Return fraud costs US retailers $103 billion in 2024, about 15 percent of all returns. The three patterns that dominate are wardrobing, wearing an item and returning it, empty-box returns, and decoy returns with a counterfeit or substitute item. All three are detectable in the data before the refund goes out.
The three patterns, and what each one looks like
Wardrobing is buying an item, using it once, and returning it as if it were unused. The classic case is a dress bought for a wedding and returned on Monday with the tags reattached. A UK survey found 32 percent of shoppers admit to it, rising to 43 percent among under-34s. The return looks clean, the tags are on, and the only evidence is a faint smell of perfume or a crease the steamer cannot quite remove.
Empty-box returns are the blunt version: the box comes back empty, or filled with rocks, old clothes, or a substitute item instead of the original product. Retailers tracking incidents report empty-box returns in 65 percent of cases they review, and overstated quantities in 71 percent. Decoy returns are the sophisticated version: a counterfeit or substitute item goes back in the box, and the original is kept or resold. The three patterns together are why the fraud rate is 15 percent, not 1 percent.
Related reading on this site: Home and Invalid Clicks and ROAS: When Bot Traffic Eats Your Ad Budget cover the neighbouring problems in more depth.
Why the refund goes out before the fraud is found
The refund goes out fast because the customer expects it fast, and the fraud is found slow because the inspection happens after the refund. A retailer that waits for a full inspection before refunding will lose good customers who expect the money back in days. A retailer that refunds on receipt will pay the fraudsters first and discover the problem later. The tension is real, and it is the reason the fraud rate stays high: the system is built to refund first and ask questions later.
The numbers show the scale. Total US merchandise returns were $685 billion in 2024, 13.21 percent of retail sales. Fraudulent returns and claims were $103 billion of that, 15.14 percent. The fraud is not a rounding error on the returns process. It is a business in its own right, with its own patterns, its own repeat offenders, and its own supply chain of counterfeits and decoys.
The signals that catch it in time
The fix is to score the return, not the customer. Return timing, value, frequency, and weight mismatch are signals, and they are available before the refund goes out. Loop's fraud model found 28 percent of returns over $800 contain fraud signals. A return that arrives two days after delivery, at high value, from an account with three returns in a month, is a return worth inspecting before the refund. A return that arrives three weeks after delivery, at low value, from a customer with no history of returns, is a return worth refunding fast.
The inspection has to be proportionate. A high-risk return gets a weight check, a serial number check, and a photo. A low-risk return gets the fast refund. The retailer that inspects every return will lose good customers to slow refunds. The retailer that inspects none will pay the fraudsters. The score is the compromise, and it has to be built on the data the retailer already has: the order, the return, and the weight.
Frequently asked questions
What is wardrobing?
Buying an item, using it once, and returning it as if it were unused. The classic case is a dress bought for a wedding and returned on Monday with the tags reattached. A UK survey found 32 percent of shoppers admit to it, rising to 43 percent among under-34s.
How big is return fraud?
Appriss Retail and the National Retail Federation put fraudulent returns and claims at $103 billion in 2024, 15.14 percent of all US returns. Total US merchandise returns were $685 billion, 13.21 percent of retail sales.
What is an empty-box return?
A return where the box comes back empty, or filled with rocks, old clothes, or a substitute item instead of the original product. Retailers tracking incidents report empty-box returns in 65 percent of cases they review, and overstated quantities in 71 percent.
How do retailers catch return fraud without punishing good customers?
By scoring the return, not the customer. Return timing, value, frequency, and weight mismatch are signals. Loop's fraud model found 28 percent of returns over $800 contain fraud signals. The fix is to verify high-risk returns and let the rest flow.