Veikko Thiele
Professor & Distinguished Research Fellow of Business Economics
- Adapted from: “Product Returns, Customer Segmentation, and Dynamic Pricing in the Online Retail Market”
- Based on Research by: Otte, J., Serfes, K., and Thiele, V.
- Journal: Journal of Economics and Management Strategy
Key Takeaways
- Product returns signal preferences. A customer who returns a product reveals a low product valuation. Retailers can use this to offer personalized discounts at the next purchase — profitably segmenting the market without surveys or tracking cookies.
- Retailers may rationally overcharge for returns. When shipping costs are low, the optimal restocking fee exceeds the actual cost of a return — generating revenue from price-sensitive returnees while high-value customers keep the product and pay full price.
- Retailers may also subsidize returns. When shipping costs are high, charging below actual cost encourages more customers to return, enabling better customer segmentation and more profitable future pricing.
- Strategic customers change the calculus. If customers anticipate that returning today earns a discount tomorrow, many will return even products they value. The retailer should respond by reducing the initial price and raising the return fee to deter gaming.
- Transparency can backfire. Because return decisions are the mechanism for segmenting customers, retailers have an incentive to conceal the use of personalized pricing — openly advertising the discount program invites strategic returns and erodes profitability.
Online retail in the United States surpassed $1 trillion in 2022, and with it came a staggering volume of returns — roughly 20.8% of all merchandise purchased online is sent back, at a total cost of $218 billion in 2021 alone. Many retailers respond by charging a "restocking fee" deducted from the refund. But why? And when should retailers subsidize returns instead?
This study develops a formal economic model showing that product returns do far more than imposing shipping costs — they reveal valuable information about a customer's willingness to pay. When a shopper returns a product, the individual is signalling that the product's value falls below its price. The retailer can exploit this signal to offer a discounted price on a future purchase, while charging full price to customers who kept the product. This is a form of third-degree price discrimination enabled entirely by return behaviour.
The authors analyze two types of shoppers: myopic customers who make return decisions purely on today's payoff, and strategic customers who may game the system — returning a product they actually value just to secure a lower price later. The model derives optimal pricing and restocking fee strategies for both scenarios.
For managers, the research reframes the cost-benefit analysis of return policies. A generous returns program is not just a customer-service expense — it is a data-collection mechanism. Retailers that track return behaviour and use it to personalize future pricing can recover the cost of returns (and more) through better-targeted offers. The finding that overcharging for returns can be optimal challenges the intuition that return fees should simply cover logistics. A restocking fee above cost extracts value from price-sensitive customers while retaining high-value shoppers — a subtle but powerful pricing lever available to any e-commerce operator.
The work also has implications for platform design: algorithms that automatically surface personalized discounts to past returnees can boost profitability, but only if they operate quietly. Prominently advertising discount programs invites gaming — a lesson relevant to loyalty systems, clearance sales, and subscription winback offers more broadly.