Expect about a fifth to a quarter of your online clothing sales to come back. The best published figures sit between 19% and 24%, though none of them is a true match for your store, and your own rate (measured properly and split by product) will tell you far more than any benchmark.
The figures, and how far to trust them
| What it measures | Rate | Source |
|---|---|---|
| UK non-food online purchases, 2025 forecast | 19.5% | Retail Economics and ZigZag, down from 21% in 2024 |
| US online sales, all categories, 2025 | 19.3% | NRF and Happy Returns |
| US sales, all channels, 2025 | 15.8% | NRF and Happy Returns, down from 16.9% |
| US online apparel, year to March 2023 | 24.4% | Coresight Research, as reported by EightX |
Every one of these needs a health warning. We couldn't get to the original Coresight report, so that figure is second-hand. The NRF numbers are estimates that retailers gave in a survey, which isn't the same as audited data. The report behind the first row was produced with a returns company, as a lot of the research in this field is, and it covers all non-food rather than clothing alone.
We also looked hard for a UK clothing figure and didn't find one we were willing to print. You'll see 23.6% and 26% quoted all over the place, but when we followed them back they usually ran out before reaching anyone who had counted anything. What we'd say with some confidence is that clothing runs above the online average, and that the average itself is about one in five.
Three ways to calculate a return rate, and three answers
Before you compare yourself with anyone, decide what you're counting, because the choice moves the answer a long way. Say a store ships 1,000 orders in a month, each with two items and each worth £80, and on 300 of those orders the customer sends back one item worth £40.
| Method | The sum | Rate |
|---|---|---|
| By order | 300 of 1,000 orders had a return | 30% |
| By item | 300 of 2,000 items came back | 15% |
| By value | £12,000 of £80,000 was refunded | 15% |
So the same store, in the same month, can honestly report 15% or 30%, and we suspect a fair number of arguments between ecommerce and finance teams come down to this. The order rate tells you how many parcels your warehouse will open, and the value rate tells you what happened to revenue. We'd use value for finance and items for merchandising, and write down which one you mean. Most published benchmarks appear to be by sales value, so that's the one to hold up against them.
Three mistakes that make the rate look better than it is
Counting too soon. If you give customers 30 days, last month's orders are still on their way back, so any recent month will flatter you. Either wait until the window has closed or report only on orders old enough to have matured, even though that means your dashboard is always a little out of date.
Ignoring what was exchanged. A size swap isn't a lost sale, but the first item still came back and it cost you the same to handle. Count exchanges separately, so that you can see both things at once.
Using the average at all. A store at 22% is often a blend of a few products at 45% and a great many at 10%, and nobody can act on 22%. The number that tells you what to do sits down at the level of the product, the size and the customer.
Where to look in your own numbers
- By product. A dress that comes back twice as often as the rest of the range probably has a sizing or photography problem, and you can fix either.
- By size. If the returns on one style cluster in one size, the size chart for that style is wrong.
- By customer. ZigZag puts 22% of shoppers behind nearly half of returns. Yours will differ, and it's worth finding out by how much.
- By basket. Compare orders with two sizes of the same item against everything else. These are the returns you can see coming before the parcel has even left.
- By discount. Check whether your deepest promotions bring back more than they sell. Some do.
Once you have the rate, the next job is to put a price on it, and our returns cost calculator will turn it into pounds. If you'd rather know what's behind the number first, start with why clothes get returned.