On this page
A single-brand store has one size chart to get right. A marketplace has hundreds, each written by a different seller with a different tape measure and a different idea of what "medium" means. That inconsistency is quietly one of the biggest drivers of returns and support tickets on multi-seller platforms, and it gets worse with every new seller onboarded.
Multi-seller marketplaces — from general resale platforms to vertical apparel aggregators — face a sizing problem that single-brand ecommerce simply doesn't have. A brand can standardize its own size chart once and enforce it across every SKU. A marketplace can't. Every seller brings their own catalog, their own measurements (or lack of them), and often their own regional sizing convention. The result is a fit experience that is inconsistent by design, and shoppers feel it immediately when a "size M" from one seller fits nothing like the "size M" from another.
Why Marketplace Sizing Breaks Down at Scale
The core issue isn't bad sellers — it's heterogeneity. A marketplace might have a seller in Turkey listing in EU sizes, a reseller in the UK using UK conventions, and a boutique brand using its own vanity sizing, all under the same product category page. Shoppers comparing three near-identical jackets from three different sellers have no reliable way to know which one will actually fit, because the size label means something different in each listing.
This is compounded by resale and secondhand inventory, where the original brand's size chart may not even be listed, or the item has been relisted with only a photo and a guess at size. Apparel and footwear already carry return rates well above other retail categories — shoes are consistently among the highest — and marketplaces amplify that baseline problem because there's no single source of truth for fit. Understanding why clothing sizes are different between brands is useful background, but on a marketplace the variation isn't just brand-to-brand — it's seller-to-seller on the exact same nominal size.
The Cost of Inconsistent Sizing Across Sellers
Returns on marketplaces aren't just a shopper inconvenience — they're an operational and trust problem that compounds across every seller relationship. Globally, returns cost fashion brands roughly $40 billion or more each year, and in the US alone, returns totaled about $890 billion in 2024, with the online return rate running near 19-20% according to the National Retail Federation. Marketplaces sit at the sharp end of that trend because they aggregate the fit risk of every seller they host.
There's also a trust dimension that's specific to marketplaces. When a shopper has a bad fit experience, they don't necessarily blame the individual seller — they blame the platform. A pattern of sizing surprises erodes confidence in the marketplace's search and recommendation experience broadly, not just in one listing. That's a different kind of damage than a single-brand store faces, because the platform's entire value proposition rests on shoppers trusting that browsing across sellers is safe and consistent.
| Challenge | Single-brand store | Multi-seller marketplace |
|---|---|---|
| Size chart consistency | One chart, brand-controlled | Hundreds of charts, seller-controlled |
| Measurement data quality | Usually complete | Often missing or inconsistent |
| Return attribution | Direct to brand | Diffused across sellers and platform |
| Fit trust | Built once, applies everywhere | Must be rebuilt per seller, per category |
Building a Fit Layer That Normalizes Inconsistent Data
The practical fix isn't asking every seller to standardize their size charts — that's a coordination problem no platform has fully solved. Instead, the more scalable approach is a fit layer that sits above seller-provided data and normalizes it against a shopper's actual body measurements, regardless of how inconsistent the underlying listings are.
This works by treating each seller's size chart (where one exists) as an input rather than as ground truth, and cross-referencing it against garment-level measurements — chest, waist, inseam, and so on — when they're available, or against category-level fit patterns when they're not. A shopper's saved body profile then gets matched independently against each listing, so the recommendation for Seller A's medium and Seller B's medium can legitimately differ, because they should. This is a similar problem to the one addressed by how to read a size chart correctly, except it has to run automatically, at catalog scale, across sellers who will never see or coordinate on that guidance themselves.
Virtual try-on adds a second layer on top of numeric recommendation: letting a shopper visualize how a garment sits on a body approximation, which matters even more on a marketplace where product photography quality and consistency vary wildly seller to seller. Where a size chart alone struggles with missing or unreliable inputs, a visual render gives shoppers a second, independent signal to sanity-check the numeric recommendation against — see virtual try-on vs. size recommendation for how the two approaches complement rather than replace each other.
Onboarding New Sellers Without Resetting the Fit Experience
Every new seller a marketplace onboards is, in effect, a new sizing dialect the platform has to absorb. A fit layer that requires manual size-chart mapping for every new seller doesn't scale — it becomes a bottleneck on growth. The more resilient design ingests whatever measurement data a seller provides (full chart, partial chart, or none), infers reasonable defaults from category and brand patterns where data is thin, and improves its confidence over time as return and fit-feedback data accumulates for that seller.
This also means the fit experience shouldn't visibly degrade for a shopper just because they're browsing a newer or smaller seller's catalog. Bracketing — buying multiple sizes with the intent to return the ones that don't fit — is already common, with roughly half of Gen Z shoppers reporting they do it according to the NRF, and it tends to spike precisely in the categories and sellers where sizing confidence is lowest. A marketplace that can hold fit confidence steady across its full seller base, not just its top sellers, reduces that behavior platform-wide rather than seller-by-seller. The same logic applies to how to reduce bracketing in fashion ecommerce, just applied at a scale where the platform, not a single merchandising team, owns the fix.
Where YourSizer Fits
YourSizer's sizing and virtual try-on layer is built to sit above catalog data rather than depend on every seller providing a perfectly standardized size chart. It ingests available measurements per listing, infers reasonable fit ranges where data is incomplete, and matches each recommendation independently against a shopper's saved body profile — so a marketplace can offer a consistent fit experience even when the underlying seller data is anything but consistent. For platforms managing hundreds or thousands of sellers, that normalization work is the difference between a fit tool that scales and one that requires manual size-chart cleanup on every seller signed. The try-on visualization layer runs alongside the numeric recommendation, giving shoppers a second confidence signal on listings where photography or chart data is thin — without requiring any extra work from individual sellers.
Frequently Asked Questions
How does a fit tool work when a seller hasn't provided a size chart at all?
When explicit measurements aren't available, the system falls back to category- and brand-pattern-based estimates built from comparable listings, and it improves its confidence for that seller over time as more fit and return signals accumulate. This isn't as precise as a seller-provided chart, but it still gives shoppers a meaningfully better starting point than no guidance at all, and it never requires the seller to do extra setup work before their listings can be recommended against.
Can a single fit recommendation apply consistently across sellers with different size conventions?
The recommendation itself has to be seller-specific even when the size label looks identical, because a medium from one seller and a medium from another may correspond to different actual garment measurements. What stays consistent across sellers is the underlying method — matching a shopper's body profile against whatever measurement data exists for that specific listing — not the output label, which is exactly the behavior a marketplace needs.
Does virtual try-on help more on marketplaces than on single-brand stores?
It tends to help more precisely because marketplace product data is less uniform. On a single-brand store, a shopper can lean on brand familiarity built from past purchases; on a marketplace, every seller is potentially a first-time interaction with unfamiliar sizing. A visual try-on rendering gives shoppers an independent way to judge fit even when they have no prior relationship with that particular seller's sizing conventions.
Conclusion
Marketplaces will never fully standardize how their sellers size garments, and trying to force that standardization is a losing battle against catalog growth. The more durable path is a fit layer that treats seller inconsistency as the default condition to design around, not an edge case to patch later — normalizing whatever data exists, filling gaps with pattern-based inference, and giving shoppers a visual second opinion where the numbers alone aren't enough. Platforms that get this right turn sizing chaos from a growth tax into a solved problem that scales as smoothly as the seller base does.