Why 72% of Fashion Returns Are Fit Problems — And What to Do About It

Returns are eating fashion margins. The average UK fashion brand loses £25 per returned item when you factor in reverse logistics, repackaging, and restocking. At scale, that's not a rounding error — it's a business model problem.

But here's what most brands miss: the majority of returns aren't preference problems. Customers aren't changing their minds. They're sending things back because the garment didn't fit.

Research consistently puts the figure at 72%. Nearly three quarters of all fashion returns come down to fit. Not quality. Not colour. Not a better price found elsewhere. Fit.

That number should reframe how you think about the returns problem entirely.

The Wrong Fix

Most brands respond to high return rates with policy changes. Charge for returns. Tighten the window. Make the process less convenient. These tactics reduce the volume of returns processed, but they don't reduce the underlying reason customers are returning — they just shift the friction onto the customer.

The result is predictable: lower satisfaction scores, reduced repeat purchase rate, and a growing number of customers who simply don't buy again because they're not confident they'll get the right size.

Charging for returns is treating the symptom. The cause is that customers couldn't accurately predict fit before they bought.

Why Size Guides Don't Work

The standard solution — detailed size guides — has been around for decades. It doesn't work well for three reasons.

First, customers don't measure themselves accurately. Most people don't own a tape measure, and even those who do rarely measure in the right positions or at the right tension.

Second, size guides describe the garment, not the fit. Knowing a medium has a 96cm chest circumference tells you nothing about how that garment will feel on your specific torso length, shoulder width, or arm proportions.

Third, fit is subjective in ways that measurements alone can't capture. A fitted medium on one body shape feels completely different on another — even if the chest measurement is identical.

The result is that customers are making purchase decisions based on guesswork, and returning items when the guess turns out to be wrong.

The Purchase History Approach and Its Limits

Some enterprise brands have moved to AI-powered size recommendations based on purchase history. The logic is reasonable: if someone bought a medium three times and kept it, recommend medium.

But this approach has significant gaps for any brand that doesn't have years of transaction data per customer, or for any customer who is new to the brand or who has recently changed shape. It also can't account for different fits across different products — the medium in your relaxed hoodie range is not the same as the medium in your fitted technical jacket.

Purchase history tells you what someone bought. It doesn't tell you why they kept it or whether it actually fit.

What Actually Works

The only way to reliably reduce fit-related returns is to give customers confidence in fit before they purchase. Not after. Before.

This means connecting the customer's actual body measurements to the specific garment they're considering, at the moment they're considering it.

The technology to do this is no longer limited to enterprise. Body measurement AI — the kind that reads shoulder width, chest, waist, hips, torso length, arm length and more from a single photo — is now available as a lightweight embed that sits directly on a product page.

GarmCheck

UI screenshotGarmCheck widget — upload screen on product page

When a customer can see how a garment actually looks on a body similar to theirs, and receive a size recommendation based on their own measurements rather than a generic chart, the dynamic changes. They buy with confidence. And confident buyers don't return.

The Numbers

Brands implementing virtual try-on with body measurement are seeing 20–30% reductions in fit-related returns. On top of that, conversion rates improve — customers who engage with a try-on tool are 27% more likely to purchase, because uncertainty is removed from the decision.

The economics work quickly. A brand processing 500 returns per month at £25 per return is spending £12,500 per month on the problem. A 25% reduction saves £3,125 per month. At £99 per month for a Starter plan, that's a 31x return.

GarmCheck

UI screenshotGarmCheck size recommendation panel — XS → XXL with limiting-factor breakdown

The Actionable Takeaway

If your return rate is above 20% and you sell fitted clothing, the overwhelming probability is that fit is the primary driver. Policy changes will reduce your returns volume at the cost of customer trust. Size guides will continue to underperform.

The fix is upstream: give customers a way to see how your garments look on their body, with a size recommendation based on their actual measurements, before they add to cart.

That's the problem GarmCheck was built to solve.