Size and fit come first, by a distance. After that come colour and looks, damage or the wrong item, and second thoughts. Each has a different owner in your business and a different moment when it can be fixed, and most of them can be dealt with before the parcel comes back.
What the numbers say
The figure you'll see most often comes from Coresight Research's 2023 survey of US apparel sellers, which we've only read second-hand, where 53% named size and fit as the top reason. Narvar found much the same in 2019, with wrong size, fit or colour the top reason for 46% of shoppers at retailers other than Amazon. ZigZag, which handles UK returns, says the same thing without giving a number.
None of this is precise, and we'd be wary of anybody who says theirs is. A return reason is a box ticked by someone who wants their money back and would like to get on with their day, so they pick the first option that will do. "Too small" might mean too small, or it might mean "I tried it on in the bedroom mirror and felt terrible". Use the reasons to rank your problems, but don't expect them to tell you how big each one is.
The reasons, and what can be done
| Reason | Who can fix it | When |
|---|---|---|
| Too small or too big | Merchandising, product pages | Before purchase |
| Ordered two sizes on purpose | Customer service, your policy | Before dispatch |
| Does not look like the photo | Photography, copy | Before purchase |
| Fabric or quality disappointed | Buying, copy | Before purchase |
| Damaged or wrong item | Warehouse | Before dispatch |
| Arrived too late | Delivery promise | Before purchase |
| Changed mind | Mostly nobody | Accept it, and offer an exchange |
Size and fit
A size chart is a table of a garment's measurements, but what the shopper wants to know is whether it'll fit them, and most returns start somewhere in the gap between those two things.
What helps is unglamorous, which may be why it so often doesn't get done. Measure each style, because a brand's size 12 is rarely one size across a whole range. When a style runs small, say so next to the size selector, where people will actually read it, rather than on a separate page they'll never open. Show the model's height and size, and publish what past customers said about the fit (including the ones who weren't happy). Then look at your returns by style and size, and fix the worst ten before you do anything cleverer. A fit tool may well help, but test it against a control group, since the vendors' case studies tend to be the good ones.
Bracketing
Bracketing is buying a jacket in two sizes and sending one back, and in that same 2019 Narvar report 48% of US consumers said they did something like it. It isn't fraud. Happy Returns' David Sobie told the NRF it's a normal part of modern ecommerce, and if you think about it from the shopper's side it's perfectly sensible: nobody has told them which size to buy, so they've built their own changing room in the spare bedroom.
From your side it guarantees a return, but it's also the one return you can see coming, because it's sitting right there in the basket. That makes it the cheapest kind to deal with. Between checkout and dispatch, you can send a message saying that they've ordered this in M and L, that most people their height keep the M, and asking whether they'd like to change. Some will say yes, in which case you've saved the postage both ways and they've saved a trip to the parcel shop. Some will say no, and you should let them.
Looks different in the flesh
Colour shares the top line with size and fit in Narvar's report, and it's a problem you can mostly photograph your way out of. Shoot every colourway rather than the best one, and shoot them in daylight as well as under studio lights. Describe the fabric the way a good shop assistant would, who might tell you it's "soft but quite thin", and whom you'd never catch saying "premium feel". Copy that sells harder than the garment can deliver gets its reward in the returns queue.
Damaged or wrong
These ones are your fault, and in the UK they're free to the customer whatever your policy says. They're also the easiest to measure, so report them by warehouse and by carrier, separately from everything else, or they'll hide in the headline rate where nobody has to answer for them.
Changed mind
Some of this is simply what it costs to sell clothes without a changing room, and you won't get rid of it. You can convert some of it, though. A customer who wanted the green one may be happy with the blue, and someone who was about to ask for a refund may take a size up if it's one tap and in stock. Do keep the refund easy to reach, because a retailer that hides it gets the refund anyway, along with a complaint.
What to do on Monday
- Export the last 12 months of returns with reason, style, size and basket.
- Check the reasons you collect. Shopify now has category-specific return reasons, which are a good deal better than a single dropdown.
- List the ten styles that cost you the most in returns, in pounds rather than as a rate.
- Read the customer comments on those ten. Fix the pages, then measure again once the return window has closed.
- Count how many baskets hold two sizes of one item, because that's your opening for a message before dispatch.
If you'd like to know what all this is worth to you, try the returns cost calculator.