a successful service offering, managers need

to determine which attributes to target for excellence and

which to target for inferior performance. These choices

should be heavily informed by the needs of customers.

Managers should discover the relative importance custom-

ers place on attributes and then match the investment in

excellence with those priorities. At Wal-Mart, for example,

ambience and sales help are least valued by its customers,

low prices and wide selection are most valued, and several

other attributes rank at points in between. (See the exhibit

“Wal-Mart’s Value Proposition” in David J. Collis and Michael

G. Rukstad’s article “Can You Say What Your Strategy Is?” in

this issue.) The trade-offs Wal-Mart makes are deliberately

informed by these preferences. The company optimizes spe-

cifi c aspects of its service offering to cater to its customers’

Frances X. Frei ([email protected]) is an associate professor of busi-

ness administration in the Technology and Operations Management

unit at Harvard Business School in Boston.

Service excellence can be defi ned as what a business chooses not to do well.

1084 Frei.indd 721084 Frei.indd 72 3/4/08 10:12:40 PM3/4/08 10:12:40 PM

hbr.org | April 2008 | Harvard Business Review 73

priorities, and it refuses to overinvest in underappreciated

attributes. The fact that it takes a drubbing from competi-

tors on things its customers care less about drives its overall

performance.

The phenomenon, of course, has a circular aspect. Shop-

pers whose preferences match Wal-Mart’s strengths self-

select into its customer base. Meanwhile, those who don’t

prefer Wal-Mart’s attributes buy elsewhere. It is important

therefore to identify customer segments in terms of attri-

bute preferences – or as some marketers prefer, in terms

of customer needs. Identifying what might be called cus-

tomer operating segments is not the same exercise as tradi-

tional psychographic segmentation. Rather than stressing

differences that enable increasingly targeted and potent

messaging, this type of segmentation aims to fi nd popula-

tions of customers who share a notion of what constitutes

excellent service.

Once an attractive customer operating segment is found,

the mission is clear: Management should design a new offer-

ing or tweak an existing one to line up with that segment’s

preferences. Look, for example, at the fi t achieved by Com-

merce Bank, which has been able to grow its retail customer

base dramatically even though its rates are among the worst

in its markets and it has made limited acquisitions. Com-

merce Bank focuses on the set of customers who care about

the experience of visiting a physical branch. These customers

come in all shapes and sizes – from young, fi rst-time bank-

ing clients to time-strapped urban professionals to elderly

retirees. As an operating segment, however, they all believe

that convenience is a bank’s most important attribute and

choose Commerce Bank because of its evening and week-

end hours. Second most important to them is the friendli-

ness of interactions with employees, and so the promise of

a cheerful, familiar teller has become part of the bank’s

core offering. Commerce has added to its branch ambience

with interior elements both lovely (high ceilings and natural

light) and fun (an amusing contraption for redeeming loose

change). When it comes to attributes less important to the

bank’s customers – price and product range – management

is willing to cede the battle to competitors.

It is tempting to think, “If I’m a really good manager,

then I don’t have to cede anything to the competition.” This

well-intentioned logic can lead, ironically, to not excelling

at anything. The only organizations I have seen that are

superior at most service attributes demand a price premium

of 50% over their competitors. Most industries don’t support this type of premium, and so trade-offs are necessary. I like

to tell managers that they are choosing between excellence

paired with inferior performance on one hand and medi-

ocrity across all dimensions on the other. When managers

understand that inferior performance in one dimension fu-

els superior performance in another, the design of excellent

service is not far behind.

2 The Funding Mechanism All managers, and even most cus-

tomers, agree that there is no

such thing as a free lunch. Excel-

lence comes at a cost, and the

cost must ultimately be covered.

With a tangible product, a compa-

ny’s mechanism for funding superior

performance is usually relatively simple: the price tag. Only

the customers who forfeit the extra cash can avail them-

selves of the premium offering. In a service business, devel-

oping a way to fund excellence can be more complicated.

Many times, pricing is not transaction based but involves

the bundling of various elements of value or entails some

kind of subscription, such as a monthly fee. In these cases,

buyers can extract uneven amounts of value for their money.

Indeed, even nonbuyers may derive value in certain service

environments. For example, a shopper might spend time

learning from a knowledgeable salesperson, only to leave

the store empty-handed.

In a service business, therefore, management must give

careful thought to how excellence will be paid for. There

must be a funding mechanism in place to allow the company

to outshine competitors in the attributes it has chosen. In my

study of successful service businesses, I’ve seen the funding

mechanism take four basic forms. Two are ways of having

the customer pay, and two cover the cost of excellence with

operational savings.

Charge the customer in a palatable way. The classic ap- proach to funding something of value is simply to have the

customer pay for it, but often it is possible to make the form

that payment takes less objectionable to customers. Rarely

is that done with à la carte pricing for the niceties. A large

part of Starbucks’s appeal is that a customer can linger al-

most indefi nitely in a coffeehouse setting. It’s unthinkable

that Starbucks would place meters next to its overstuffed

chairs; a better way to fund the atmosphere is to charge

more for the coffee. Commerce Bank is open late and on

weekends – earning it high marks on extended hours – and

it pays for that service by giving a half percentage point less

in interest on deposits. Could it fund the extra labor hours

by charging for evening and weekend visits? Perhaps, but a

slightly lower interest rate is more palatable. Management

in any setting would do well to creatively consider what

feels fair to its customers. Often, the least creative solution

is to charge more for the particular service feature you are

funding.

Create a win-win between operational savings and value-added services. Very clever management teams dis- cover ways to enhance the customer experience even while

spending less (fi nding, in other words, that there can be such

a thing as a free lunch). Many of these innovations provide

1084 Frei.indd 731084 Frei.indd 73 3/4/08 10:12:47 PM3/4/08 10:12:47 PM

The Four Things a Service Business Must Get Right

74 Harvard Business Review | April 2008 | hbr.org

only a temporary competitive advantage, as they are quickly

recognized and copied. Some are surprisingly durable, how-

ever. An example is the immediate-response service provided

by Progressive Casualty Insurance. When someone insured by

Progressive is involved in an auto accident, the company im-

mediately sends out a van to assist that person and to assess

the damage on the spot – often arriving on the scene before

the police or tow trucks. Customers love this level of respon-

siveness and give the company high marks for service. But

in anticipation of such a need someday,

would they pay more in insurance pre-

miums? Unfortunately, no. People are

pathologically price sensitive about car

insurance and almost never select any-

thing but the rock-bottom quote. The

key to Progressive’s ability to fund this

service is the cost savings it ultimately

yields. Normally insurance providers

are subject to fraud, with criminals

making claims for accidents that were

staged or never happened. Because of these and other types

of disputed claims, fi rms also incur high legal fees – which,

combined with the other costs of fraud, add up to some $15 out of every $100 in insurance premiums across the industry. Since deploying its vans, Progressive has seen costs in both

categories plummet. Sending a company representative to

the scene pays for itself.

Progressive offers another customer convenience that

many competitors have so far shied away from: giving

quotes from other providers alongside its own when a po-

tential buyer inquires about the cost of insurance. It’s not

that Progressive is determined to go one better than rivals

to win the business. In fact, Progressive’s is the lowest quote

only about half the time. What Progressive does believe is

that its quote is the right one given the probability of that

person’s getting into an accident – a probability that the in-

surer is best in class at determining. If indeed its quote is

spot-on, then allowing a competitor to insure the customer

at a lower rate is doubly effective: It frees Progressive from

a money-losing proposition while burdening its competitor

with the unprofi table account. Thus a level of service that

looks downright altruistic to the customer actually benefi ts

the company. This is an example of leveraging operations

into a value-added service.

How can your management team fi nd win-win solutions

of its own? When I pose this question to managers, their

impulse is to imagine what new value could be created for

customers and then to ponder how that could be funded

through cost savings. I suggest beginning instead by asking,

“Where are our biggest cost buckets?” With these in mind,

managers can then simultaneously determine how to reduce

costs and create a value-added service. A good fi rst place

to look? Anywhere that time is a large component of cost.

Removing time is often fruitful, since it can directly improve

service even as it cuts costs.

Spend now to save later. Often it is possible, if somewhat painful, to make operational investments that will pay off

eventually by reducing customers’ needs for auxiliary ser-

vice in the future. A classic example is Intuit’s decision to

provide free customer support, in defi ance of the software

industry norm. Call centers are expensive to staff because

of the combination of technical knowledge and sociability

required to fi eld inquiries effectively. Customers meanwhile

are extremely uneven in their neediness vis-à-vis informa-

tion technology. For most software makers this adds up to

the obvious conclusion that customers should be charged

for support.

Intuit founder Scott Cook sees the matter differently.

Those needy calls, he believes, are a useful form of input

to continued product development – the engine of future

revenues – and that justifi es an even greater expense outlay.

Intuit has its higher salaried product-development people,

not solely customer service people, fi elding calls so that sub-

sequent versions of its offerings will be informed by direct

knowledge of what users are trying to accomplish and how

they are being frustrated. This is part of a broader commit-

ment to feedback-driven improvement that Cook refers to as

“DIRST” – for “do it right the second time.” The investment

has paid off in better software, which means a lower call

volume. “Our competition thinks we’re crazy,” Cook says, and

he understands why. “If we got as many calls as they do, we’d

be out of business.”

Have the customer do the work. One other type of fund- ing mechanism for enhanced service puts the cost back in

the customer’s court, but in the form of labor. Offering self-

service, from pump-your-own gas to self-managed broker-

age accounts, is a well-established way to keep costs low. If

the goal is service excellence, though, you must create a

situation in which the customer will prefer the do-it-yourself

capability over a readily available full-service alternative.

Airlines have achieved this, at last, with fl ight check-in kiosks,

although the value proposition they initially presented was

dubious. At fi rst, passengers felt compelled to use the rela-

tively unappealing kiosks only because carriers had allowed

the lines in front of manned desks to become intolerable.

If a self-service option is truly preferable, customers should be willing to take on the work for nothing or even pay for the privilege.

1084 Frei.indd 741084 Frei.indd 74 3/4/08 10:12:52 PM3/4/08 10:12:52 PM

hbr.org | April 2008 | Harvard Business Review 75

Today, however, frequent fl iers prefer the kiosks because they

provide readier access to useful tools like seat maps. Businesses

looking to achieve service excellence in other settings

should not take such an indirect route. They should set

themselves the challenge of creating self-service capabilities

that customers will welcome. Indeed, if a self-service option

is truly preferable, customers should be willing to take on

the work for nothing or even pay for the privilege. When

managers designing self-service solutions are not permitted

to add the inducement of price discounts, they are forced to

focus on improving the customer experience.

Whatever funding mechanism is used to cover the costs

of excellence, it is best thought out as thoroughly as possible

prior to the launch of a new service, rather than amended

in light of experience afterward. When a service that’s been

perceived as free suddenly has fees associated with it, cus-

tomers tend to react with disproportionate displeasure. And

since companies cannot thrive by offering service gratis, it

is vital that they not set expectations that can’t be sustained.

With careful analysis and design, a company can offer and

fund a better service experience than its customers would

enjoy elsewhere.

3 The Employee Management System Companies often live or die on the quality of their workforces,

but because service businesses are typically people intensive,

a relative advantage in employee management has all the

more impact there. Top management must give careful at-

tention to recruiting and selection processes, training, job

design, performance management, and other components

that make up the employee management

system. More to the point, the decisions

made in these areas should refl ect the

service attributes the company aims to

be known for.

To design a well-integrated employee

management system, start with two

simple diagnostic questions. First: What

makes our employees reasonably able to

achieve excellence? And then: What makes our employees

reasonably motivated to achieve excellence? Thoughtfully

considered, the answers will translate into company-specifi c

policies and programs. Companies that neglect to connect

the dots between their employee management approaches

and customers’ service preferences will fi nd it very hard to

honor their service promises.

At one large international retail bank I studied, a se-

nior manager had come to a depressing realization. “Our

service stinks,” she told me. Under her guidance the bank

took various measures, mainly centering on incentives and

training, but the problem persisted. Customer experience

in the branch did not improve. Perplexed but determined,

the executive decided to become a frontline employee

herself for a month. She thought it would take that much

time to experience a typical range of service interactions

and see the roots of the problem. In fact, it took one day.

“From the time the doors opened, customers were yelling

at me,” she reported. “By the end of the day, I was yelling

back.” What became clear was that employees were set up

to fail. Recent cross-selling initiatives had created a set of

customers with more complex needs and higher expecta-

tions for their relationship with the bank, but employees

had not been equipped to respond. As a result of decisions

made by the management team (all individually sensible),

the typical employee did not have a reasonable chance

of succeeding. The bank’s employee management system

was broken.

If your business requires heroism of your employees to

keep customers happy, then you have bad service by de-

sign. Employee self-sacrifi ce is rarely a sustainable resource.

Instead, design a system that allows the average employee

to thrive. This is part of Commerce Bank’s competitive for-

mula. Recall that the bank chooses to compete on extended

hours and friendly interactions and not on low price and

product breadth. Now think how that strategy could inform

employee management; the implications are not hard to

imagine. For instance, Commerce concluded that it didn’t

require straight-A students to master its limited product

set; it could hire for attitude and train for service. In job

interviews, its managers could use simple weed-out criteria –

like “Does this person smile in a resting state?” – rather

than trying to maximize across a wide range of positive

characteristics. The bank’s current employees could be de-

ployed as talent scouts, on the principle that it takes one

to know one. (When people from Commerce see someone

providing great service in another setting, whether at a res-

taurant or at a gas station, they hand out a card printed

with a compliment and a suggestion to consider working

for Commerce.)

It’s a simple reality that employees who are above aver-

age in both attitude and aptitude are expensive to employ.

They are not only attractive to you but also attractive to

your competitors, which drives up wages. A business that

wants to maintain a competitive cost structure will prob-

ably need to compromise on one quality or the other (or,

if it insists on having both, fi nd a way to fund that luxury).

If, as Commerce Bank does, you choose to hire for attitude,

then you must engineer things so that even lower-aptitude

employees will reliably deliver great service. Like managers

who don’t want to admit that their service is designed to be

inferior on some attributes, many people are reluctant to

acknowledge a trade-off between aptitude and attitude. But

failure to accommodate this economic reality in the design

of the employee management system is a common culprit

in fl awed service.

1084 Frei.indd 751084 Frei.indd 75 3/4/08 10:12:58 PM3/4/08 10:12:58 PM

The Four Things a Service Business Must Get Right

76 Harvard Business Review | April 2008 | hbr.org

4 The Customer Management System In a service environment, employees aren’t the only people

affecting the cost and quality of service delivered. The cus-

tomers themselves can be in-

volved in operational processes,

sometimes to a very large ex-

tent, and their input infl uences

their experiences (and often

other customers’ too). For ex-

ample, an architectural fi rm’s

client may explain the purpose

of a new facility well or poorly,

and that will affect the effi ciency of the design process and

the quality of the end product. A customer who dithers at

a fast-food counter makes the service less fast for everyone

behind him.

Customer involvement in operations has profound im-

plications for management because it alters the traditional

role of the business in value creation. The classic product-

based business buys materials and adds value to them in

some way. The enhanced-value product is then delivered

to customers, who pay to receive it. In a service business,

however, employees and customers are both part of the

value-creation process. A main benefi t is that customer

labor can be far less expensive than employee labor. It

can also lead to better service experiences. When students

participate more in a classroom environment, for example,

they learn more. But there are challenges, as well. Designing

a system that explicitly manages these challenges is essential

to service success.

Consider the issue of customer selection. Service designs

may call for customers to perform important tasks, but for

the most part customers have no interview, no background

check, and no personality profi le. As a former senior execu-

tive from Nestlé now working in fi nancial services put it,

“I could control who was in my factory at Nestlé; I have no

such control over the customers in my bank’s branches.”

In addition, despite many organizations’ best efforts, cus-

tomers are not as easy to train as employees. There are usu-

ally many times more customers than employees, and creat-

ing effective training materials for such a large, dispersed,

unpaid, and often irrelevantly skilled workforce is diffi cult.

When this holds true, fi rms must accommodate the limited

training in the design of the service experience. If tasks are

shifted from employees to customers – from higher-skilled

to lower-skilled people – then they must be adjusted accord-

ingly. Airlines seem to get this right. Recall (if you can) the

last time you checked in with an agent at the full-service

counter. Chances are you witnessed the agent complete a

dizzying sequence of keystrokes. It would not seem reason-

able to expect customers to perform these same steps, and so

when the check-in role was transferred to customers, it was

dramatically simplifi ed. By contrast, think of the self-service

supermarket checkout. Here customers are asked not only

to do what trained employees have done previously but also

to shoulder the additional responsibility of fraud preven-

tion through a complicated process of weighing bags. Asking

customers to perform more-complicated tasks than higher-

skilled employees contributes to the disarray and anxiety

that surrounds these checkout lines.

Customers also have a great deal of discretion in their op-

erational activities, usually far more than employees. When

a company introduces a new process that it wants employ-

ees to use, it can simply issue a mandate. When customers

are involved, transitions like this can be signifi cantly more

complicated. Look at Zipcar, the popular car-sharing service.

To keep costs low, its service model depends on customers

to clean, refuel, and return cars in time for the next user.

Motivating employees to perform these tasks would be rou-

tine; motivating customer-operators has required a complex,

evolving mix of rewards and penalties.

In managing customers in your operations, then, you’ll

need to address a few key questions: Which customers

are you focusing on? Which behaviors do you want? And

which techniques will most effectively infl uence behavior?

For example, a company whose business model depends on

customers’ timeliness – whether it’s a dental offi ce packing

its appointment calendar or a video store circulating hit

fi lms – may use more- or less-heavy-handed tactics to ensure

compliance. In a previous article for Harvard Business Review

(“Breaking the Trade-Off Between Effi ciency and Service,”

November 2006), I related lessons from several companies

that have used a range of techniques to modify customer

behavior. These techniques can be divided into two basic cat-

egories: instrumental (the carrots and sticks we commonly

see play out as discounts and late fees) and normative (the

use of shame, blame, and pride to motivate us to return

shopping carts and pick up trash even when no one is look-

ing). The important thing is to manage customers in a way

that is consistent with the service attributes you’ve chosen

to emphasize overall.

Integrating the Elements Successful service companies have a working plan that incor-

porates all four elements of service design. Within each of

those areas, however, it is hard to spot any best practice. This

is because the whole business depends more on the intercon-

nection of the four than on any one element.

A standout example of effective overall integration is

the Cleveland Clinic, which is consistently ranked among

America’s most eminent hospitals and has been a leader in

pioneering cardiac care for decades. It’s hard to put a fi nger

on the source of that advantage. The fact that the clinic

has specialty centers focusing on diabetes, for example, or

cardiac care is not exceptional in itself. Its refusal to attach

1084 Frei.indd 761084 Frei.indd 76 3/4/08 10:13:04 PM3/4/08 10:13:04 PM

hbr.org | April 2008 | Harvard Business Review 77

fi nancial rewards to doctors’ productivity is unusual but

might not be effective elsewhere. Step back from the de-

tails, however, and the bigger picture emerges. Attracting the

highest-severity patients means that doctors will always face

a challenging environment in need of innovative solutions.

Organizing into disease centers rather than narrower, more

traditional lines of specialization (such as kidneys or blood)

sets the stage for cross-disciplinary collaboration – and thus

for novel perspectives – within those centers. Removing

productivity incentives gives doctors license to spend time

on innovation, which is enhanced by their close work with

specialists from other fi elds. The particular choices made on

methods, processes, and personnel are the right ones for the

Cleveland Clinic because they complement one another and

come together in a smoothly operating system.

Any service company, no matter how long established, can

benefi t from a review of its operations using the framework

laid out in this article. Bringing the four elements of service

design into tighter alignment can be an ongoing process of

small tweaks and experiments in change, inspired by the

kinds of questions included in the sidebar “Diagnosing Ser-

vice Design.” A management team planning to launch a new

service will fi nd the framework particularly helpful. It fl ags

the decisions that should be made early and in tandem so

that they don’t clash down the road. And at the highest level,

it underscores two very important principles of service de-

sign. First, there is no such thing as a good idea in isolation;

there is only a good idea in the context of a specifi c service

model. Second, it is folly to attempt to be all things to all

customers.

The fi rst point notes the importance of fi t, mentioned

earlier as a key strength of the Cleveland Clinic. At the clinic,

management knows that extensions to its core business must

be examined closely for their fi t with its existing service

model. The organization recently abandoned the concept of

a high-end wellness and spa offering because it didn’t build

on the hospital’s core operational strengths. In some ways

this seems like an obvious point, but managers often stray

into areas of relative weakness, particularly when they see a

fi rm they consider to be a direct competitor succeeding with

a service they don’t yet offer. Progressive made this mistake

when it decided to venture into the home insurance market.

No question, there is money to be made in home insurance,

as innumerable fi rms have shown. But Progressive failed in

its attempt because the challenges of that business did not

match up with the company’s competitive strengths. Recall

that Progressive is justifi ably proud of its analytics advan-

tage, which enables it to effectively size up the risk that a

given policyholder will fi le a claim. Unfortunately, that kind

of actuarial prowess is not as central to making a profi t on

insuring homes. Home insurers rise or fall on the manage-

ment of their investment portfolios – and that is a relative

weakness of Progressive. (Firms typically lose money on the

THE SUCCESS OR FAILURE of a service busi- ness comes down to whether it gets four things right or wrong – and whether it balances them effectively. Here are some questions that will sharpen managers’ thinking along each dimension and help

companies gauge how well their service models are integrated.

The Offering Which service attributes (convenience? friendliness?) does the fi rm target for excellence?

Which ones does it compromise in order to achieve excellence in other areas?

How do its service attributes match up with targeted customers’ priorities?

The Funding Mechanism Are customers paying as palatably as possible?

Can operational benefi ts be reaped from service features?

Are there longer-term benefi ts to current service features?

Are customers happily choosing to perform work (without the lure of a discount) or just trying to avoid more-miserable alternatives?

The Employee Management System What makes employees reasonably able to produce excellence?

What makes them reasonably motivated to produce excellence?

Have jobs been designed realistically, given employee selection, training, and motivation challenges?

The Customer Management System Which customers are you incorporating into your operations?

What is their job design?

What have you done to ensure they have the skills to do the job?

What have you done to ensure they want to do the job?

How will you manage any gaps in their performance?

Diagnosing Service Design

The Whole Service Model Are the decisions you make in one dimension supported by those you’ve made in the others?

Does the service model create long-term value for customers, employees, and shareholders?

How well do extensions to your core business fi t with your existing service model?

Are you trying to be all things to all people – or specifi c things to specifi c people?

2

1

3

4

1084 Frei.indd 771084 Frei.indd 77 3/4/08 10:13:09 PM3/4/08 10:13:09 PM

The Four Things a Service Business Must Get Right

insurance but make money investing pre-

paid premiums.) The fi t, in retrospect,

was a bad one. It should have been seen

that way early on.

Just as common a failing is the mis-

guided desire to be all things to all

people. In today’s service economy, it is

nearly impossible to design a service model

to cover a huge range of customers and re-

main competitive across them. Instead, fi rms should

design their service models for more targeted excellence

by being specifi c things to specifi c people.

Great service companies are, almost without excep-

tion, very clever about selecting their customers. We saw

this in Progressive’s highly informed choice of whom

to do business with. Commerce Bank, from its begin-

nings in 1973, knew it should stake out its own claim on

the market. “The world,” its founder Vernon Hill said,

“did not need another ‘me-too’ bank. I had no capital,

no brand name, and I had to search for a way to dif-

ferentiate from the other players.” Shouldice Hospital, a

Canadian specialist in hernia operations, is highly selec-

tive about its customer base. Not only does it serve just

patients experiencing a certain type of ailment, it has the

luxury of operating on otherwise healthy people. It has

skimmed the cream of the market.

Becoming a Multifocused Firm Inevitably, companies that attempt to be all things to all

people begin to struggle when upstart competitors like

Shouldice start picking off profi table niches. Often, the

decline is not taken seriously until it’s too late. (See the

sidebar “Coming to Terms with the Threat.”)

However, some incumbents have managed to com-

pete effectively with their more-focused rivals, and there

is much to learn from their experience. The common

thread in their competitive responses to upstarts is

the capacity to become “multifocused.” In other words,

they stopped trying to cover the entire waterfront with

a single service model. Instead they pursued multiple

niches with optimized service models – each designed to

achieve excellence on some dimensions at the expense of

inferior performance on others. The secret to success in

a multifocused fi rm is the ability to benefi t from having

various service models under one house umbrella. This

benefi t often comes in the form of shared services (that

is, internal service providers), which enable a fi rm to

generate economies of scale and economies of experi-

ence across its service models. Effectiveness at utilizing

shared services to the advantage of the individual service

models can determine the success of a multifocused fi rm.

(See the exhibit “Are Focused Competitors Nipping at

Your Flanks?”)

HIGHLY FOCUSED FIRMS are the bane of big, established companies. Because they laser-target certain customer segments, they are able to optimize their service mod- els. The service quality they provide, using specialized employees and a customized product set, is potent. By contrast, incumbent fi rms typically attract a mix of custom- ers, hire and develop a variety of employees, and – as a result of excellent, well-intentioned suggestions from both groups – are rampant product proliferators.

Covering the waterfront like this can dilute your excel- lence in every area. Companies that try to do it all…

…are vulnerable to attacks by highly focused entrants, who pick off niche businesses.

Your best defense is to concentrate on multiple niches, shoring them up with the economies gained through internal shared services.

Are Focused Competitors Nipping at Your Flanks?

Focused firms

Nonshared (model-specific) services

Shared services Finance Purchasing IT HR Executive training

Customer facing

Back office

A

B

C

1084 Frei.indd 781084 Frei.indd 78 3/4/08 10:13:15 PM3/4/08 10:13:15 PM

hbr.org | April 2008 | Harvard Business Review 79

The shared services architecture can

be seen in multifocused corporations

across industries – from Yum Brands,

a collection of fi ve fast-food companies,

to Omnicom, which consists of hun-

dreds of companies in the interactive-

marketing space, to GE, which seems

to have no limit on the markets it can

enter. Each corporation has created

distinct service models for distinct cus-

tomer operating segments and gauges

the overall benefi t of the models by as-

sessing how much they gain from one

another. What determines whether a

company has assembled the right port-

folio of service models? It comes down

to a critical test: Is each of the fi rm’s

distinct service models better off as a

result of the others? If the answer is no,

it signals that performance is about to

decline or that the company may want

to spin off some service models. If the

answer is yes, it’s almost always thanks

to superior management of shared ser-

vices, and the incumbent thrives.

The services shared in multifocused

companies typically include business

functions like fi nance, purchasing, in-

formation technology, human resources,

and executive training. The scale advan-

tages they provide are straightforward

and include pooled purchasing, pre-

ferred access to credit, and other cost-

related benefi ts. Economies of experi-

ence are more diffi cult to realize but

can also be more valuable. Here, the

challenge is to use knowledge gained

in one service model to strengthen the

performance of the others. To a limited

extent, this kind of knowledge transfer occurs informally;

this has always been the hope and promise of diversifi ed

companies. The important difference in successful multi-

focused fi rms is that they formalize the process, designing

very explicit ways of leveraging experience across service

models. Knowledge transfer is facilitated by deliberate in-

vestments in such programs as formal best-practice sharing;

centralized, dynamic employee training; and the rotation of

managers among models.

My research convinces me that the best means of sus-

taining growth in a service business is to employ the multi-

focused model, yet it is also evident that this model requires

concentrated effort to defend. Leaders of individual service

models constantly assert that dedicated, rather than shared,

resources would do more to strengthen their own businesses.

Operations managers, meanwhile, raise a chorus of complaint

that shared services require more- vigilant control “below the

line” if they are to deliver the necessary economies of scope

and experience. Given the perpetual assault on the model,

it may not be surprising that another common characteristic

of successful multi focused fi rms is directive (even autocratic)

leadership. This leadership style accommodates different per-

sonalities, but it always relies on senior managers who are

able and willing to exert strong infl uence on subordinates.

They must be, in order to balance the competitive autonomy

of individual service models with the collective value of shared

services. Without strong, centralized leadership, revenue-

generating line managers typically overrule shared-services

HOW DO INCUMBENTS REACT

when a focused entrant appears on the scene? The usual response seems to follow four distinct stages of “strategic grief.”

Dismissal. The incumbent perceives the entrant as a non- threat. It is a deceptively easy assessment to make, given that the focused fi rm has optimized its service model to be deliber- ately good – and bad – at certain aspects of the incumbent’s business.

Sadness. Next comes a sense of loss as profi table customers start to defect. They are willing, if not eager, to make the trade-offs inherent in the entrant’s service model.

Relief. Sadness is replaced by relief as the realization dawns that only one of the incumbent’s cus- tomer segments is being targeted by the focused entrant. The new competitor may win on a few di- mensions of value and take certain customers away, but there are still many other segments to serve!

Dread. Finally, the larger threat reveals itself. The problem is not this single entrant; it’s the inevi- table attack of focused fi rms on other fronts.

Spotted in time, the threat of focused competitors can be met effectively. Is there a troubling area of competitive activ- ity on your radar screen? If so, don’t be lulled by its small scale and isolation. Move quickly to understand what’s going on there. In particular, focus on the entrant’s rate of improvement along critical measures like market share, share of wallet, and service quality. Benchmarks of absolute difference can fool managers into believing that the threat is not im- minent. But when a new competi- tor improves faster than you do, the gap soon closes.

Once you learn the threat is real, explore your potential advantages. Can you compete effectively as a “multifocused” fi rm (one that targets multiple niches rather than trying to tackle everything)? The threat needs to be addressed with humility. The temptation will be great to believe that “our way” remains the better way. If anything, overstate the fact that it is not, and proceed from that assumption to craft a competitive response.

Coming to Terms with the Threat

1084 Frei.indd 791084 Frei.indd 79 3/4/08 10:13:20 PM3/4/08 10:13:20 PM

The Four Things a Service Business Must Get Right

80 Harvard Business Review | April 2008 | hbr.org

managers, particularly in moments of strategic distress. In-

deed, companies often stack the deck by placing stronger

leaders in the service models than in the shared services,

effectively undermining the performance of the system.

The Management-Practice Frontier Management scholars, and not a few practitioners, have

taken up an interesting debate in recent years: Is the dis-

cipline of management fundamentally different in service

businesses than in product businesses? The way in which

management is studied and taught in graduate business

schools was forged in the context of the industrial economy.

Are the approaches that worked for manufacturing compa-

nies equally applicable to services?

As service businesses continue to innovate, succeed, and

be studied, the answers are becoming clearer. The frame-

work presented here suggests why the traditional techniques

have proved as durable as they have and why they still leave

sophisticated managers wanting more. Much of what de-

termines the health of a product business – the soundness

of its offering and the management of its people – is just

as indispensable in a service business and can be addressed

with a similar tool kit. But whole new areas involving

the roles of customers have opened up, and their tool kits

are only now being assembled.

Reprint R0804D

To order, see page 139.

“I’m here to defrag the Magic 8 Ball.” D av

e C

ar p

e n

te r

1084 Frei.indd 801084 Frei.indd 80 3/4/08 10:13:27 PM3/4/08 10:13:27 PM

Harvard Business Review Notice of Use Restrictions, May 2009

Harvard Business Review and Harvard Business Publishing Newsletter content on EBSCOhost is licensed for

the private individual use of authorized EBSCOhost users. It is not intended for use as assigned course material

in academic institutions nor as corporate learning or training materials in businesses. Academic licensees may

not use this content in electronic reserves, electronic course packs, persistent linking from syllabi or by any

other means of incorporating the content into course resources. Business licensees may not host this content on

learning management systems or use persistent linking or other means to incorporate the content into learning

management systems. Harvard Business Publishing will be pleased to grant permission to make this content

available through such means. For rates and permission, contact [email protected].