Retailers Lose $29M to Returns Fraud

New research from ReBound Returns revealed that 1 million returned orders on behalf of retail clients between July 2025 and May 2026 resulted in $29 million of potentially fraudulent returns.

“The returns process has become a blind spot for retailers. Investment in fraud prevention is focused almost entirely on the point of purchase. Once a return is initiated, most retailers are operating without the visibility they need to catch what is actually coming back to them. The result is that many are absorbing losses they cannot identify, let alone measure or tackle the root cause,” says Wouter ten Heggeler, product manager at ReBound Returns. “Returns fraud does not need to be widespread to cause serious financial harm; the problem scales directly with sales. A fraud rate that looks manageable on paper actually represents millions in losses, and without the right systems in place, most retailers have no way of knowing how exposed they actually are. Current returns systems are not built for the increasing scale and sophistication of modern returns fraud. Most rely on static rules and manual review, with limited ability to predict behavior or flag risk before a refund is issued. However, with a more effective model - combining behavioral analysis and risk scoring at the point of return initiation, physical verification at hub level, and connected data across every channel - it is possible to catch fraudulent activity before it becomes a financial loss, without adding friction for legitimate customers.”

Key takeaways:

·      Return rates are now approaching 20% of all online sales, with a total market value nearing $850 billion in the United States alone. As returns rates increase, instances of returns fraud and abuse are growing too, with the global Merchant Risk Council deeming 'refund and returns policy abuse' to be the most prevalent fraud type facing merchants today. This is despite retailers investing heavily in fraud prevention at the point of purchase, with widespread adoption of identity verification, authentication and risk-scoring tools.

·      Returns are frequently refunded before physical inspection has taken place, and data sits in silos across e-commerce platforms, stores, and third-party marketplaces, leaving operational teams without a unified view of returns.

·      A mid-market fashion retailer with $100 million in annual sales, a 20% return rate, and a 5% fraud rate would face a projected loss of $1 million each year to fraudulent returns. For a large omnichannel retailer with the same fraud rate, the loss would rise to $3.5 million annually. At enterprise scale, with $960 million in sales and a 7% fraud rate, the projected loss reaches $20 million a year.

·      The longer a customer holds a product before initiating a return, the higher the likelihood of returns fraud or abuse. The median lead time for a normal return is 9.5 days, but for returns flagged as potentially fraudulent, this almost doubles to 18 days.

·      Poland recorded the highest rate at 6.6%, followed by Denmark at 5.3%. Both were considerably higher than the overall average returns fraud rate of 3.9%.