Customer rewards have been reviled in the business press as cheap promotional devices, short-term fads, giving something for nothing. Yet they’ve been around for more than a decade, and more companies, not fewer, are jumping on the bandwagon. From airlines offering frequent flier deals to telecommunications companies lowering their fees to get more volume, organizations are spending millions of dollars developing and implementing rewards programs.

Company interest is justified. The theory is sound. Rewards can and do build customers’ loyalty, and most companies now appreciate how valuable that loyalty can be. As Frederick F. Reichheld and W. Earl Sasser, Jr., documented in “Quality Comes to Services” (HBR September–October 1990), a company’s most loyal customers are also its most profitable. With each additional year of a relationship, customers become less costly to serve. Over time, as the loyalty life cycle plays out, loyal customers even become business builders: buying more, paying premium prices, and bringing in new customers through referrals.

In practice, however, rewards programs are widely misunderstood and often misapplied. When it comes to design and implementation, too many companies treat rewards as short-term promotional giveaways or specials of the month. Approached that way, rewards can create some value by motivating new or existing customers to try a product or service. But until they are designed to build loyalty, they will return at best a small fraction of their potential value.

A rewards program can accelerate the loyalty life cycle, encouraging first-or second-year customers to behave like a company’s most profitable tenth-year customers—but only if it is planned and implemented as part of a larger loyalty-management strategy. A company must find ways to share value with customers in proportion to the value the customers’ loyalty creates for the company. The goal must be to develop a system through which customers are continually educated about the rewards of loyalty and motivated to earn them. Achieving sustainable loyalty, measured in years, requires a strategic sustainable approach.

The Rules of Rewards

Some of the best examples of building customer loyalty through value sharing can be found in traditional small businesses. For many years, successful neighborhood merchants and restaurateurs have understood intuitively the broader strategic purpose behind an effective rewards program. Such businesspeople make it a point to get to know their best customers personally and often reward them with special services and attention—notifying them when sought-after merchandise arrives, for example, or giving them a free drink or a special dessert. They know that delivering increased value to profitable customers turns them into loyal customers; and that loyal customers become even more profitable over time.

But as companies increase in size and complexity, their ability to detect which of their customers are the most valuable falls precipitously. High turnover of sales and customer service employees exacerbates the problem. Personalized customer relationships and the accompanying keen judgment on value sharing disappear.

Large companies striving for increased market share, scale, and efficiency try to compensate for the loss of personal relationships by using database marketing or sophisticated market research techniques to target valuable customers. For those investments to pay off, however, companies must also keep in mind the following principles of effective value sharing.

All customers are not created equal.

Realizing the benefits of loyalty requires an admission that not all customers are equal. In order to maximize loyalty and profitability, a company must give its best value to its best customers. That is, customers who generate superior profits for a company should enjoy the benefits of that value creation. As a result, they will then become even more loyal and profitable.

For example, a company might consider offering better prices to loyal customers. Credit card companies often offer lower interest rates to customers with better credit profiles and superior payment histories. Taking into account long customer tenure and good accident history, State Farm Insurance provides individual discounts on its auto insurance policies. It also discourages business from problem drivers by not offering competitive prices for that customer segment. State Farm’s competitors are thus put in the position of serving a less attractive base of remaining customers.

Unfortunately, most companies inadvertently treat all customers as equal, providing them with products of equivalent value regardless of how much they spend or how long they’ve been customers. A company that offers average-value products and services to everyone wastes resources in over-satisfying less profitable customers while under-satisfying the more valuable loyal customers. The outcome is predictable. Highly profitable customers with higher expectations and more attractive choices defect, and less desirable customers stay around, diluting the company’s profits.

Value created must exceed cost of value delivered.

Most companies launch rewards programs without assessing their own needs and the economics of cause and effect. They haven’t thought through the links between the value delivered to customers and the value created for the company. A rewards program should not give something for nothing: The profits will be illusory, but the costs will be real.

Consider restaurant discount cards such as the one introduced by Transmedia in the United States two years ago. With this card, consumers are entitled to discounts of up to 25% at participating restaurants. Hundreds of restaurants in major metropolitan areas have signed on with Transmedia and similar programs, believing they could bring increased volume or a shift in share sufficient to offset the lower margins that accompany the deal.

A closer assessment of the real proposition for restaurants, however, reveals that the cards have more to do with cash management than with customer loyalty. Transmedia offers restaurants upfront cash in exchange for hefty discounts down the road. In effect, it is lending restaurants money at very high interest rates. Financially troubled restaurants find the offer appealing. Stable establishments, feeling intense competitive pressure, have joined as well.

But in the long run, neither the individual restaurant nor the industry as a whole is likely to benefit. Why? These discount cards represent a transfer of value from the restaurant owner to the consumer (and to Transmedia, which is now expanding globally) but no accompanying creation of value for the restaurant owner. In fact, by encouraging consumers to shop around, the cards discourage loyalty. The flaw is in the design of the incentives. Transmedia offers restaurant-goers the same discount at all restaurants regardless of how often they visit or how much they spend. Such discounts differ from traditional perks, which individual restaurateurs offer only to their regular customers.

Restaurants join discount card programs to attract new customers by shifting market share away from restaurants that don’t participate. But the structure does not ensure either that customers will dine out more often than they used to or that they will focus their patronage on a single restaurant—the two acts required to create real value.

Customer behavior should drive value sharing.

Rewarding—and thus reinforcing—desirable behavior may seem like an obvious tenet of program design, but the marketplace is full of companies that reward the talk instead of the walk. One credit card company, for example, recently launched a campaign that gives new customers 10,000 bonus points redeemable for airline miles and other rewards. That initial bonus puts customers well on their way to earning an airline ticket, so the value of the offer is quite high. Yet there is nothing to preclude customers from signing up, redeeming their points, and then defecting. That behavior does not benefit the company in the long run.

Such promotions are common today in consumer product businesses, and the results should sound a warning. Customers are so inured to offers promising everything from a free vacation in Florida to a free credit card that they either yawn when they see a new one or become experts at getting something for nothing. Long-distance phone companies routinely offer $50 checks or coupons for switching to their service. The strategy has brought to light, perhaps even created, a segment of chronic switchers, who routinely shop for the lowest price. No business should want those customers: The economics of loyalty ensure that no business can make money on them.

In order for a rewards program to be a profit center instead of a cost center, the payout must be inextricably linked to desired behaviors. The American Express Company learned that lesson with its Membership Miles program—a rewards system that allows Amex cardholders to earn points toward a variety of awards by charging purchases. Although Amex management had defined and communicated a companywide loyalty vision of achieving 100% share of customers’ card spending, concern over program costs hindered rapid progress toward that goal. Because Amex originally developed Membership Miles in response to competitive pressures from other card issuers in the United States, many in the organization viewed the program as a purely defensive move. They didn’t fully recognize its importance to the overall loyalty strategy. Until about a year ago, in fact, the organization treated its rewards program as a cost center, not a profit center: it cost Amex hard cash to purchase miles from the airlines, but many of the benefits of rewards were difficult to quantify.

Discontinuity between strategy and implementation is not uncommon. Tracking the benefits of loyalty requires new tools that go beyond traditional financial analysis. At Amex, quantifying the results of Membership Miles was an ongoing goal, but it wasn’t easy to measure the full effect of value sharing on customers’ behavior. Everyone acknowledged that customers who enrolled in the program charged more of their purchases to the card, but it wasn’t until the company began tracking higher retention, incremental upgrades and purchases of Amex products, and acquisition of new customers that the true profitability was clear. Only when Amex understood these relationships did it begin to use the program more efficiently to encourage profitable customer habits such as referrals and to discourage unprofitable habits such as attrition.

Long-term perspective is critical.

Onetime promotions can cost a great deal of money and do not, as a rule, generate loyalty. They do indeed change customer behavior but often in ways that are undesirable in the long run. Any positive impact is washed away as soon as competing companies launch their next promotions.

Many credit card companies have used lotteries, for instance, to increase response rates from potential new customers. A display of brochures urging people to “take one” may feature a chance at a free trip to London or a vacation in a luxury condominium. Such offers understandably generate more excitement than a standard description of a credit card product does. Yet once the lottery ends (with most customers turning out to be losers), people who signed up only to be in the drawing may defect. In the short run, the organization hits its numbers, but, in the long run, it is no better off.

The full potential of value sharing through rewards is realized only when customers change their habits to become sustainably loyal. And that shift occurs only when the company has developed and communicated a proposition that clearly has long-term benefits for the consumer.

Offers must target attractive customers.

Progressive companies understand the superior economics of targeted marketing and the need for a segmented approach to developing products and value propositions. Many invest heavily in market research to generate elaborate demographic or psychographic segmentations. But too often these models prove unsatisfying because there is no practical way to identify a customer by segment. Defining a segment as “big backyards” or “personal-computer junkies” may call to mind a picture of the lifestyle or mind-set of people in those groups, but it does not provide a useful means of finding the customer.

In addition, as anyone familiar with market research knows, what customers say and what they do are two different things. The most sophisticated research methodologies in the world do not always yield accurate predictions of real-life behavior. For example, most people, when asked, will say they don’t like companies calling them on the phone to sell products. Yet, in numerous situations, telemarketing has generated very high response rates.

A well-designed rewards program can target—and attract—valuable customer segments. At the same time, it can save the company money by discouraging those customers who would prove to be less valuable. Such a program is self-selecting and individually correcting.

MCI’s Friends and Family is a good example of a series of self-selecting, individually correcting programs. It offers customers a significant incentive to enroll their friends and relatives with MCI. All calls within the specified network of friends and family members cost 20% to 50% less than other calls. The proposition is clearly most appealing to customers who use long distance heavily, a core segment with which MCI wants to build loyalty. In fact, Friends and Family has helped MCI lure away from AT&T a disproportionate share of profitable customers. In addition, the value to a customer individually corrects to match the customer’s behavior. The more family members and friends a customer recruits for MCI, the more discounted calls that customer earns. Although the cost to MCI is substantial, there is no cost at all until the customer actually exhibits the desired behavior.

MCI shifted marketing dollars from sales-force expense to customer value by turning its customers into highly effective salespeople. It also provided pressure to keep customers from switching to competitors. At a time when most long-distance players were using competing coupon promotions to encourage switching, that was a significant accomplishment. (See the insert “Customer Segmentation at Amex” for another self-selecting, individually correcting rewards program.)

Customer Segmentation at Amex

The Strategy Behind the Program

A rewards program is a competitive strategy and, as such, it must meet certain criteria. Does the program align with company capabilities? Will customers value the program? Can competitors offer a more desirable alternative? Would partnering make the program more competitive? Any rewards program that does not address such criteria is unlikely to succeed.

Does the program align with company capabilities?

Just because your customers need something that you’re not currently providing doesn’t mean you should develop that capability. The need may already be satisfied with products your competitors are offering or planning, and your company may not be capable of meeting the need profitably, or as profitably as competitors can. Rewards programs do not exist in a vacuum; they must dovetail with a company’s overall strategy and capabilities. The right question, therefore, is not What do our customers need? but Which of our customers’ needs can we profitably and differentially meet?

For example, research that American Express conducted on one core segment, affluent business travelers, revealed that rewards offered by competitors were one of the major reasons why customers were defecting. A number of credit card companies had copied the airlines’ frequent flier programs and had begun to offer their customers rewards ranging from airline miles to discounts on new cars. The offers represented a significant enhancement in the basic value proposition and were attractive to high-spending business travelers, who could earn rewards quickly and were already familiar with frequent flier programs.

When American Express decided to invest heavily in a rewards program, however, it considered not only customer demand but also its own capabilities and competitive advantages. Unlike the banks that issue credit cards through Visa and Master-Card and make money by charging interest on revolving balances, American Express must rely more heavily on the fees it charges merchants because its charge card customers pay their balances in full each month. However, since American Express is the leading payment product among affluent consumers and business spenders, it generally charges merchants higher transaction fees than credit cards do. As a result, increases in spending for customers who pay off their balances are more profitable to American Express than to its bank card competitors. In a world where customer rewards are based on dollars spent, American Express can afford to offer more generous value sharing to the cardholder.

Will customers value the program?

Many of the rewards and loyalty programs in the marketplace today reveal a limited understanding of customer needs and desires. From a customer’s perspective, five elements determine a program’s value. They are cash value, choice of redemption options, aspirational value, relevance, and convenience. Few programs today offer all five, but companies that want to play the rewards game should be sure their value measures up to customers’ alternatives.

There is more than one way to measure the first element, cash value. Although the Discover credit card offers a rebate in cash (up to 1% of all spending), most companies’ rewards call for a yardstick other than cash, one that compares the underlying value of different redemption options. A simple rule is to think of the value of a reward (what the customer would have to pay in cash to acquire it) as a percentage rebate on what the customer spent to earn that reward.

Programs that offer airline miles as a reward are tricky because they require some assumptions about what an airline ticket is worth. Since airline seats are so often discounted, customers wouldn’t take full retail value as the cash measurement. If, to earn a domestic round-trip ticket valued at $400, one must spend $25,000 with Citibank AAdvantage, the cash value of the reward is 1.75%. That is considerably higher than Discover’s cash value, but there are other elements at play. With Citibank AAdvantage, redemption choice is limited to tickets on American Airlines.

The success of Citibank AAdvantage and other programs with single redemption options—the General Motors card, for one—demonstrates that although customers value choice, they also respond well to rewards that have aspirational value. Rewards that motivate a customer to change his or her behavior have as much to do with psychology as with economics. A discount on a telephone bill does not have the same aspirational impact as exotic free travel or a hot new car. The key is offering the right choice of aspirational redemption options. American Express is trying to address choice by converting Membership Miles into Membership Rewards and by adding more redemption options, such as hotels, resorts, entertainment, and discounts on automobiles.

AT&T has spent millions to launch and promote its True Rewards program. With True Rewards, customers have a choice of redemption options, including free minutes of calling, a 5% cash rebate, and airline miles. The program offers choice, aspiration, and competitive cash value, but customers can earn rewards only on their long-distance spending. For all but a few heavy-volume long-distance customers, it will take many years to earn an airline ticket. Thus, for most consumers, the program lacks relevance.

The bottom line is that customers don’t want to play in 20 different games or wait 20 years to accumulate airline tickets. Even though frequent fliers have the option to participate in ten or more airline programs, they tend to focus their purchases. If they were to spread their purchases out evenly, it would take them ten times as long to attain rewards. That’s too long to wait.

The final factor in determining customer value is convenience. To understand its importance, one need only observe Air Miles’ different levels of success in different markets. Air Miles, an independent company founded in the United Kingdom, put together networks of wholesalers and retailers offering customers rewards for purchases within the participating network. The cash value of the program (5%on spending at most participating partners) was highly competitive when it was launched in 1990.

In the United States, however, the Air Miles membership card had no transaction capability (it was neither a charge card nor a credit card), and therefore tracking rewards required an entirely new infrastructure at point of sale. Customers wearied of mailing in receipts and redeeming coupons. Merchants resisted additional signage and tedious paperwork. Air Miles in the United States was not only inconvenient for consumers and participating merchants but also costly for the company to administer. As a result, the company folded its U.S. franchise within two years. Air Miles franchises in both the United Kingdom and Canada took the lesson to heart, subsequently partnering with banks for the critical transaction capability.

Rewards linked to a charge card or credit card have a clear convenience advantage because neither the customer nor the merchant needs to exert any incremental effort. Another advantage is that customers can accumulate rewards in a single program based on all their card spending. In fact, one of the reasons rewards programs work so well for card issuers is that they motivate customers to consolidate all their spending onto a single card rather than use two or three cards.

Would partnering make the program more competitive?

To make value sharing work, the value a company delivers to its best customers should be competitive in all five dimensions: cash value, choice, aspiration, relevance, and convenience. Although few companies have all those capabilities in their own arsenals, that shouldn’t prevent a business from attaining access to the full set.

Take the GM card, for example. General Motors, on its own, would not seem a likely candidate to offer a rewards program. Car companies are trying to influence a major but infrequent purchasing decision, whereas the typical loyalty-building program gives customers rewards for frequent spending on one company’s products. But General Motors, accustomed to offering costly purchase incentives in the form of rebates, saw a better opportunity to shift share through a rewards program that locks in customers who might not otherwise purchase a GM product.

GM can’t offer customers a free car for every ten cars they buy. But the company wanted to encourage repeat purchases and attract new customers. So it entered into a partnership with Household Finance. By launching a cobranded credit card that put 5% of spending toward purchase or lease of a new car, GM was able to share value with loyal customers and attract new customers as well. GM estimates that it will make money on the card if one of six cars sold through points redemption would not have been sold otherwise.

General Motors clearly understood that as a single entity, it could provide aspiration and cash value. It also understood that, to be competitive, it needed convenience, relevance, and, ultimately, choice. Household Finance saw an opportunity to acquire new customers, who would be attracted by the 5% rebate on spending toward a new car.

But sharing value through cobranding is not the only way a company can join with other organizations to create value through a rewards program. Moreover, customers do not want a wallet full of cobranded credit cards today any more than they wanted a wallet full of retailer charge cards 20 years ago. Recognizing that, some companies have chosen to become part of a rewards network, such as a credit card program, through which customers can earn one or more of a host of rewards provided by different companies. In a network, each company brings different capabilities to the table—and each may take away a different form of value. For example, a card-issuing partner might be building loyalty through the program in a straightforward way: customers like the rewards offered, so they use the card more frequently. But auto or entertainment partners might use the program to encourage consumers to try out their services. Identifying potentially profitable customers who can be turned into loyal customers is worth a lot. Exactly where companies plug into the loyalty equation matters less than the value that participation creates. (See the exhibit “Rewards Have Evolved Toward Greater Customer Value.”)

Consider the American Express Membership Miles network, through which participating companies with differing capabilities and needs gain access to Amex’s base of affluent members. Across 19 international markets, including the United States, the program has generated a bank of billions of redeemable points. To companies in the network, that bank represents purchasing power. American Express, in a sense, serves as a broker, lining up valuable offers for its cardholders while simultaneously creating value for its rewards partners.

The partners also benefit from the network’s flexibility. For example, auto partners, being in a highly cyclical industry, can’t afford to have too many points redeemed for rebates at a time when capacity utilization is tight. A network such as Membership Miles has the flexibility to steer customers toward the product that will generate the greatest value for a partner company at any given time.

Partner companies in Membership Miles also have access to detailed data on customers’ spending habits that they never had before. That data allows partners to identify potentially valuable customers and differentiate the value proposition through rewards, service, and a range of value-sharing, loyalty-building options.

Undoing Flawed Value Sharing

Designing and implementing an effective rewards program that encompasses every internal and external factor is not an easy process, and some businesses must first undo past damage. In retailing, for example, bankruptcies, mergers, and inroads from large-scale discounters have ended the heyday of large department stores. What has happened to stores such as Macy’s, which for so many years were both profitable and esteemed by their customers? The creation of value for customers has become unlinked from value sharing.

Macy’s begins a season selling high-priced merchandise to its most profitable, loyal customers, who pay full price because they want to ensure a full selection of sizes and styles. At the end of the season, Macy’s marks down merchandise by as much as 80% in a desperate attempt to clear excess inventory before the start of the next season. Some might argue that markdown sales are an effective application of customer segmentation: they cut prices for the price-sensitive. But the customers who are extracting the most value from the system aren’t the loyal, highly profitable shoppers who frequent Macy’s year-round. They are bargain hunters who feel little or no commitment to Macy’s or to any other store.

The first step toward realigning the customer value proposition and turning around the profitability of chains such as Macy’s is better information. Discounting has become a fact of life in retail clothing partly because the stores do not know enough about their customers to predict demand accurately. If they knew their most profitable customers and those with the potential to become highly profitable, they could maintain a database on their style preferences and sizes. They could reduce the costs of excess inventory, while offering superior value to their loyal customers in the form of selection, recognition and service, and perhaps price.

Retailers have begun to recognize the importance of improved information in achieving the profits of loyalty. Programs such as Neiman Marcus In-Circle and Saks Fifth Avenue SaksFirst offer rebates and rewards as incentives to use proprietary-label credit cards. Although department stores have had their own charge and credit cards for years, most traditionally focused on generating more profit through finance charges and have not used the information created to share value with their best customers.

These retail rewards programs are a step in the direction of value sharing, but, to be competitive, they must overcome gaps in customer value and restructure their incentives to create new customer habits. SaksFirst, for example, links the rewards customers can earn on their spending to their level of spending. A customer who spends more than $2,000 per year receives a certificate worth 2%, or $40, toward future services. At $5,000 per year, the reward increases to 4%. The tiered system motivates customers to spend more (and, in all probability, generates additional profits through interest on revolving balances).

The problem is that the level of discounting is still very modest compared with alternatives such as the GM card’s 5% rebate on all purchases. And, when you consider that end-of-season shoppers are offered discounts of 50% or more, it’s clear that these loyalty programs don’t go far enough.

For the most profitable customers, retailers should be willing to provide even stronger incentives to encourage specific customer behaviors such as referring other customers, buying items not ordinarily purchased (or private-label items with higher margins), or paying full price rather than waiting for end-of-season markdowns.

Anyone who has attempted to get an organization to focus on customer loyalty and to design and implement an effective rewards program is already aware of the barriers. Each functional department views loyalty differently and is likely to promote its own set of solutions. And since customer loyalty is still a new concept to many companies, many managers are prone to use familiar but inappropriate measurement systems and incentives in their new programs. In addition, growth is a top priority at many companies and can lead to an emphasis on acquiring new customers at the expense of retaining current ones or building loyalty.

We’ve laid out the steps required to break this cycle. But for a company to commit to the steps that build loyalty, senior managers must agree that loyalty pays. Then they must be unrelenting about focusing the organization and all their marketing programs on the goals and measures that will develop a loyal customer base.