Returns are retention
An independent take on Loop's 2026 Retention Report
Loop’s just published its 2026 Global Ecommerce Retention Report.
It’s mostly benchmarks. Paid returns are now the default, and exchange behaviour varies wildly by region.
But the report keeps coming back to two important moves brands should prioritise. Charge for returns more often. Convert refunds into exchanges or credit so the cash stays with you.
Not revolutionary concepts, but the report backs up the claim with data.
Loop also says brands kept $516m in 2025 by improving return outcomes. The headline numbers are punchy, but they’re what you’d expect from a returns vendor.
Why we’re doing independent peer-led research
Returns are framed as money you keep, not just a cost you cut.
That part’s right. But the returns space is also full of noise, and most of it comes from vendors, not the people managing carriers and trying to make disparate systems talk to each other.
That’s why we’re doing peer-led research with brand operators, and publishing the full benchmark in March 2026. We shared a first read late last year, and this is the next step.
A return isn’t just a cost line. It’s the moment you either keep the customer, or you hand them a clean exit and call it inevitable.
Returns are now a leadership problem
Loop’s point is simple. Treat the return as part of the order, not the annoying bit you dump on ops after the money’s been taken.
No operator will argue with that. Returns touch ERP, WMS, finance, 3PL relationships, and CX, and you don’t get to improve one without making trade-offs in the others.
If you’re in finance or commercial leadership, you can’t pretend it’s just an ops workflow anymore. Your returns policy is pricing, loyalty, and cash flow.
Paid returns are now normal
Loop’s strongest point is that paid returns aren’t a debate anymore. They’re spreading because the maths stopped working.
Loop puts it at 70 percent of brands now charging return fees. That lines up with what we’ve heard directly from operators.
The bit that matters is what happens next. Brands aren’t just charging to cover costs; they’re trying to nudge behaviour without triggering a PR sh*tstorm.
The more interesting bit is tiering. Some brands are quietly making free returns a loyalty perk, and charging everyone else a modest fee.
Loop sells return fees as a retention move, not just a margin move. That only works if your policy’s clear and consistent.
If you roll it out in a panic, or you change it twice in six months, customers notice.
Exchanges aren’t a feature, they’re a behaviour change
Loop’s also right to push exchanges and store credit as the win. Converting a refund into an exchange keeps money with you, and sometimes turns a churn moment into another purchase.
In Loop’s benchmarking, UK shoppers exchange about 5.8 percent of returns, versus roughly 17 percent in the US. The UK is basically refund-first, by habit.
Here’s the rub. If you just switch exchanges on in the portal and call it a day, refunds still win.
We’ve heard versions of the same story from brands. One UK operator told us only about 2 to 3 percent chose exchange or credit at first, and it only moved after deliberate work.
Then they had to do the unsexy work. Redesign the portal flow, add incentives, and make the exchange feel like the obvious choice.
If you want the teaser for March, it’s this. The exchange rate is rarely a tooling problem.
It’s a commercial decision that shows up in UX, fulfilment constraints, and how much hassle you’re willing to take out for the customer.
Loop’s report also throws out ideas like opt-in return coverage at checkout. It’s clever, and it’ll work for some brands.
But it’s still the same game. You’re shaping expectations before the order is placed, and you’re trying to avoid the post-purchase argument later.
Capable tools inside fragile systems
Here’s the part vendor reports always gloss over. Retention doesn’t come from having a returns platform.
It comes from the whole chain doing its job, including the boring bits between systems and teams.
In our first-look write-up we used a phrase that’s stuck for a reason. Returns platforms are capable tools inside fragile systems.
Most operators feel broadly positive about the tool they use, but less positive about the whole setup. The friction tends to show up where integrations behave inconsistently across regions, and where international processes never quite align.
This is where Loop’s report is right on paper, but it leaves the hard bit out. A slower international returns loop can absolutely drag repeat purchasing, and UK return habits don’t change just because you’ve got a nicer portal.
Zoom out and it’s the same story across returns and cross-border. Integration quality dictates most of the bottlenecks, and the level of manual work is still higher than anyone wants to admit.
That’s why vendor benchmarks often feel glossy. They don’t capture the week you spend unpicking why refunds are out of sync between your returns portal, your finance rules, and the warehouse reality.
Bring this to your next leadership meeting
If you want to get anything out of Loop’s report, use it to force a decision. Stop arguing about whether returns matter and start deciding what outcome you want.
Quick gut-check. Look for these three blockers.
You can’t pull returns outcomes split by refund, exchange, and credit without manual wrangling.
A basic return has handovers and dead time you’ve normalised between the portal, the warehouse, and finance.
Refund is still the easiest path in your portal, so you’ve built the outcome you’re getting.
Why the peer benchmark will be different
Loop’s report is useful. It’ll start the right fight internally, and it’s encouraging the industry in a direction that makes sense.
But it’s still a vendor report. It can’t show you the messy edges where the commercial outcome gets decided, and that’s exactly where operators keep pointing.
In March 2026 we’ll publish the full peer-led benchmark across returns and cross-border platforms. We’ll go a lot deeper on what it takes to turn a platform feature into retention you can actually bank.
Vendors can’t rate themselves. Operators can. Subscribe to get the research first: