The story is repeating across U.S. retail. Sub-scale DTC brands closing. Established DTC names cutting staff. Category leaders adjusting return policies to add fees, restocking charges, or shorter windows.
Best Buy, Macy’s, Kohl’s, JCPenney, J. Crew, and Urban Outfitters have all added return fees in the last two seasons. NRF says 72% of merchants now charge a return fee, up from 66% a year ago. Amazon has ended its Try Before You Buy program, tightened return windows on electronics and fashion, and started banning serial returners outright. REI and Target are cracking down on serial returners too. And the reason cited across the category is the same: returns.
The industry consensus, or what looks like consensus, is that “free returns” broke DTC. That the policy the category adopted to compete with Amazon a decade ago became the cost that ate its unit economics.
Did Free Returns Really Break DTC?
I don’t think that’s what happened. And if that’s the diagnosis U.S. retail is heading into peak season 2026 with, the corrective actions will be wrong.
The category was broken by static returns, not free returns.
Here is what actually happened. Over the last decade, U.S. DTC brands built a category on customer-friendly policies. Free shipping. Free returns. Ninety-day windows. The math worked when acquisition costs were low, when brands were sub-scale enough to eat the return cost, and when returns were 8% of retail.
It stopped working when acquisition costs tripled, when brands became large enough that returns became a material line item, and when returns climbed to 15% to 20% of retail. NRF’s 2026 forecast for U.S. retail returns is $849.9 billion. Online is worse, with 19.3% of every e-commerce order coming back.
Why Are Static Returns Costing DTC Brands More Than Free Returns?
But scaling costs is not the same as scaling intelligence.
Most brands still run the same returns portal for every customer. A first-time buyer with a size mismatch and $400 in projected lifetime value sees the same flow as a serial returner with a 78% return rate and $15 in net revenue. Same free label. Same ninety-day window. Same refund pathway. This is the software problem underneath the policy problem. It is not a policy problem at all.
Across the D2C brands we work with in the U.S., 54% of refund-intended returns convert into exchanges when the software reads the customer at the moment they click start return. Not one policy for everyone. A specific offer, generated in real time, based on what the software knows about that customer. The brand keeps the sale. The customer feels seen. And the brand’s return rate, the metric everyone tracks, doesn’t move. But the metric that actually matters, return revenue recovery, moves dramatically.
Why Will Across-the-Board Return Fees Backfire This Holiday Season?
The mistake most brands are about to make in Q4 is to tighten policy across the board. Add fees. Shorten windows. Make everyone pay for the abusers. This is a rational response to a mispriced diagnosis. Free returns are not the cost. Undifferentiated returns are the cost.
Consider what the 2026 holiday season will look like. Close to a trillion dollars in U.S. holiday retail. NRF forecasts 17% of it, more than $170 billion, will come back as returns between late November and January. Q4 is when brands acquire the most customers. Q1 is when they lose them, through refund flows that treat every customer identically. The math of Q4 is decided in Q1.
If your January 2027 returns process reads a $400-LTV first-time buyer and a $15-LTV serial returner the same way, you’re losing the customer you just spent your acquisition budget on.
How Can DTC Brands Protect Q1 2027 Revenue Before Peak Season?
If you’re heading into peak season and you want to protect Q1 2027, here is what I would push you to do before December.
- Stop tightening policy across the board. A generic fee on all returns will lose you your best customers along with your worst. Instead, price the return based on the customer. High-LTV first-time buyers should still see free returns. Serial returners should see a fee. Same portal from the front. Different math underneath.
- Move to exchange-first defaults for the customers who deserve it. The default question in most portals is “why are you returning this,” followed by a shipping label. For a customer with three prior orders and a size mismatch, the default should be “would you like a size 8 instead of the size 6?” A specific alternative that costs the brand almost nothing to offer, and saves the sale 54% of the time.
- Measure return revenue recovery, not return rate. Return rate tells you how much you’re paying. Return revenue recovery tells you how much you’re keeping. If you optimize the first without the second, you will fix a cost line and lose a growth line at the same time.
- Prepare your CX team for the January cliff. Every brand that has a strong Q4 will see a wave of returns hit in January. Most CX teams are still staffed for standard volume. Add capacity, or add automation, but do not let January’s returns queue become February’s cancellation queue.
Sub-scale DTC brands will keep closing over the next twelve months. The story will keep getting told the same way: “returns broke us.” What the story is actually telling U.S. is that a decade-old software approach broke against a category that grew ten times more complex than the software allowed for. That gap is the opportunity.
The fix is to make returns smarter, not to make them harder. The brands that will still be here in 2027 are the ones running software that can read the customer at the moment the return starts. Not the ones with the strictest policies.
Lokesh Kumar is a founding member of ClickPost, the post-purchase decision engine for commerce. He works with mid-market DTC brands across the US, India, and the Middle East on retention economics: how software can turn every return, delivery, and exchange after checkout into a decision that keeps the customer. He writes about post-purchase economics and the metrics that decide whether a first purchase becomes a second one.
