In the first week of January, the back doors of American warehouses look more like the front doors did in November. Trucks back in, boxes come off, and somewhere a clerk starts sorting through what used to be Christmas.
US shoppers returned roughly $890 billion in merchandise in 2024, per the National Retail Federation. That is more than Switzerland’s GDP, sent back, often with the tags still on.
In 2025, the number is expected to come down. NRF’s 2025 Retail Returns Landscape report, produced with Happy Returns, puts the projected total at $849.9 billion, with the return rate dropping from 16.9 percent to 15.8 percent of sales. That is the first real decline in years. It is also nowhere near a fix.
The holiday season is where this gets ugly. Retailers expect 17 percent of holiday sales to come back, in line with the last several years. Online returns run higher still, at 19.3 percent. About 9 percent of all returns are fraudulent, by retailers’ own count.
Most coverage frames the problem as a logistics one. Trucks, warehouses, processing software. That part is real. It is also the part that gets the most investment, and it is not where the actual money is leaking.
The leak is in the conversation, not the warehouse
The actual leak is upstream of logistics. It is in the conversation the customer has, or fails to have, before the return label gets printed.
Walk through what happens when a shopper decides to return something. The size is off. The color is wrong. The product arrived two days late, and the occasion had passed. They open the app, find the return page, select a reason from the dropdown, and the box is prepared. Nobody asks why. Nobody offers an exchange in the moment. Nobody catches the buyer who would have kept the product if a single human had explained how the sizing runs.
This is the gap a serious retail call center operation closes. Not by stopping the return, since most returns are legitimate and trying to block them damages the relationship, but by catching the small fraction that were avoidable, surfacing the exchange option early, and capturing the reason data that the product team will need in February to stop the next wave.
That last piece matters more than retailers usually credit. Return reason data, gathered now of contact rather than from a dropdown, is some of the cleanest customer feedback a retailer can collect. It tells you which SKUs are mis-sized, which photos are misleading, and which delivery promises are missing the mark. Most retailers throw that data away. The ones who staff it build a flywheel.
What the retailers handling returns well actually do
Three traits show up consistently. None of them are about software.
They staff for the spike, not the average.
Inbound contact volume in the first ten days of January can run three to four times the November baseline. Retailers who scale their service team for the average month spend the first two weeks of the year apologizing. The ones who scale for the spike resolve, exchange, and retain.
They treat the return as a sales conversation.
A well-trained agent on a return call closes exchanges at meaningfully higher rates than a self-service flow. The exchange protects revenue. It also protects the customer relationship, because the buyer leaves the call feeling heard rather than processed.
They look at the fraud number honestly.
Nine percent of returns being fraudulent is a security problem, but it is also a process problem. Retailers using AI for return fraud detection (85 percent are, per NRF) get more value when the detection model is paired with trained agents who can de-escalate borderline cases without alienating real customers caught in the net.
Ecommerce-first brands feel this most acutely. Online return rates running near 20 percent mean that for every five orders shipped, one comes back. That math destroys margins quickly, which is why ecommerce customer service outsourcing has shifted over the last two years from a cost-saving move to a margin-protection move. The right partner is not the cheapest one. It is the one that treats the post-purchase window as a revenue opportunity and staffs accordingly.
The retailers who get January right do not see it as the end of the holiday season. They see it as the first transaction of the next one. The customer who calls in to return a coat in week one is the customer who will or will not buy from the same brand in October. The handling of that call decides which.
$890 billion is a lot of merchandise. It is also a lot of conversations. The retailers who treat the conversations as seriously as the warehouse will be the ones whose next holiday season looks a little less like Switzerland walking back through the door