If you run a fashion online store, you see it in your orders every week: the same item, in the same colour, in two neighbouring sizes. A few days later one of them comes back. The shopper did nothing wrong. They simply chose to try at home rather than guess on the product page.
This habit is called bracketing, and it costs more than it seems. Here is why it happens, why a size chart on its own does not fix it, and what shoppers actually need to pick one size with confidence.
Why bracketing costs more than an ordinary return
A normal return starts with a problem: the shopper did not like the item, the colour looked different from the photo, the quality was disappointing. Bracketing is planned. The shopper knows, at the moment of ordering, that something is going back.
For a brand, that turns into very practical costs:
- Stock tied up. The second size is out of the warehouse for days, often in exactly the weeks when it could have filled another order.
- Double handling. Shipping, collection, inspection, refolding and restocking of an item that was never really sold.
- Less reliable data. Gross sales look healthy, net sales much less so, and buying plans by size become harder to get right.
- Items you cannot sell at full price. Some returns come back tried on, creased or without tags.
There is a quieter cost too. People order two sizes because they do not trust the information on the page. It is a sign of doubt, not convenience, and doubt sometimes ends in an abandoned basket instead.
Why the size chart is not enough
Almost every online store has a size chart. The trouble is that it asks shoppers to do a job they either cannot do or do not want to do.
To use it properly, they would need a tape measure, take chest, waist and hip measurements in the right place, compare them with the chart and then decide what to do when they fall between two sizes. Very few people go through that. Most glance at the chart, think of the size they usually wear, and order that one plus the next one up.
Then there are limits no chart can get around:
- It is generic. It often covers the whole collection, while a technical jacket, a slim jean and an oversized sweatshirt fit very differently under the same label.
- It describes the body, not the garment. It says a medium matches a certain chest measurement, but not how much ease that particular style has.
- Every brand has its own scale. The medium a shopper wears from another label is not your medium, and they know it. That is exactly why they order two.
- Non-standard sizes are left out. Shoppers in plus or petite sizes rarely get clear guidance, so they buy with more doubt than anyone else.
What shoppers need on the product page
Shoppers do not need more numbers. They need an answer: which size of this item should I buy. The information that helps them get there is short and specific.
- Guidance for the individual style, not the category. If an item runs small, say so on its own page.
- Real garment measurements, taken on the finished product: length, width, waist.
- The size and height of the person in the photo. It is a quick reference, and shoppers rely on it more than you might expect.
- An honest description of the fit: regular, close or relaxed, with a line on how it hangs.
- A size suggestion right where the decision is made, next to the size selector and the add to basket button.
That last point is the one that changes behaviour, because it turns everything above into a single choice without asking the shopper to do the maths.
How a size recommendation works
A size recommendation tool does the comparison that a chart leaves to the shopper. It takes a few details about the person, matches them against the measurements of that specific garment and points to the best size.
To work in practice, it needs three things.
- Very little input. Every extra question loses people along the way. Height, weight and age are details anyone knows without measuring themselves.
- Reliable garment data. The suggestion is only as good as the tech packs behind it. If those are incomplete, fix them first.
- One answer per item. The same person can be a medium in a shirt and a large in a padded coat. The suggestion has to change from product to product.
A good recommendation will not remove every size related return, and it should not be presented as perfect. Its job is to give the shopper enough confidence to order one size.
What to check
Before you switch on a size recommendation tool, go through this list:
- Make sure every item has real measurements for every size, including plus and petite.
- Check that the suggestion appears close to the size selector, on mobile as well as desktop.
- Count how many steps a shopper needs to get an answer. Fewer is better.
- Try two items with different fits and confirm the suggestion actually changes.
- Before launch, record size related returns and orders containing several sizes of the same item, so you have a baseline.
- Plan an A/B test: part of your traffic with the suggestion, part without, over a period long enough to cover a meaningful number of orders.
- Judge the results on returns as well as conversion. Returns arrive weeks after purchase, so allow time before drawing conclusions.
How We Wear does it with SizeFlow
SizeFlow is the We Wear size recommendation tool. The shopper enters height, weight, age and gender directly on the product page and gets the suggested size for that item before adding it to the basket. There is no body scan and no long questionnaire.
The calculation is mathematical: the algorithm compares the shopper's details with the real measurements of each garment. That makes the answer consistent, repeatable and specific to each item. It also covers technical garments as well as plus and petite sizes.
SizeFlow is a JavaScript widget for Shopify, WooCommerce, Magento, PrestaShop and custom platforms, with no backend development needed. Our advice is to launch it with an A/B test, so you measure the effect on returns and multi-size orders with your own data.
If you would like to see how it would work on your catalogue, book a 30-minute demo.
Read also: How to build a size chart · A/B testing for fashion e-commerce · Size conversion IT, EU, UK and US
Frequently asked questions
What is bracketing in fashion e-commerce?
It is the habit of ordering the same item in two or more sizes to try them at home and return the ones that do not fit. Unlike an ordinary return, it is planned at the moment of ordering.
Why is a size chart not enough to prevent size related returns?
It asks shoppers to measure themselves and compare the numbers alone, it is often generic for a whole collection, and it describes the body rather than the fit of the individual garment. Many shoppers only glance at it and order two sizes.
How does SizeFlow work?
The shopper enters height, weight, age and gender on the product page and gets the suggested size for that item before adding it to the basket. A mathematical algorithm compares these details with the real measurements of each garment, with no body scan and no long questionnaire.