The big idea: Coupons customized to consumer preferences should be less about providing discounts and more about marketing to your best customers.

The scenario: The current economic environment and the growth of digital and mobile technology have led to a resurgence of coupons. Retailers, coupon delivery service providers, and manufacturers of consumer packaged goods are competing to deliver better and relevant coupons to consumers. Trends in customer relationship management technology have also allowed retailers to track consumer purchases through shopper cards and provide discounts that are customized to their consumers’ preferences. Several retailers, including CVS, Kroger and Safeway, have implemented such programs with the hope of improving customer retention and lifetime value. The profitability of these programs is unclear. Without a model of how they should work, retailers will find it hard to monitor and manage these campaigns successfully.

The resolution: Retailers would need to design careful experimental studies to understand how consumers react to customized coupons and to estimate the returns from their customized coupon campaigns. We analyzed more than 500,000 customized coupons that were sent by a consortium of eight non-competing retailers over 16 months. More than 300 national and regional brands participated in these campaigns.

The study found that a substantial portion of consumers who are exposed to the marketing messages that accompany coupons are more likely to buy the product featured, even if they do not redeem the coupon. The “exposure” results from the advertising messages accompanying coupons and is most powerful when coupons are: (a) tailored for individual tastes; (b) distributed selectively rather than mass-marketed, (c) presented as unexpected rewards for customer loyalty, “delighting” consumers, and (d) offer significant savings. Although mere exposure to customized coupon campaigns can improve sales (regardless of redemption), we caution against eliminating discounts and merely mailing advertisements. Our results show that having discounts featured in the campaigns gives consumers a reason to pay attention and improves the exposure effect. These campaigns delivered net returns in excess of 15 percent, which is in the upper end of most customer-centric retailer initiatives. The customized coupon campaigns therefore represent a promising application of customer relationship management data for retailers.

This exposure effect of customized coupons questions the current business model of online, social-media coupon distributors. It urges them to reconsider how they should measure the success of coupon campaigns. The current model for estimating returns from coupons relies on redemption rate as the primary success metric. We propose that coupon sponsors as well as distributors should also be interested in all of the consumers who received the coupon e-mail and did not redeem the coupon, but visited the store.

The lesson: Industry is focused on coupon redemption rates and sales lift among coupon redeemers to measure the effectiveness of coupons. But they are looking at the wrong place, and redeemers are a small portion of the consumers affected by coupons. Brand managers and retailers need to be looking at all of the consumers who received a coupon. Exposure effect is due to effective use of coupons as a marketing communication rather than merely as a discount. Recognizing exposure effect can change the design and implementation of coupons.

Rajkumar Venkatesan and Paul Farris

Venkatesan and Farris are professors of business administration at the University of Virginia’s Darden School of Business.