When a retailer sells products from dozens of different vendors, "medium" doesn't necessarily mean the same thing from one product to the next. Each vendor has its own blocks, grading, and fit standards, so a single published size chart can only capture so much of that variation. Shoppers end up trying to figure out whether the chart actually applies to the specific product they're looking at.
For this retailer, we took a different approach. We didn't change the garments, vendor specifications, or grading rules. We started with the size charts they already had and used real customer measurement data to make the sizing information more accurate over time. In a controlled A/B test against the retailer's original static size chart, dynamic sizing drove a 3.4% site-wide conversion lift and a 10.8% increase in average order value.
This retailer carries a broad assortment, with products sourced from many different vendors. That's a common and efficient way to build a large catalog, but it also creates a sizing challenge. Every vendor has its own approach to fit. One vendor's medium may be based on one block and one set of ease assumptions, while another vendor's medium is based on something completely different.
When all of those products are presented on the same site, however, shoppers often have one size chart to reference. That chart may be a reasonable average across the assortment, but an average doesn't necessarily tell a shopper whether a specific product will fit them. Over time, shoppers learn this. They may start guessing between sizes, hesitate before purchasing, or decide not to buy. And without customer measurement data, it's difficult for the retailer to know exactly where the sizing is missing the mark.
One of the most important parts of this test is that the products themselves didn't change. We didn't ask the retailer to re-cut garments or standardize grading across vendors, and we didn't change vendor specifications or run a new round of fit sessions. Any of those could be useful initiatives, but they're also significant undertakings across a large multi-vendor assortment.
Instead, we treated the problem as a sizing data problem. The retailer's existing size charts became the starting point, and the garments stayed the same. What changed was the quality of the information used to describe who those sizes actually fit.
We deployed TrueToForm on the product page, where shoppers can either enter measurements or complete a quick body scan. That gives them a personalized fit prediction for the product they're viewing instead of asking them to interpret a generic size chart.
But the fit prediction is only part of the value. With every order, the retailer can also collect real customer measurement data tied to the size purchased. When return information is available, that data can be further separated by whether the order was kept or returned. Over time, this creates a much clearer picture of who is actually buying and keeping each size.
Dynamic sizing uses that information to improve the sizing guidance. Rather than relying entirely on a static average, the sizing data can increasingly reflect the real customers associated with each size. As more data accumulates, it can also be segmented more specifically, starting with broader categories and narrowing toward individual product groups where there is enough volume to support it.
That's particularly useful for a multi-vendor catalog. Instead of forcing products from different vendors into the same average, the sizing information can reflect differences across the assortment based on the customers actually buying those products.
The retailer ran dynamic sizing against its original static size chart in a controlled A/B test. The test showed a 3.4% site-wide conversion lift and a 10.8% increase in average order value.
The fact that these were site-wide results is important. This wasn't just a lift among shoppers who interacted with the fit experience. The test compared the overall performance of the two groups, with the same products, vendors, pricing, and assortment. The difference was the sizing information available to shoppers and the underlying data used to inform it.
The AOV increase is also something we see when shoppers have an easier path to confident purchasing. Getting an immediate size result, rather than having to dig through a size chart for each product, can make it easier to shop across categories and add more items to an order. We also see higher confidence in purchasing higher-priced items, where fit uncertainty can be a bigger barrier to purchase.
When sizing is inconsistent across vendors, the natural solution is to try to fix the problem upstream. Standardizing vendor grading, tightening specifications, and improving fit processes can all be valuable. But those changes take time, and they're often difficult to implement across a large vendor network.
Dynamic sizing offers another option: improve the sizing data without changing the products. Instead of asking every vendor to fit into the same standard, retailers can use their own customer data to understand what the existing sizes actually look like on the people buying and keeping them.
Every order adds another data point. Over time, that makes the sizing information more representative of the actual customer base. And because the data is connected to products and order outcomes, retailers can begin to understand where a broad size chart is accurate and where it isn't.
For shoppers, the experience is simpler too. Instead of asking them to figure out whether an average size chart applies to them, they can see how a specific product is expected to fit their body.
The products didn't change in this test. The information about the products did.
No. The products, vendors, grading, and specifications stayed the same. The test started with the retailer's existing size charts, and the change was in how the sizing data was informed by real customer measurements.
The test compared dynamic sizing, informed by real customer measurement data, against the retailer's original static size chart. The assortment, products, vendors, and pricing were the same on both sides.
Customer measurement data can be segmented as enough data accumulates, starting with broader categories and becoming more specific where there is sufficient volume. This allows sizing guidance to reflect differences across the assortment rather than treating every product as though it follows the same sizing standard.
No. Dynamic sizing can start with the retailer's existing size charts and internal data. As more customer measurements and order outcomes are collected, the sizing guidance can become increasingly representative of the retailer's actual customers.
-
Want to see what your customers’ measurements look like across your current size charts? TrueToForm can help you understand where your sizing data is accurate, where it may be missing the mark, and how dynamic sizing can improve it over time. Get in touch with our team here.
Subscribe to our mailing list for news and updates
Access on mobile at web.truetoform.fit (no install required).
Also available on the App Store.