Return rates are making repeat customers overvalued

Most retailers assume that returning customers become more profitable over time.

New research by El Kihal et al. (2025) suggests that this can be dangerously incomplete.

In e-commerce, repeat customers are usually treated as a valuable asset. They are cheaper to retain than new customers are to acquire, and they are often expected to buy more confidently as they get to know the brand. That logic makes sense. A customer who has already bought several times should, in theory, make fewer mistakes and fewer returns.

But that is not always what happens.

The study followed the full purchase and return history of customers at an online fashion retailer over six years. The researchers wanted to understand whether return behavior improves over time or gets worse as the customer relationship develops.

The answer was clear. Return rates increased.

The average customer returned 25 percent of items on their first purchase. By the tenth purchase, that number had risen to 37 percent. That is a 48 percent increase.

At first glance, this seems counterintuitive. Shouldn’t repeat customers learn what fits, what they like, and what to expect from the brand?

The study found that they do. Customers gain what the researchers call brand experience. The more items people have bought in the past, the more familiar they become with the brand’s quality and fit, and that tends to reduce returns.

But another force turns out to be stronger.

Customers also learn how easy it is to return. Over time, they become more comfortable with the retailer’s return process. And once that pattern forms, returns can become habitual. Customers who returned a larger share of items in the past were more likely to return more items in the future.

That second effect outweighed the first one.

In other words, familiarity with the brand reduces returns, but familiarity with returning increases them even more. The result is that return rates rise as the customer relationship continues.

This matters because many retailers estimate customer value as if repeat customers become steadily more profitable. The study shows that this can be a costly mistake.

For the retailer in the study, ignoring this return-rate increase would have led managers to overestimate cumulative customer value by about 40 percent after ten purchases. That is not a small forecasting error. It changes how much a customer is actually worth.

The broader lesson is simple. Repeat customers are not automatically better customers. In some businesses, especially those with easy returns and tight margins, the opposite can happen. Customers may become easier to sell to, but also more expensive to serve.

Returns are not just a logistics problem. They are part of customer economics.

The paper is especially useful because it avoids a simplistic conclusion. It does not argue that repeat customers are bad. It shows that their value depends on the balance between two learning processes. One helps the retailer. The other helps the customer to exploit flexibility. The study finds evidence for both forces, but the return-habit effect is stronger, which is why average return rates rise over time rather than fall.

That leads to a more realistic view of loyalty. Loyalty is not always a sign of lower cost-to-serve. In some settings, it can mean the customer has become highly skilled at using the system in their own favor. That is why this paper matters beyond returns. It pushes against the comfortable assumption that a longer customer relationship automatically becomes more profitable. In this case, the relationship becomes more efficient for the customer too.

For marketers, the practical implication is that return behavior should not be treated as an operational side issue. It should be built into customer strategy. A customer who buys often but returns heavily should not be valued in the same way as a customer who buys often and keeps most of what they order.

This affects how lifetime value should be estimated. It also changes how success should be measured. Repeat purchases alone are too crude. What matters is repeat purchases net of return behavior. 

The study also points to a more subtle strategic problem. Retailers often work hard to remove friction from the return process because easy returns improve trust and conversion. That can be the right decision. But it also teaches customers how to use returning as part of their shopping routine. In some categories, especially fashion, a lenient return policy may help growth in the short term while making customer economics worse over time.

That does not mean retailers should make returns difficult. It means they need to understand the trade-off clearly. An easier return process may increase acquisition and conversion while also encouraging habits that reduce long-term profitability.

The most useful response is not to punish returns indiscriminately. It is to separate productive loyalty from costly loyalty. Some customers are becoming more efficient buyers. Others are becoming more efficient returners. Those are not the same type of relationship, even if both look active in the purchase data.

The broader lesson is that customer relationships do not always improve in the way managers expect. Over time, customers do learn. But what they learn may not always benefit the firm.

What, then, does marketing research suggest retailers should actually do to reduce return rates?

The most consistent answer is to reduce uncertainty before purchase. Returns often happen because customers are unsure about fit, quality, or whether the product will match what they imagined. Research on online retail returns shows that tools that reduce fit uncertainty, such as accurate product description (Leeuw et al., 2016; Hjort et al. 2019) digital fitting (Gustafsson et al., 2021) and recommendation systems (Sun & Chen, 2025) can lower return-related costs by helping customers make better choices before they buy . 

A second lesson is that better information matters, but only if it is honest (Minnema et al., 2016). Reviews can reduce mismatch by helping customers judge fit and product quality more accurately. But there is an important catch: overly positive reviews can increase returns because they inflate expectations that the product then fails to meet. So the goal is not simply more persuasive product pages, but more calibrated.

Pricing strategy matters too. Recent research suggests that discounts at the moment of purchase can reduce later returns, while post-purchase price drops inside the return window can increase them by giving customers another reason to send items back (Gijsenberg et al., 2025). In other words, sloppy discounting can quietly train both bargain-seeking and return-seeking behavior at the same time.

There is also a segmentation lesson hidden in all this. Not every return should be treated the same way. The paper on return-rate evolution suggests that past return behavior is one of the strongest predictors of future returns. That means marketers should identify customers whose behavior signals growing return habits and treat them differently from customers who simply needed time to learn the brand. The right response may be more sizing help, more product detail, or more careful assortment guidance, rather than a blanket policy change for everyone.

The deeper implication is that the best return-reduction strategy is usually not harsher policy. It is better expectation management. When retailers help customers choose better, describe products more accurately, and avoid creating artificial enthusiasm that the product cannot live up to, they reduce returns without damaging trust. That is a much stronger long-term strategy than simply making the return process more painful.

References:

de Leeuw, S., Minguela-Rata, B., Sabet, E., Boter, J., & Sigurðardóttir, R. (2016). Trade-offs in managing commercial consumer returns for online apparel retail. International Journal of Operations & Production Management, 36(6), 710-731.

El Kihal, S., Erdem, T., Schulze, C., & Zhang, W. (2025). Customer return rate evolution. International Journal of Research in Marketing.

Gijsenberg, M. J., Bijmolt, T. H., & Hirche, C. F. (2025). Promoting product returns? The impact of at-purchase and post-purchase discounts on customers’ return behavior. Journal of Retailing.

Gustafsson, E., Jonsson, P., & Holmström, J. (2021). Reducing retail supply chain costs of product returns using digital product fitting. International Journal of Physical Distribution & Logistics Management, 51(8), 877-896.

Hjort, K., Hellström, D., Karlsson, S., & Oghazi, P. (2019). Typology of practices for managing consumer returns in internet retailing. International Journal of Physical Distribution & Logistics Management, 49(7), 767-790.

Minnema, A., Bijmolt, T. H., Gensler, S., & Wiesel, T. (2016). To keep or not to keep: Effects of online customer reviews on product returns. Journal of retailing, 92(3), 253-267.

Sun, M., & Chen, J. (2025). Managing returns by selling and pricing strategies with online product reviews. European Journal of Operational Research.