Research
Useful virtual try-on statistics need a source, a date, a clear denominator and an explanation of what was measured. This page brings together Rendered Fits original storefront research and external returns evidence, reviewed on 1 October 2026. Sizing information, appearance previews and commercial outcomes are different measures.
The Shopify Size Chart Report 2026 scanned public information on 24 September 2026: 829 stores with clothing sizes and 107,138 sized products. It used public feeds, product pages and linked guides without executing JavaScript. The non-random sample is not an estimate for all Shopify stores, and it did not measure physical fit, sales or returns.
31.6% of the 829 stores had a chart the scanner could read as page text. It found no chart for 27.7% of those stores. A chart missing from raw HTML can still be available to shoppers after a click or JavaScript. Source, method and all published measures.
20.3% of the 107,138 sized products contained recognised fit notes. Separately, 47.7% of the 262 stores with a readable chart used one measurement unit only. These percentages have different denominators and must not be added together. Values are rounded; do not derive exact counts from them. Download the aggregate data.
The returns evidence and measurement guide distinguishes published studies from results on your own store. Hwangbo and colleagues (2020) reported a 27% decrease at one different casualwear retailer using 3D virtual try-on. That study does not establish a universal returns reduction, a Rendered Fits merchant result or a guarantee for your store. Cite the study context alongside its figure.
A return rate depends on the category, market, customer mix and measurement window. State whether you count orders, items or revenue and allow the return window to mature. For conversion, distinguish attributed purchases after a fitting-room interaction from a causal uplift. The merchant measurement plan explains the definitions and a controlled comparison.
Calculate the cost of a return from your own postage, handling, payment costs and recoverable stock value. Segment mobile traffic using your store analytics. Measure image latency with start and completion timestamps, including failed and abandoned attempts. A vendor headline is not a universal cost, traffic-share or speed benchmark.
For reuse, cite the canonical research report with its collection date, sample, denominator and relevant limits. Its plain-text summary, CSV data and BibTeX citation are available under CC BY 4.0. For current product capabilities and pricing, use the separate product facts.
This page does not establish an industry-wide fashion return rate. Measure your own mature order or item cohorts, state the denominator and segment by category, market and return reason before comparing results.
There is no universal figure. A 2020 study reported 27% at one different retailer using 3D virtual try-on; it is not a controlled Rendered Fits merchant result, an industry average or a guarantee.
Calculate your own return postage, handling, payment costs and recoverable stock value. State the category, market and costing method. This page does not establish a universal cost per return.