A Japanese retail experiment involving nearly 130,000 customers found that redeeming a targeted coupon increased estimated customer lifetime value for the promoted product, while the gains were smaller or absent at retailer level.
The research by Rin Tsujikawa, Ryoya Nakano and Takahiro Hoshino examines what happened beyond the immediate discounted purchase. It illustrates why a promotion can work for a manufacturer without producing the same return for the store carrying its products.
Published in 2026 in the Journal of Retailing and Consumer Services, the study used a randomized field experiment in Japan and statistical methods designed to estimate the effects of redeeming a coupon.
Following customers beyond the discounted sale
Customer lifetime value is an estimate of the economic contribution a customer makes over a relationship with a business. It can include expected purchases and the costs of acquiring, serving and retaining the customer. The calculation depends on the information and assumptions used.
The researchers estimated that coupon redemption reduced the probability of customer churn by approximately 10 percentage points and increased the promoted product’s estimated lifetime value by more than 30% in their analysis. Churn refers here to a customer stopping purchases under the study’s definition.
The estimated gains were larger for customers with lower pre-promotion lifetime value, though the authors also found positive estimated effects for some higher-value customers. These are results from a specific promotion and market, not a promise that every coupon will produce a comparable increase.
Random assignment helps researchers compare groups, but redemption itself is a customer choice. Someone who uses a coupon may already be more likely to buy. The authors used an instrumental-variable approach, a statistical method that uses the assigned offer to help estimate the effect of redemption, to address that selection problem.
A brand and a retailer can see different results
A manufacturer wants customers to buy its product. A retailer also needs to know whether the promotion increased spending at the store as a whole. A customer who switches from one brand to another may improve the promoted brand’s results without adding an equivalent amount to the retailer’s sales.
The study found that manufacturer-level lifetime value was more likely to rise, while estimated retailer-level gains were smaller or not evident. The summary does not establish that all retailers lose money on coupons. The outcome depends on the product, campaign costs and what customers would have purchased otherwise.
A business can examine that question by tracking the promoted item, other items in the category and the rest of the basket. It should also account for who funded the discount, platform fees and any sales shifted from a later week.
Our earlier coverage of discounts for infrequent buyers examined another part of promotion design: how customers who do not receive an offer react. Both questions are missed when a campaign is judged only by the number of coupons redeemed.
Test incremental value, not just activity
A coupon can reward a purchase that would have happened anyway. It can also bring an existing purchase forward, or attract a customer who buys once and leaves. A retailer and manufacturer may therefore disagree about whether the same campaign paid for itself.
Comparing a randomly selected group offered the promotion with a similar group not offered it can help estimate additional sales. Follow-up purchases, product margins and the cost of the discount matter alongside the first transaction. Businesses should also check whether data collection and targeting comply with the rules in their markets.
The Japanese findings give both partners a reason to define their own outcome before splitting a promotion budget. More sales of the featured product and more profit for the shop are separate questions.
