October 29, 2026
Returns are a data problem before they're a logistics problem
Most return-reduction effort goes into the warehouse and the policy. The causes start much earlier: in fit data, size information and how the product is shown.
The short answer
Fashion returns are usually treated as a logistics cost, but most of them are caused upstream: fit that does not match expectation, sizing data the brand does not control, and product pages that misrepresent colour or proportion. The Interline's 2026 spotlight on fit cited NRF and Statista estimates that returns cost US retail around $890 billion in 2024. Reducing them starts with better product and fit data, not a stricter returns policy.
Key takeaways
- Returns cost US retail an estimated $890 billion in 2024 (NRF and Statista, as cited by The Interline).
- Fit is a data and preference problem: brands often lack current size data or control of their blocks.
- Accurate imagery addresses part of the problem (colour, proportion, detail); it cannot fix fit.
- Start with the product record: one accurate set of facts feeding the page, the size guide and the factory.
How big is the returns problem?
The Interline's Spotlight On: Fit & Body Data cited figures from the National Retail Federation and Statista estimating that returns cost US retail alone around $890 billion in 2024. That figure covers all of retail, not only fashion, and we have not seen the underlying method, but it is consistent with what apparel teams report: returns are one of the largest controllable costs in online fashion.
The instinct is to treat returns as a logistics problem: better reverse logistics, stricter policies, return fees. Those help at the margin. They do not touch why the item came back.
Why do fashion items come back?
| Cause | Where it starts | What can reduce it |
|---|---|---|
| Fit not as expected | Size data, blocks, grading, preference | Better fit data and size guidance; consistent blocks |
| Looks different from the picture | Product imagery and description | Faithful imagery reviewed against the product |
| Wrong colour | Imagery and colour management | Colour checked against the approved swatch |
| Quality not as expected | Materials and description | Accurate material and detail shots |
| Bought several to choose | Shopper behaviour | Confidence from accurate information |
Why is fit so hard to fix?
The Interline's spotlight argues that fit is mainly a data and preference problem rather than a pure technology one. Brands often lack current size data, suppliers may control the blocks, and fit is a personal preference that varies by shopper. Body scanning meets resistance from shoppers, so the data that would help most is the hardest to collect. The instalment is sponsor-supported, but the diagnosis is one we hear widely.
The practical lesson is that no single tool fixes fit. It improves when the brand owns consistent blocks and grading, keeps measurement data current, and shows shoppers honest information about how a product fits.
What can better imagery fix, and what can it not?
Accurate imagery addresses the returns that come from a gap between picture and product: colour shifts, misleading proportions, details that were not visible. On Apparel Resources, the founder of a virtual try-on company reported improvements in engagement, conversion and returns for its clients; those are the company's own claims, but they point in a plausible direction.
What imagery cannot do is change how a garment fits. A beautiful on-model image of a dress that runs small will still come back. We think brands should be clear-eyed about this: faithful images reduce one class of returns, fit data reduces another, and the two need different work.
Where should a brand start?
1. Find out why things come back
Code return reasons by style and size. Many brands have the data and do not analyse it.
2. Fix the record first
Make sure the product facts (colour, material, measurements) are correct in one place.
3. Make the page match the product
Review every image against the approved product before publishing.
4. Own the fit basics
Keep blocks and grading consistent across suppliers and seasons.
5. Close the loop
Feed return reasons back to design and development, not just to the warehouse.
What does this look like in practice?
An illustrative example. The brand and the details are invented to show the idea; they are not a case study.
An online dress brand codes a season's returns by reason and finds two clusters. One style comes back for being tight across the shoulders in two sizes, a grading issue traced to a supplier using an older block. Another comes back for colour: the product page showed a dusty pink that arrived as a warmer coral.
The fixes are different. The fit problem goes to development, which aligns the supplier on the current block and updates the size guidance. The colour problem goes to the content team, which reshoots or regenerates the images against the approved swatch and adds a note about the shade. A stricter returns policy would have touched neither cause.
Where CotonAI fits.
CotonAI works on the part of the problem that sits in the product record and the product page. Product photography and AI fashion models start from the real product and are built to keep colour, shape, texture and labels true; every image belongs to the right style and colourway and is approved before it ships. CotonAI does not measure bodies or recommend sizes, and we do not claim it fixes fit. See CotonAI for retailers.
Sources.
We read these pieces and wrote this analysis in our own words. Figures are credited to the source each article gives; vendor and company figures are their own claims.
- The Interline, Spotlight On: Fit & Body Data, 22 September 2026. Sponsor-supported instalment
- Apparel Resources, AI, computer vision and 3D modelling are boosting VTO accuracy, Dheeraj Sharma, 23 April 2025. Interview with a vendor
Common questions
How much do returns cost retail?
The Interline cited NRF and Statista estimates of around $890 billion for US retail in 2024. The figure covers all retail, not only fashion.
What is the main cause of fashion returns?
Fit not matching expectation is the largest driver in apparel, followed by products looking different from their images, including colour.
Can AI imagery reduce returns?
Faithful imagery can reduce returns caused by colour, proportion or detail mismatches. It cannot fix fit, which needs better size and fit data.
How do brands start reducing returns?
Analyse return reasons by style, fix product facts in one record, review images against the product, keep blocks and grading consistent, and feed reasons back to design.
Keep reading
Make the page match the product. Then measure the returns. We'll show you the workflow end to end and recommend where to start. Book a demo