Conclusion: most apparel P&Ls carry one blended “returns” line, which hides the part you can actually move. Split it and the fit slice stands out: online apparel is returned at roughly 24% (Coresight), and size/fit is the #1 cited reason for about 53% of those (Coresight) to ~70% including style (EEA). Multiplied out, fit alone accounts for ~13–17% of every apparel order returned — a pre-purchase confidence cost, not a logistics one. Size it with one line of arithmetic before spending on a fix.
A return where the item was fine but wrong — too big, too small, or it sat differently than the photo suggested. Not a defect and not buyer's remorse: a “couldn't picture it” problem, which is the return category most influenced by pre-purchase confidence. Note the language: size/fit is the cited reason in surveys, which is weaker than saying fit caused the return.
Two sourced numbers set it. Coresight put the U.S. online apparel return rate at ~24% (24.4% for the 12 months ended March 2023) — the most-returned category. Of those returns, size/fit is the top cited reason: 53% in Coresight's survey (vs 16% color, 10% damage), rising to ~70% in the European Environment Agency's read when style is bundled with fit. Multiply: fit-driven returns run ~13% of orders at the low end and ~17% at the high end.
Four inputs, one multiplication — three from your own store, one benchmark you should replace with your real number.
| Input | What it is | Where to find yours | Benchmark if you can't |
|---|---|---|---|
| Units | Apparel units sold in the period | Store analytics / orders export | use your own |
| Return rate | Share of apparel orders returned | Returns report | ~24% online apparel (Coresight) |
| Fit share | Share of returns tagged size/fit | Return-reason codes | 53% cited / ~70% incl. style (Coresight/EEA) |
| Per-return cost | All-in cost to process one return | Ops + finance | your number — include markdown, not just postage |
Fit-return cost = units × return rate × fit share × per-return cost. The per-return input most teams use undercounts: for apparel it should carry the markdown or write-down on goods that come back out of season or can't be resold at full price, not just the return label.
Worked example: a store selling 40,000 apparel units a year at a 24% return rate takes back 9,600 items; at the midpoint fit share (~60%), 5,760 are fit returns. At an illustrative all-in $20 per return (an assumption — plug in your own), that's ~
Returns hit the P&L as one blended number, so the biggest and most-fixable slice hides in plain sight: fit isn't a logistics cost, it's a pre-purchase confidence cost wearing a logistics disguise.
It's the only large return category decided before checkout. A defect is a QC problem and a changed mind is often unavoidable, but “it didn't look like I expected” is a confidence gap on the product page. In a 4,028-person study, try-on users were 80% more confident and 67% less likely to return (Snap + Publicis); 59% say a try-on helps them picture the item (Nosto). That's the mechanism, not a promise — a try-on can't erase genuine size mismatches, and we won't quote a return-reduction percentage we can't defend.
“$890 billion in returns” is real but it's ALL U.S. retail (16.9% of 2024 sales, NRF/Happy Returns), not apparel — quoting it as an apparel number is a common error. The honest apparel figure is a rate (~24% online) applied to your volume, which is what the formula does.
A return where the size or fit was wrong — too big, too small, or it sat differently than the photo — rather than a defect or a changed mind. A “couldn't picture it” problem, and the return category most influenced by pre-purchase confidence.
Multiply apparel units sold × return rate × the share of returns tagged size/fit × all-in cost per return. Benchmarks: ~24% online apparel return rate (Coresight) and 53–~70% fit share (Coresight/EEA). Make the per-return cost your true figure, including markdown on stock you can't resell at full price.
Size/fit is the #1 cited reason: ~53% of online apparel returns in Coresight's survey (vs 16% color, 10% damage), up to ~70% in EEA data when style is bundled with fit. “Cited reason” is a survey answer, not a proven cause.
It targets the pre-purchase confidence gap that drives fit returns, and studies show try-on users are more confident and less likely to return (Snap + Publicis, N=4,028). But it can't erase genuine size mismatches, and any specific reduction depends on your store — measure it with a holdout.