Loyalty programs, which give consumers rewards for repeat purchases from particular merchants, are almost as common among retailers as cash registers. McKinsey research found that about half of the ten largest US retailers in each of seven sectors have launched such programs, and the rate is similar among top UK retailers (Exhibit 1). Moreover, loyalty programs are popular with customers: in the United States, 53 percent of grocery customers are enrolled in them, to say nothing of 21 percent of the customers of casual-apparel retailers. Of those who join grocery programs, McKinsey research indicates that 48 percent spend more than they would otherwise, though the figure is only 18 percent in casual-apparel programs. Yet even 18 percent represents a sizable number.

If anything, we expect interest in loyalty programs to intensify. Faced with slowing revenue growth in many categories and the emergence of competing Internet start-ups, retailers are eager to deepen their relationships with existing customers and to increase their share of wallet. Retailers that have stand-alone loyalty programs—in other words, programs without corporate partners—must work out ways of dealing with alliance programs, which bring groups of retailers together to offer rich rewards as well as rebates beyond the reach of most individual retailers.

But retailers are reluctant to face some serious questions. Is heavy investment in loyalty programs likely to make a difference in the behavior of customers? Do they really value these programs? Are any retailers getting their money’s worth? The answers are not straightforward. Many loyalty programs are successful; they enhance the value propositions of retailers, capture valuable data, draw in higher-profit customers, and persuade those customers to spend more. But as many or more programs unwittingly deploy value-destroying mass-discount vehicles, and the problem is likely to worsen as retailers respond to sluggish growth. Every such retailer would do well to consider some hard facts about the real profitability of most programs, the economic traps behind the numbers, and what it takes to ensure success.

The trouble with loyalty

Many loyalty programs struggle with three stubborn facts.

First, the programs are expensive. Our research revealed that 16 major European retailers had a total of some $1.2 billion tied up in annual discounts to customers, with several supermarket chains devoting some $150 million. Costs are about the same in the United States. Given large sales volumes, even programs with modest rebates (up to 1 percent) can cost a great deal of money. Then come the costs of marketing and managing a program—investment in systems, fulfillment support, and so forth—which usually run well into the millions.

Second, loyalty programs take on a life of their own once they start, so mistakes can be very difficult to correct. Even programs with low benefits become entrenched in the minds of customers, who must be informed when they change or come to an end. Customers react negatively to any perceived "take away" once a program is in place, even if they are not actively involved in it. Curiously, the successful launch of a program worsens the problems of ending it. And a negative experience with a program heightens the skepticism of consumers when they are offered a follow-up program and can undermine their trust in a retailer more broadly.

Third, these programs, despite their number and apparent popularity with customers, often fail to increase customers’ loyalty. In fact, 79 percent of customers in casual apparel and 70 percent in grocery say they are always seeking alternatives to their current retailers—percentages that far exceed the percentage of customers actively seeking alternatives in other categories (Exhibit 2). Nor do consumers who join a loyalty program necessarily increase their spending (Exhibit 3).

Understanding the economic traps

Although retailers spend large amounts of money on loyalty programs, many of these programs don’t appear to be very successful. Ineffective programs often fall into one or more of four traps.

Free riders

As many as half of all members of loyalty programs are free riders, enjoying benefits without spending more at the store that provides them. Since these free riders get more and give nothing in return, the incremental sales from program members who do spend more must cover the costs of the program not only for themselves but also for the free riders.

The implications for the numbers needed to break even are sobering. Take a retailer providing a 2 percent rebate to program members. At variable margins of 30 percent, all of its program members would have to spend about 6 percent more if it wanted to break even on the cost of rewards alone. But if only half of the members spent more money, that average spending increase would have to double, to 12 percent. And in reality, a 12 percent sales increase is rare. Consumers bombarded by discounts and other promotions have access to an increasingly wide range of competitive alternatives in nearly every shopping category, especially with the growth of the Internet.

McKinsey research found that the earliest US and European retailers to offer loyalty programs enjoyed average sales increases of 1 to 3 percent in groceries and 5 to 8 percent in department stores during the first year. Later entrants eroded a good part of even these modest gains.

Slim margins versus attractive rewards

Modest per-customer sales (on average, less than $500 a year per household) and slim margins limit the ability of retailers to offer attractive benefits in a cost-effective way. A retailer’s 2 percent rebate on $500 worth of annual household sales, for example, translates into a rebate check of $10 a year—not much in absolute terms and trivial compared with the 25 to 40 percent markdowns now common in a price-driven retail environment (Exhibit 4). Margins are too slim to afford greater rebates.

In addition, the rewards bar is increasingly set by industries whose players can provide higher-value rewards at a far lower cost. Many of these industries have predominantly fixed cost structures and excess capacity: airlines and entertainment venues with unfilled seats, hotels with empty rooms, rental businesses (such as cars) with excess inventory, even telecommunications providers with unused capacity. All these can release inventory for rewards at very low cost.

Unlike such industries, whose variable costs are a fraction of the customer value created, retailers must purchase a reward from manufacturers at some 60 to 70 percent of its retail value. The alternative—using unsold inventory—is more attractive economically but makes for uninspiring rewards.

Failing to track expenses properly

Many retailers seriously underestimate the full cost of setting up and sustaining loyalty programs, so even those that increase sales might actually be draining money. For a large multisite retailer, the launch and maintenance investment (store training, marketing, fulfillment support, and information technology and systems costs) can easily reach $30 million in the first year. Thereafter, annual maintenance costs can reach $5 million to $10 million when marketing, program support, offer fulfillment, customer service, and IT infrastructure costs are figured in. Very few retailers fully account for these incremental costs—especially for the marketing support required to sustain awareness of the programs as well as their momentum and impact.

Competing on unequal terms with the virtual world

Dot-coms with venture capital and an intense need to create buyer awareness commonly offer cash rebates of 5 percent and more—a level beyond the reach of most retailers. These rewards serve as advertising that generates expenses for e-tailers only when a customer actually makes a purchase. This is an effective way to link burgeoning Internet marketing budgets more tightly to profit-generating customer behavior.

In addition, e-tailers are creating on-line reward alliances to encourage customers to consolidate their spending and rebate points across several sectors, such as books, compact discs, electronics, and toys. Cybergold counts Chef’s Catalog, Hammacher Schlemmer, and OshKosh B’Gosh among its e-tailers. MyPoints.com, which has 7.1 million members as well as partnerships with more than 70 retailers offering on-line transactions, is expanding its reach both on- and off-line; consumers and retailers can get points and sales, respectively, through more channels than ever. MyPoints now offers a branded credit card that can be used to earn points for on- and off-line purchases. As a result, customers can aggregate their spending at different retailers in a single program and get back an amount that stand-alone "landed" (physical world) retailers can’t match. The promotional costs are spread among all the participating e-tailers.

Beyond one size fits all

It shouldn’t be surprising, then, that few bricks-and-mortar retailers succeed with anything like a mass-rebate proposition. Changes in consumer behavior resulting from rebates are too small to break even on the investment, and retailers don’t have the margins or the share of spending that would allow them to be more generous.

What can be done? We have found that companies with successful programs solve the problem of low margins by doing three things. If they have a value proposition on which loyalty programs can piggy-back, they exploit it. If their value proposition doesn’t offer a distinctive foundation for a loyalty program, they find mechanisms to provide genuinely attractive rewards without going to great expense. And if their customer economics make this strategy impossible—if such mechanisms simply do not exist—they emphasize the data-gathering value of a loyalty program by designing one that provides small but instant rewards and by making participation extraordinarily easy. The resulting data can be used to tailor value propositions to the most important customer segments.

Use rewards to strengthen the value proposition ...

Some retailers have found simple, elegant ways to sidestep the traps and use rewards to enhance their value proposition.

For example, Target’s Guest Card, a private-label credit card, lets customers donate 1 percent of their card purchases to a local school of their choice. Target describes the program as "school fund-raising made simple" and as "a cornerstone of our commitment to the communities we serve." Rewards that would be insignificant at a personal level are aggregated to create significant contributions from all Target customers. Since the program began in 1995, Target has opened more than 11 million new accounts, and schools have benefited to the sum of $23 million. Thus, at relatively low cost, the program gives all customers the feeling that they play a part in Target’s community service activities while simultaneously reinforcing Target’s core value proposition of community service. This charity program would have much less power if Target hadn’t spent upward of 30 years cultivating its support of local communities.

Neiman Marcus’s InCircle Rewards loyalty program is based on two insights: first, a small percentage of customers generates a majority of the company’s sales and, second, the brand connotes exclusive merchandise and excellent service. The InCircle program thus uses exclusivity and attention-grabbing rewards to reinforce the company’s value proposition among high-spending segments and to create a "halo" effect among the broader customer base. Membership is limited to customers who spend at least $3,000 annually. Basic rewards like in-store parties are available for all InCircle members, but exclusives, such as a month on a remote island or a customized painting by a famous artist, are available only to the biggest spenders.

... to change customer behavior ...

In the examples above, retail companies are using not only the reward but also their value proposition to change the behavior of customers. Not all companies are blessed with a value proposition distinctive enough to accomplish that purpose. Those that are not must provide rewards that are sufficiently attractive in themselves.

Companies trying to do so on their own face quite a challenge: as we have noted, the industry’s economic structure typically makes this road ruinously expensive. But successful models have offered loyal customers small but attractive rewards, such as movie tickets that can be earned in just months, as well as more valuable awards that might take a year or two to earn.

Even so, there is a limit to how far companies can go without overspending. A better solution was found by Air Miles, a Canadian alliance that was started in 1994 and now counts nearly half of Canadian households as members. It stresses the benefits of combining expenditures at more than 100 sponsors, such as Bank of Montreal, The Bay, and Shell Canada, to earn rewards quickly. When members switch their spending to retailers in the program, they build their point totals. The Air Miles World Wide Web site states, "From long-distance calling to travel, car rentals to movie passes, your reward miles can mean free rewards sooner than you think." Air Miles surmounts the cost problem by aggregating retailers, so that customers spend thousands rather than hundreds of dollars a year within the program, allowing those customers to earn rewards that would be economically out of reach for a single-retailer program.

Tesco, the UK grocery retailer, solved the cost problem differently by deploying a two-tier Clubcard program. The first tier is a straightforward effort to gather data, though it does give points (one for every pound sterling customers spend) that can be redeemed for vouchers once 150 points have been accumulated. The second tier, targeted at frequent spenders, is more innovative. Customers earn a "key" when they spend $38 in a single transaction. Fifty keys make the customer a "keyholder," 100 keys a "premium keyholder." Besides earning vouchers, keyholders get discounts on top leisure attractions, theater tickets, sporting events, hotel stays, and other activities. The key program seeks to change the behavior of heavy spenders by encouraging them to spend their money more often at Tesco. In the four years since starting Clubcard, Tesco has raised its market share from 13 percent to more than 17 percent, and about 75 percent of its sales now come through the program. The successful design of the program is evident, as customers spend on average $38.70 per trip. To make it past the threshold, many of them are clearly spending more than they otherwise would.

... and to learn more about your customers

Even very small rewards, if they are immediate, can get customers to participate in a loyalty program if it is extremely easy to do so. The advantage to the retailer is the generation of an enormous amount of data that can be mined for insights into customers’ spending and behavior patterns.

Traditionally, retailers have drawn most of their customer data from mass-market sources: basket analysis, customer surveys, demographic profiles, and so on. But these sources don’t give retailers insight into the behavior of individual customers over time—the critical marker of a retailer’s ability to attract, develop, and retain its customers. Loyalty programs can provide this data.

Once retailers get a sharper image of their customers, they can adjust their operations—merchandising, selection, advertising, and promotions—to appeal more effectively to the most desirable segments. Tesco, for instance, uses Clubcard data to tailor 80,000 variations of a letter offer (based on each member’s profile) and the magazine it sends to Clubcard members.

A tighter focus should in turn enhance loyalty and generate more relevant data in a virtuous circle of action and response. To create reliable insights, however, the loyalty program must capture the customers who generate at least 50 to 60 percent of total sales. If the data are to be used in a meaningful way, it is essential to make the program as easy to use as possible.

Sign up today!

It is hard to overstate the importance of making loyalty programs easy to join and use. To work, they must overcome the inertia that often limits enrollment and participation to avid customers—that is, those who are already loyal. An effective program must reach out to passive and uninterested customers and then build and sustain their interest and involvement. Like ties to distant relatives, customer relationships dwindle rapidly in the absence of regular contact.

But even the simplest programs involve some effort by customers—at least completing paperwork for the membership card or key. To carry customers over the bar, a program must therefore be well communicated to a properly trained workforce so that employees are motivated and able to sign up anyone who shops at or gets in touch with the store. Given the rapid turnover of frontline retail employees, simplicity is crucial.

How persistent should you be? Good loyalty programs are increasingly featured in retailers’ broadcast advertisements and other mainstay communications channels. Tesco, for example, publicizes the Clubcard program on the home page of its Web site. Target’s home page highlights the total donations made to schools through the Guest Card as well as an on-line application for the card. Shell Canada gas stations prominently display large Air Miles signs. Communicating too much is not a risk; failing to make your loyalty program central to your customer marketing and communications initiatives is.