Even as loyalty programs are launched left and right, many are being scuttled. How can that be? These days, everyone knows that an old customer retained is worth more than a new customer won. What is so hard about making a simple loyalty program work? Quite a lot, the authors say. The biggest challenges include clarifying business goals, engineering the reward structure, and creating incentives powerful enough to change buying behavior but not so generous that they erode margins. Additionally, companies have to sort out the puzzles of consumer psychology, which can result, for example, in two rewards of equal economic value, inspiring very different levels of purchasing. In their research, the authors have discovered patterns in what the successful loyalty programs get right and in how the others fail. Together, their findings constitute a toolkit for designing something rare indeed: a program that won't do you wrong. To begin with, it's important to know exactly what a loyalty program can do. It can keep customers from defecting, induce them to consolidate certain purchases with one seller (in other words, win a greater share of wallet), prompt customers to make additional purchases, yield insight into their behavior and preferences, and turn a profit. A program can meet these objectives in several ways--for instance, by offering rewards (points, say, or frequent-flier miles) divisible enough to provide many redemption opportunities but not so divisible that they fail to lock in customers. Companies striving to generate customer loyalty should avoid five common mistakes: Don't create a new commodity, which can result in price wars and other tit-for-tat competitive moves; don't cater to the disloyal by making rewards easy for just anyone to reap; don't reward purchasing volume over profitability; don't give away the store; and, finally, don't promise what can't be delivered.

This is an MIT Sloan Management Review article. In theory, trade promotions should benefit everyone involved. In practice, however, manufacturers and retailers often use trade promotions as weapons in a zero-sum game, and consumers are sometimes left out altogether. It need not be that way. Over the past three years, David Bell, an associate professor of marketing at the University of Pennsylvania's Wharton School, and Xavier Drèze, a visiting assistant professor of marketing at UCLA's Anderson School, have examined the theoretical and practical problems associated with trade promotions, and they explain how the right kind of deal can be created -- a transparent system that generates mutual trust and provides benefits to both manufacturers and retailers. The key is proper implementation of what is thus far a little understood tool: the pay-for-performance trade promotion, in which retailers get rewarded according to how much they sell, not how much they buy. The authors explain how the most accepted way of doing promotions today -- which rewards retailers for effective buying rather than effective marketing -- creates a variety of inefficiencies that drain resources from their intended purpose. Using a hypothetical case involving much-simplified mathematics, they go on to demonstrate how manufacturers can design pay-for-performance options that retailers can embrace. They also illustrate how one national beverage company made pay-for-performance deals work in practice. Finally, they offer practical advice to help senior managers rethink the elements of organizational culture that stand in the way of a more effective approach to trade promotions -- and, by extension, block better, more profitable relationships all along the channel.