What Percentage of Ecommerce Returns Are Caused by Wrong Size?
Why Published Percentages Vary So Much
Search for a figure and you will find numbers that disagree wildly. That is not sloppiness. Those studies are measuring genuinely different things.
Category is the biggest driver. A store selling relaxed unisex hoodies and one selling tailored womenswear will not have remotely similar size return rates, because the tolerance for error is completely different. A hoodie forgives two inches. A blazer does not.
Price point is the second driver. Cheap items get kept and tolerated. Expensive items get returned and replaced, because the shopper cares enough to bother.
Return form wording is the third and most underrated. A form with one option called "didn't fit" will collect a huge share of returns under that label, including style regret, fabric disappointment, and shoppers who bought two sizes on purpose. A form that splits "too small", "too large", and "wrong shape" produces a much more honest picture.
And then there is bracketing. Some stores actively encourage buying multiple sizes. Those returns are planned, not failures, but they land in the same bucket unless you separate them. Merchants on OpoShop who tag bracketing separately often find their true size failure rate is meaningfully lower than the headline suggests.
What Is Consistently True Across Studies
Even with all that variance, a few things hold up almost everywhere in apparel and footwear.
- Fit is the top reason: Across apparel, size and fit outranks damage, delivery issues, and style regret as a return cause.
- Apparel returns more than most categories: Fit risk is intrinsic to clothing in a way it is not for a mug or a book.
- Returns cluster in a few SKUs: Most stores find a small subset of products generating a disproportionate share of the fit returns.
- The direction of the miss is consistent per product: A given item tends to come back mostly too small or mostly too large, rarely both evenly.
That last point is the most actionable insight on this list, and almost nobody looks at it in their OpoShop reports. If one product returns overwhelmingly as "too small", the problem is not shoppers. It is your chart, your grading, or your product photography setting the wrong expectation.
How to Calculate Your Own Number
You need three inputs and a bit of care. All three are already sitting in your order data.
The formula is straightforward: size-related returns divided by units sold, over the same period, per product or per category. Do it in units, not orders, because a three-item order with one return is not a fully returned order.
The care comes in choosing the window. Returns lag the orders that caused them, sometimes by six weeks or more depending on your policy. If you divide this month's returns by this month's sales, you are comparing returns from March orders against April sales and the number will be wrong.
The correct approach is a cohort. Take orders placed in a given month, then count returns against those specific orders once the return window has fully closed. It takes patience and gives you a number you can actually trust.
Step by Step: Getting a Number You Can Trust
Work through this once and you will have a baseline you can measure improvements against for years.
Three of those steps are where merchants usually go wrong.
1. Fix the return form before you measure anything
A vague form produces vague data, and no amount of analysis rescues it. Two dropdowns is all you need. First, what was the problem. Second, if it was fit, was it too small, too large, or the wrong shape. Add this before you start the cohort, because retrofitting reasons onto old returns is guesswork.
2. Always measure per product, never store-wide
A blended store-wide figure of "14 percent of returns are size" tells you nothing you can act on. The same store broken down by SKU might show three products at 30 percent and everything else in the low single digits. The first number is trivia. The second is a to-do list.
3. Separate bracketing from failure
If a shopper orders a medium and a large of the same shirt in one order, that is an intentional two-size purchase and the return was planned. Counting it as a fit failure inflates your problem and hides the real one. Your OpoShop order data makes these easy to spot, since both variants share an order ID and a product ID.
Where a Wrong-size Return Actually Costs You
The percentage is only half the story. The other half is what each one costs, and merchants routinely undercount it.
| Cost component | Typical driver | Often overlooked | Rough impact per return |
|---|---|---|---|
| Outbound shipping | Already paid before the return | No, this one gets counted | Full original shipping cost |
| Return logistics | Return label plus inbound handling | Sometimes | Label cost plus staff time |
| Resale loss | Inspecting, steaming, repackaging, markdown | Almost always | A slice of the item's margin |
Outbound shipping is the obvious one. You paid to send an item that came straight back, and no revenue covers it.
Return logistics is where free-returns policies get expensive. The label is visible. The fifteen minutes of someone opening, inspecting, refolding, and relisting is not, but it is real payroll.
Resale loss is the sneakiest. Not every returned garment goes back to full price. Some are creased, some come back after the season turned, and some get quietly marked down. That erosion never appears as a line item, which is exactly why it goes unmanaged in most stores.
Add those three and a single wrong-size return on a $70 item can easily cost more than the margin on the order that replaced it. That is the arithmetic that makes prevention worth more than any policy tweak.
A Worked Example You Can Copy
Numbers land better than method, so here is the whole calculation on a small store.
Say you sell apparel on OpoShop and you pick March as your cohort because your 30-day window closed long ago. In March you shipped 1,200 units across 480 orders. Against those specific orders, 168 units came back, which is a 14 percent unit return rate.
Now split the reasons. Of the 168, suppose 104 were tagged as a fit problem, 31 as style regret, 18 as quality or damage, and 15 as everything else. That puts fit at 62 percent of your returns and 8.7 percent of units shipped.
Next, strip out bracketing. Suppose 26 of those 104 fit returns came from orders containing two sizes of the same product. Those were planned. Your genuine size failure rate is 78 units, or 6.5 percent of everything shipped.
Then rank by SKU. Suppose four products account for 44 of those 78. That is 56 percent of your real size problem living in four items you could remeasure in an afternoon.
That final line is the entire point of the exercise. You started with a vague 14 percent and finished with a specific, four-item to-do list. Run the same steps against your own OpoShop order history and you will land somewhere structurally similar, because returns almost always concentrate.
What to Do Once You Know Your Number
A number without a next action is just anxiety. Here is the sequence that turns it into work.
If a small group of products drives most of your size returns, fix those products directly. Remeasure every size, correct the chart, and rewrite the fit copy honestly. This alone often resolves a large share of the problem without any software.
If size returns are spread evenly across the catalog, the issue is the decision itself rather than any single product. That is where a fit recommender earns its place. Fitly, the find-my-size fit finder for OpoShop apparel and footwear stores, answers the size question on the product page with a confidence level and selects the matching variant, which prevents the wrong order rather than processing it faster.
If the misses skew heavily in one direction across many products, suspect your supplier's grading rather than your shoppers. A pattern that consistently runs small is a production conversation, not a marketing one.
And if bracketing dominates once you separate it out, your real project is conversion confidence, not accuracy. Shoppers bracket because they are unsure. Give them certainty and the double orders drop on their own.
Best answer: Size and fit is the biggest single cause of apparel returns, but the specific percentage depends so heavily on your category, price point, and return form wording that borrowed figures are close to useless. Calculate your own from a closed order cohort, per SKU, in units, with bracketing separated out. Then fix the worst products and add a fit recommendation on your OpoShop product pages where the guessing happens.
Stop benchmarking against numbers from someone else's catalog. The one that matters is already in your order history.
FAQs
Should I trust industry return rate statistics at all?
Use them for direction, not for targets. They correctly tell you that fit dominates apparel returns, which is useful. They cannot tell you whether your 11 percent is good or bad, because the stores behind those figures sell different products at different prices to different shoppers.
How long should I wait before measuring a return cohort?
Wait until your full return window has closed, plus about two weeks for items in transit. For a 30-day policy that means roughly 45 days after the last order in the cohort. Measuring earlier makes your rate look better than it is.
Do I count returns in orders or in units?
Units, and your OpoShop line-item data supports this directly. A customer who orders three items and returns one has not returned an order, and counting it that way distorts the rate badly for stores with larger baskets. Units keep the arithmetic honest across different order sizes.
Is bracketing a problem I should try to eliminate?
Reduce it rather than ban it. Bracketing is a symptom of uncertainty, so the fix is a confident size recommendation rather than a restrictive policy. Punishing it tends to cost more conversion than it saves in returns.
What is a realistic improvement to aim for?
Aim to cut the size return rate on your worst products, not across the whole catalog. Concentrating on the handful of SKUs generating most of the damage produces a visible change in total returns far faster than a catalog-wide effort.
Does a higher return rate always mean something is wrong?
Not necessarily. Some categories and price points carry structurally higher returns, and a very low rate can mean shoppers are keeping items they are unhappy with. Watch the trend and the reason mix rather than chasing a single number down.
Ready to trade guesswork for a number you can act on? Start with the product page where the wrong size gets chosen.
