Journal entry

90 HARVARD BUSINESS REVIEW

The way you set prices doesn't

just influence demand. It also guides

the way buyers use your product or

service-and that can have a lasting

impact on customer relationships.

Pricing AND THE ^ ^

j ^ ^ PSYCHOLOGY OF . •

Consumption by John Gourville and Dilip Soman

A SK ANY EXECUTIVE how pricing policies influence / \ the demand for a product or service, and you'll

/ \ get a confident, well-reasoned reply. Ask that same executive how pricing policies affect consump- t i o n - t h e extent to which customers use products or ser- vices that they've paid for - and you'll get a muted re- sponse at best. We find that managers rarely, if ever, think about consumption when they set prices-and that can be a costly oversight.

Consider this example. Two friends, Mary and Bill, join the local health club and commit to one-year member- ships. Bill decides on an annual payment p l a n - $ 6 0 0 at the time he signs up. Mary decides on a monthly pay- ment plan - $50 a month. Who is more likely to work out on a regular basis? And who is more likely to renew the membership the following year?

Almost any theory of rational choice would say they are equally likely. After all, they're paying the same

SEPTEMBER 2002

amount for the same benefits. But our research shows that Mary is much more likely to exercise at the club than her friend. Bill will feel the need to get his money's worth early in his membership, but that drive will lessen as the pain of his $600 payment fades into the past. Mary, on tbe other hand, will be steadily reminded of the cost of her membership because she makes payments every month. She will feel the need to get her money's worth throughout the year and will work out more regularly. Those regular workouts v̂ àll lead to an extremely impor- tant result from the health club's point of view: Mary will be far more likely to renew her membership when the year is over.

For many executives, the idea that they should draw consumers' attention to the price that was paid for a prod- uct or service is counterintuitive. Companies have long sought to mask the costs of their goods and services in order to boost sales. And rightly so-if a company fails to

91

Pricing and the Psychology of Consumption

make the initial sale, it won't have to worry about con- sumption. To promote sales, health club managers en- courage members to get the payment out of the way early; HMOs encourage automatic payroll deductions; and cruise lines bundle small, specific costs into a single, all-inclusive fee.

However, executives may be discouraging consumption when they apply those pricing practices. People are more likely to consume a product when they are aware of its cost-when they feel "out of pocket." But common pricing practices such as advance sales, season tickets, and price bundling all serve to mask how much a buyer has spent on a given product, decreasing the likelihood that the buyer will actually use i t And a customer who doesn't use a product is unlikely to buy that product again. Executives who employ those pricing tactics without considering their impact on consumption may be trading off long- term customer retention for short-term increases in sales.

The Psychology of Consumption Let's look more closely at why consumption is important and how pricing affects consumption.

Higher consumption means higher sales. One of the first steps in building long-term relationships with cus- tomers, we believe, is to get them to consume products they've already purchased. Research has repeatedly shown that the extent to which custtimers use paid-for products in, say, one year determines whether they will repeat the purchase the next year. One field study, for instance, found that health club members who worked out four times a week were much more likely to renew their memberships than those who worked out just once a week. According to another study, customers who regularly used an enhanced cable television service in one year were more likely to renew their subscriptions in the next year than those who used the service only occasionally.'

Consumption is important to the bottom line in many ways. In businesses that sell subscriptions or member- ships-like Time Warner, the YMCA, or the Metropolitan Opera-customer retention is vital. But keeping custom- ers is tough: Most magazines experience renewal rates of 60% or less, and health clubs retain just 50% of their mem- bers every year. As competitive pressures intensify and the cost of customer acquisition rises, a key to long-term profitability is making sure that customers actually use the products and services they buy.

Consumption also helps establish switching costs. In the software business, for example, companies often make more money selling upgrades than selling the initial ap-

John Gourville is an assoaate professor at Harvard Busi- ness School in Boston. Dilip Soman is an associate professor at the School of Business and Management at Hong Kong University of Science and Technology.

plication. Once customers start using an application, they have to either buy upgrades or make the painful transi- tion to another system. Companies whose software is pur- chased but never used - shelfware, as it's disparagingly called-miss the opportunity to lock customers in for the long term.

Consumption is no less important for businesses that rely on a two-part revenue stream. For movie theaters, sports arenas, and concert hails, ticket sales are just one source of revenue; parking, food and drink, and souvenir sales are a profitable second source. Team Marketing Re- port estimated that the cost for a family of four to attend a major league baseball game in 2000 was $121.36. Only half ofthat sum was used to buy tickets; the other half was spent on beer, soda, hot dogs, programs, baseball caps, and the like. Clearly, if ticket holders don't attend events, these high-margin secondary sales are lost.

Still other organizations believe that their core mis- sion includes encouraging certain kinds of consumption. HMOs, for instance, know that their clients' overall health will improve-and their own costs will be contained-if patients can be persuaded to get regular immunizations and periodic checkups. But they have limited success making this happen. By one estimate, 15% of insured chil- dren do not get all the immunizations they need, 30% of insured at-risk women fail to get mammograms within any two-year window, and 50% of insured men over 50 fail to have physical exams within any three-year period.

Finally, consumption is important to any business that relies on satisfaction to generate repeat sales and positive word-of-mouth. For products as diverse as wine, books, and PDAs, customers won't purchase again or evangelize about products if they don't use them in the first place. In- deed, it's difficult to think of any business in which con- sumption does not make a difference.

Costs drive consumption. People are more likely to consume a product if they're aware of its cost. This is known as the sunk-cost effect: Consumers feel compelled to use products they've paid for to avoid feeling that they've wasted their money. It's well documented that con- sumers routinely consider sunk costs when deciding fu- ture courses of action. In one example made famous by Richard Thaler, a behavioral economist at the University of Chicago, a man joins a tennis club and pays a $300 membership fee for the year. After just two weeks of play- ing, he develops an acute case of tennis elbow. Despite being in pain, the man continues to play, saying: "I don't want to waste the $300."

In a similar vein, Hal Arkes, a psychologist at Ohio Uni- versity, asked 6i college students to assume that, by mis- take, they'd purchased tickets for a $50 and a $ioo ski trip for the same weekend. The students were informed they'd have much more fun on the $50 trip. They were then told they had to choose between the two trips and let the other ticket go to waste. Amazingly, more than half

92 HARVARD BUSINESS REVIEW

Pricing and the Psychology of Consumption

the students reported that they would go on the less enjoyable $itxi trip. For those students, the larger sunk cost mattered more than the greater enjoyment they'd get out of the less expensive trip.

Pricing drives perceptions of cost. Our research also suggests that consumption is driven not so much by the actual cost of a paid-for product as by its perceived cost. This perception is influenced greatly by the manner in which the product is priced. Some pricing policies high- light the perceived cost of a paid- for product while other pricing policies mask the cost.

Consider something as simple as the method of payment. A $io cash transaction feels different than a $ioo cash transaction. Counting out currency and receiving change make a buyer very aware of the magnitude of a transaction. But a $10 credit card transaction is, in important ways, indistinguishable from one for $100-both involve merely signing a slip of paper.

Not surprisingly, people are better able to remember the cost of products if they pay with cash than if they pay with credit cards. In addition, they feel more pressure to consume products if they paid with cash than if they paid with a credit card. In one theater company we studied, the no-show rate for credit card customers was ten times higher than the no-show rate for cash customers. Other pricing tactics that mask rather than highlight prices-like seastmtickets, advance purchases, and subscriptions-also reduce the pressure to consume the product in order to get the money's worth.

Putting the Pieces Together Because pricing has such a powerful effect on consump- tion, managers must make careful decisions about when and bow to charge for goods and services.

Timing. Companies often have great discretion about when to bill for gtK>ds and services. Some companies de- mand payment at or near the time tbe product is to be consumed - this is true when you buy a Big Mac or a movie ticket. Other businesses require payment far in advance of consumption. Concert promoters and sports teams have long operated on this principle. Health clubs and country clubs also charge large, up-front initiation fees. Finally, some businesses allow customers to pay long after a product is purchased. Increasingly, automakers and consumer electronics companies advertise "buy now, pay later" schemes.

Companies make these timing decisions based on either financial considerations (payment sooner Is better than later) or demand considerations ("buy now, pay

The extent to which

customers use the products

they've paid for determines

whether they will repeat

the purchase.

later" increases customer demand). But that may be shortsighted. Payments that occur at or near tbe time of consumption increase attention to a product's cost, boost- ing the likelihood of its consumption. By contrast, pay- ments made either long before or after the actual pur- chase reduce attention to a product's cost and decrease the likelihood that it will be used.

We conducted a survey at the Chicago Science Museum in 1997 to determine how timing affects consumption.

We presented the following hy- pothetical scenario to 80 visitors: "Six months ago, you saw an ad for a theater event and called to reserve a $50 ticket. Yesterday, you went to the box office and paid $50 in cash for your ticket, which is nonrefundable. This morning, you woke up with the flu. The event is tonight. Will you go to the theater or stay home?" Almost 60% of the people reported

that they would go to the theater. They were not willing to let the $50 they had just paid for the ticket go to waste.

We then presented a slightly different scenario to an- other group of 80 visitors. This group was told that they had paid for tbe ticket six months prior to the event, rather than the day before. This time, less than 30% of the people surveyed told us that they would go to the theater. The only difference between tbe two scenarios was the timing of the payment. Yet that difference was sufficient to reduce the predicted consumption by 50%. The results of this as well as several similar surveys show that the im- mediacy of payment can be critical for the consumption of a paid-for product.

In fact, consumption closely tracks the timing of pay- ments by customers. We analyzed data on the payment and attendance records of 2(X) members of a prestigious Colorado-based health club. All the members were con- tractually committed to one-year memberships that cost them $600 eacb. Tbe club's pricing policies allowed mem- bers to choose ffom among four payment plans; annual, semiannual, quarterly, and monthly.

Members who made a single annual payment used the club most frequently in the months immediately follow- ing payment, reflecting a strong sunk-cost effect. But as time passed, the sunk-cost effect dissipated. By the final months, individuals seemed to be treating their member- ships as if they were free and worked out at a rate that was only a quarter of what it had been in the first few months. The same pattern held for members who had paid on a semiannual or quarteriy basis: Aftendance was highest immediately following payment, only to decline steadily until the next payment. This resulted in a sawtooth pat- tern of usage, spiking in the first and seventh months for semiarmual payment members and every three months

SEPTEMBER 2002

Pricing and the Psychology of Consumption

for quarterly members. By contrast, the usage pattern of members who paid on a monthly basis was smoother. Since they were reminded of the cost of their member- ships every month, they used the facility at a steady rate (see the exhibit "Consumption Follows the Timing of Pay- ments). The timing of payments is important because it influences the club's retention of customers. Members who paid on a monthly basis used the gym most consis- tently, making this pricing scheme the most likely to gen- erate membership renewals.

Price Bundling. Organizations often bundle prices to increase the demand for products and services. This prac- tice does increase short-term demand-but it may also re- duce consumption.

Several studies demonstrate this tendency. We con- ducted a survey in a Colorado ski town, for example, pre- senting two slightly different scenarios to two groups of 50 skiers. The first group was told: "It's early spring in Col- orado, and you're on a four-day ski vacation. The day you arrived, you purchased four one-day ski tickets for $40 each. It's now the morning of the last day. You've had three excellent days of skiing, but rain hit the area last night, making a mess of the slopes. A friend suggests that, rather than skiing, you take it easy and leave early to beat the traffic home." We presented the same scenario to the other group with one important difference. We told mem- bers of the second group that, instead of purchasing four one-day tickets for $40 apiece, they had each bought a four-day ski pass for $160.

We asked members of both groups to report their like- lihood of skiing on a scale of one to ten, where one indi-

cated "definitely would not ski" and ten indicated "defi- nitely would ski." The average response of the first group was 7-0, indicating a high likelihood of skiing on the fourth day. The average response of the second group was 3.3, indicating a low likelihood of skiing.

The two scenarios were financially identical, so why the difference? In this study and in several others, we found that price bundling infiuenced consumption con- siderably. Quite simply, it is far easier to identify and account for the cost of an individual product in an un- bundled transaction than within a bundled transaction. The one-to-one relationship between price and benefits in an unbundled transaction makes the cost ofthat item obvious, creating a strong sunk-cost eftect and a high like- lihood of consumption.

Nowhere is the impact of price bundling on consump- tion more obvious than in the case of season tickets. The purchaser pays one bundled sum for a collection of indi- vidual events, making it difficult to allocate costs to any one performance or game. This reduces the likelihood of its usage. We tested this out by analyzing ticket purchase and attendance data at a Shakespearean summer festival. The festival ran from June through August 1997 and in- volved the production of four plays. Some ticket holders had purchased tickets to a single play, some to two or three of the plays, and others to all four plays. What we found was that the no-show rate for people who had bought tickets to a single play was 0.6%, indicating that al- most all ticket holders showed up. But the no-show rate for those purchasing tickets to two plays was 3.5%; for three plays, 13.1%; and for four plays, 21.5%. As the bundling of

Consumption Follows the Timing of Payments

Our analysis of one health club's

records showed that consumption

closely follows the timing of pay-

ments. Whether members made

annual, semiannual, or quarterly

payments, club use was the highest

in the months immediately follow-

ing payment and declined steadily

until the next payment. Members

who paid on a monthly basis used

the gym most consistently, making

this pricing scheme the most likely

to generate membership renewals.

annual payment plan quarterly payment plan

semiannual payment plan monthly payment plan

month of membership

94 HARVARD BUSINESS REVIEW

Pricing and the Psychology of Consumption

tickets increased from one to four plays, the likelihiwd of a person not showing up for one of the pîays rose 35-fold.

One could argue that the higher no-show rate for those who had bought tickets to more than one play was due to other factors: boredom (ticket holders got sick of Shake- speare) or perhaps dissatisfaction (after the first play.

no-shows, a theater manager could oversell performances where the no-show rate is expected to be high.

Stagger payments to smooth consumption. A sec- ond course of action, slightly more proactive but still within current pricing practices, would be to stagger billing cycles so that demand is smoothed over time. This

Pricing tactics that mask rather than highlight prices reduce pressure

on buyers to use the product in order to get their money's worth.

ticket holders realized the quality of the performances was not very good). However, when we looked at only the first play each person had bought tickets for, the pattern remained the same. Compared with the no-show rate of 0.6% for the single-play ticket holders, the no-show rate at the first play for those patrons who bad purchased tick- ets to two, three, and four plays was 2.8%, 7.8%, and 15.8%, respectively. So, the bundling of tickets had much the same effect as the advance selling of tickets in our earlier examples: It hid the cost of each ticket. Unable to link the costs and benefits of any given play, patrons who pur- chased tickets to multiple plays increasingly treated their tickets as if they were free. With little sunk-cost pressure, many of these customers did not use tickets they had pre- viously paid for, reducing their likelihood of repeating their ticket purchases for the follov^ing season.

Linking Price and Consumption We're not suggesting that executives throw out their cur- rent, demand-centered pricing policies and focus exclu- sively on encouraging consumption through pricing deci- sions. It wouldn't be realistic, and it wouldn't be smart. Many companies lack the ability (or the desire) to restruc- ture their pricing practices. In some cases, industry norms or consumer expectations dictate the use of advance sell- ing or price bundling. However, we believe that executives should take consumption into account when they set prices. Here are some suggestions on how to do that.

Practice yield management. Managers can run oper- ations more efficiently by anticipating actual demand given the naturally occurring mix of bundled versus un- bundled purchases or the ratio of advance to current pur- chases. Consider the case of a theater manager. She might forecast a no-show rate of 2tw(j if the proportion of season ticket holders is high but a no-show rate of only 5% if the proportion of season ticket holders is low. Armed with this knowledge, she could better manage costs by staffing according to actual, as opposed to paid, demand. Alterna- tively, she could increase revenues by overselling some events and not others. In much the same way that an air- line oversells a fiight in proportion to the expected rate of

is another form of yield management. Health clubs, for example, know that most of their new members sign up at specific times of the year, most commonly in January. But many still offer discounts to members who pay in full at the start of the calendar year. The net effect is that peak usage occurs in January, February, and Marcb, which reduces customer satisfaction because of the strain it places on the facilities. Health clubs could stagger billing cycles to offset tbat trend. For instance, a health club could offer ten- or 14-month contracts, perhaps on a dis- counted basis, to break the cycle of January renewals. Over time, this change would help smooth demand and increase customer satisfaction.

Time payments to maximize consumption. Some executives can do more than react to demand: They can use their pricing policies to actively encourage consump- tion. Perhaps the most dramatic way to do this is to link payments more closely to benefits. Consider Boston Red Sox season ticket holders, who are asked to pay for tickets five months before the season begins. To promote atten- dance over the course of the seastm. Red Sox executives could spread out that one large payment. They could, say, bill patrons in four installments. Customers might even prefer this approach because smaller installments are fi- nancially more manageable. Similarly, the health club we studied could increase usage and, thus customer reten- tion, by promoting annual memberships with monthly, rather than annual, payments.

Psychologically link payments to benefits. Some companies view price bundling as a necessary ttxil to pro- mote initial sales: If they eliminate price bundling, they could eliminate the sale. However, organizations could psychologically unbundle those offerings to promote con- sumption. One way of doing this would be to highlight the prices of individual items in tbe bundle after the pay- ment has been made. For instance, travel companies could itemize the approximate cost of offerings in vaca- tion packages. Some all-inclusive cruise ships already make guests pay for drinks, meals, and entertainment with beads to highlight the fact that nothing is really free. Restaurants could offer the same dishes both à la carte and as part of a fixed-price dinner, so that the cost of each

SEPTEMBER 2002

Pricing and the Psychology of Consumption

item in the latter becomes clear. In the same way, HMOs could promote preventive care by itemizing the cost of in- dividual services within the bundled fee, making the cost of those services more apparent to the customer. This would increase enrollees' likelihood of consuming the benefits (getting checkups, immunizations, and so on) they've already paid for.

Reduce consumption. Not all organizations want to encourage consumption all the time. Consider the man- ager of a truly scarce resource, such as a private golf course on a beautiful Sunday morning in June. Managing peak demand is the main concern. The current alterna- tives are to limit the number of customers admitted, as when a country club caps membership, or to accept all customers and run tbe risk of dissatisfaction when the fa- cility is at capacity. The first solution limits revenues; the second increases customer frustration. By managing when and how payments are made, executives can maxi- mize the total number of customers who pay for their services and, at the same time, limit peak demand. If the golf course manager charges annual membership fees in January or February, long before the golf season has begun, a member's pain of payment will fade by the time the peak summer months come, reducing the need to get his money's worth. That should allow the club to maxi- mize its membership base without turning away mem-

bers wanting to play during the peak period. By contrast, if the club bills its members just prior to the peak season, say in May or June, it will be inadvertently promoting demand at its busiest time. Similarly, when vacationers at a ski resort buy a week's worth of lift tickets, the resort has the option of providing them with seven daily tickets or a bundîed pass. The former will encourage consump- tion every day ("I'm not going to let today's ticket go to waste!"), while the latter will mask the cost of skiing on any given day and reduce crowds on the slopes.

« • •

Managers spend a lot of time thinking about how to get customers to buy their products and services. But that's Just half the battle. If organizations wish to build long-term relationships with customers, they must make sure their customers actually use their products. A first step is pricing. ^

1. Stefano DellaVigna and Ulrike Malmendier,"Seif-Control in the Market Ev- idence from the Health Club Industry," Working Paper (Harvard University, 2(K)i). Katherine N. Ismon/liffany lîamett White, and Russell S. Winer,"Dy- namic Customer Relationship Management: Incorporating Future Considera- tions into the Service Retention Decision" Journal of Marketing, January 2002.

Reprint R0209G ' To order reprints, see the last page of Executive Stinunaries.

For more on this topic, go to http://explore.hbr.org.

"Have a seat, and Mr. Schwartz wilt be with you as soon as he can."

96 HARVARD BUSINESS REVIEW

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].