Data
Sizing advice and virtual try-on support different shopping decisions. A size recommendation uses the product's sizing information; a generated image previews appearance. Neither guarantees physical fit. This guide explains the published evidence, its limits and a practical way to evaluate outcomes on your own store.
Hwangbo and colleagues (2020), in IEEE Access, studied a women's casualwear brand using a 3D body model and reported a 27% decrease in the return rate. This was a different technology and retailer from Rendered Fits. See the university research record and original paper.
A separate study by Gallino and Moreno (2018) used randomised field experiments to examine virtual fit information. It found improvements in conversion and order value, with lower costs from returns and customers ordering multiple sizes. Read The Value of Fit Information in Online Retail. Its findings concern fit information; they should not be treated as proof that every generated try-on image reduces returns.
Start with the product data that helps a shopper choose one size. Publish body or garment measurements with clear units, say which they are, and add fit notes such as relaxed cut, stretch or a recommendation for someone between sizes. Check that every size sold appears in the chart.
Our Shopify Size Chart Report 2026 found a text-readable size chart on 31.6% of the 829 stores scanned. This was a non-random sample and a raw HTML scan, with limits described in the report. It identifies information gaps, rather than proving they cause returns. Use the free size chart audit to check your own store.
A product-specific size recommendation turns measurements and fit notes into advice at the size picker. A virtual try-on preview helps the shopper picture style and proportion on themselves. These support different decisions: an appearance preview does not establish physical fit.
Use a relative reduction, not a percentage-point change. For illustration, if a store has 1,000 delivered orders and 250 are returned, its order return rate is 25%. A 20% relative reduction means 200 returned orders and a 20% return rate. A 30% reduction means 175 returned orders and a 17.5% return rate. These are worked examples, not Rendered Fits results.
Keep the denominator consistent. Do not mix returned items with returned orders, or all returns with fit-related returns. Multiply the avoided returns by your own handling and postage cost, then subtract the app and operating costs. Keep retained revenue and saved handling costs separate to avoid double counting.
Rendered Fits combines personalised sizing, Complete the Look and virtual try-on for Shopify fashion brands. For a store evaluation, read the Shopify returns guide and measurement checklist. We have not published a controlled merchant return-reduction result for Rendered Fits.
About Rendered Fits. Rendered Fits is a Shopify app for fashion brands that recommends a size for supported products from the brand's own size chart, and can show shoppers an optional AI try-on preview. There is a Free plan, and Complete is $349 a month with a 14-day trial.
There is no universal figure. Use a defined comparison, consistent order or item counts and a completed returns window. Report uncertainty and costs alongside any observed difference.
Fit and size are among the most common causes of fashion returns. Others include quality expectations, colour mismatch, and changed mind. Check your own returns reasons.
It varies by business. Include shipping, inspection, re-pressing, restocking, and losses on items that cannot be resold, and use your own figures.