Market report
Why clothes get sent back — and how try-on helps
Apparel is the most-returned category in online retail, and the reasons are remarkably consistent: it didn't fit, or it didn't look the way the shopper expected. This report summarizes the scale of the problem and how virtual try-on is reported to help close the gap.
Last updated June 2026 · wearfits.me editorial
The scale of the problem
Returns are a structural cost of selling clothing online. Industry estimates consistently put online apparel return rates well above other ecommerce categories — often roughly two to three times the rate of general merchandise — because shoppers can't try anything on before it ships. A common pattern is "bracketing," where a shopper deliberately orders several sizes or colors intending to keep one and send the rest back, which inflates return volume on top of genuine fit misses.
Each return carries return shipping, inspection, repackaging, and markdown or write-off costs, plus the lost margin on items that can't be resold at full price — so even a modest reduction in returns has an outsized effect on profitability and on the environmental footprint of shipping garments back and forth.
Why shoppers send clothes back
Across reported breakdowns, the leading reasons cluster tightly around fit and appearance. The table below shows the typical pattern of apparel return reasons (illustrative shares, reported ranges):
| Return reason | Typical share | What it means |
|---|---|---|
| Doesn't fit (too big / too small) | Largest single group | Size guessed wrong from a chart or label |
| Doesn't look as expected | Second-largest group | Color, cut, drape, or style differed from the photo |
| Ordered multiple sizes (bracketing) | Notable share | Planned returns to compensate for sizing uncertainty |
| Quality / other | Remainder | Damage, changed mind, late delivery, etc. |
The headline takeaway is that fit and "look" together dominate apparel returns. Both are forms of the same problem: the shopper couldn't accurately picture the garment on their own body before buying.
How virtual try-on is reported to help
Because the leading return reasons are fit and appearance, the most direct lever is to let shoppers see the garment on themselves before they buy. Generative-AI virtual try-on renders the item on the shopper's own body — photorealistically, including how the fabric drapes — which is designed to shift the decision from "I'll order it and see" to "I can see this works." Reported benefits include higher purchase confidence, more engagement on product pages, and fewer "doesn't fit / doesn't look right" returns compared with static photos and size charts alone.
These are directional, reported outcomes rather than a guaranteed figure for any single store; impact depends on the category, the catalog, and how prominently try-on is offered. The mechanism, though, lines up cleanly with the data: address the expectation gap and you address the biggest slice of returns.
What the published data shows about services like this
The case isn't just mechanism — it's measured. Independent reports on deployed virtual try-on and fit services (the same category as the generative-AI apparel try-on demonstrated here) point the same way:
| Reported finding | Context | Source |
|---|---|---|
| 55% returned an item because it looked different on them than expected | Online apparel shoppers | eMarketer, 2025 |
| −40% returns on jeans after rolling out virtual try-on | Zalando, cited in Savills report | Savills, "AI and the Future of Physical Retail," 2026 |
| 12% conversion with virtual advisors vs 3% for traditional ecommerce | Virtual advisor / try-on assisted journeys | Savills, 2026 |
| 80% higher post-try-on confidence; up to 8× return on ad spend | Snap AR try-on (Alter Agents study) | Snap for Business, 2022 |
Adoption is following the evidence: Zalando plans to roll virtual try-on out to all customers in 2026, Walmart offers try-on across 270,000+ items, and Google has added a virtual dressing room to its AI-mode shopping search. The generative-AI apparel try-on you can try in our demo sits in exactly this category.
Methodology & sources
The figures on this page are presented as industry estimates and reported ranges, synthesized from widely published ecommerce and retail-returns commentary, not as precise measurements from a single named study. Return rates and reason breakdowns vary by retailer, region, season, and product mix, so we describe patterns and relative magnitudes rather than exact percentages. Where we describe what virtual try-on does, those capability claims reflect the generative-AI apparel try-on documented in our live demo. Shoppers looking to understand the root cause should read Does it fit me?, and the method-by-method view is on the comparison page.