Size questions in a clothing store: answer faster and reduce fit returns

Clothing size questions in an online store arrive daily and sound alike: "I usually wear M, will this dress fit?". Replying "yes, M should be fine" is a guess, because M in your store need not match M in a brand the customer knows. A good reply gives the measurements of the actual item plus one follow-up question, so the shopper can compare them with something they already own.

This guide covers the conversation, product page content and whether anyone notices your size chart. General description gaps are covered in our product page guide, return terms in the article on your return policy.

What to share instead of just "M"

A size label says which body an item was designed for, not how big it is. Two size M shirts from different makers can differ noticeably in width, because one is cut relaxed and the other slim. So reply with flat-lay garment measurements: chest width from armpit to armpit, total length and sleeve length, or waist and inside leg for trousers. Say how you measured, because a flat width and a circumference are different numbers.

EN 13402-3 contains body measurement tables used to set clothing sizes, and the DIN summary states that garment dimensions are not included. This does not show your suppliers follow the standard; it shows that a body chart and garment measurements are separate information. DIN EN 13402-3:2014.

Next comes the fit: loose or close to the body, and whether the fabric stretches. If a style runs small, say so plainly. Do not promise the size will "definitely be right"; numbers and a short comment beat reassurance.

What to ask before you recommend a size

When someone just writes "which size?", one good question saves several messages. Do not open with a request for full body measurements. Start with details people know by heart and save measuring for cases where two sizes are close.

1. Height and usual size."How tall are you and which size do you usually wear?" For trousers and dresses, height often decides the length more than the size itself.
2. Size in a familiar brand."Which size do you wear in a similar item from another brand?" A better reference than a generic M, if your team knows the brand or the shopper shares its measurements.
3. Measure a favourite item."Could you lay a shirt that fits you well flat and measure it?" Comparing garments is easier than measuring a body and matches your chart's format.
4. Preferred fit."Do you like it looser or more fitted? Will you wear a jumper underneath?" The same person may deliberately choose different sizes of the same item.

In Meetlead, agents see which page the visitor is on, so they need not ask which product is meant. If the shopper is viewing two variants, confirm which one before sending numbers. When the doubt is not about size but about choosing between products, use the three questions from the guide to pre-purchase questions in chat.

Cheat sheet: question, data needed, place on the page

This cheat sheet groups common size questions in a fictional linen clothing store. It is a template, not research: every agent answers with the same data and knows what to add to the page.

↔ Scroll the table sideways to see all columns.

The last column matters most. A chat reply closes one conversation; a row in the measurement table can spare you many of the ones that follow. Where chat helps and where page content does the job is covered in our guide to live chat for an online store.

How to turn repeat questions into page content and knowledge

To know which questions recur, count them. Tag conversations with one tag, such as "size", and note the product name internally. In Meetlead you then filter the inbox by that tag and review those conversations once a month. Our guide to tagging conversations and tickets shows how to keep the tag list short.

• A product with several size questions a month gets a garment measurement table, not just a body chart.
• Every table says how it was measured: flat or as a circumference, in centimetres.
• Photos state the model's height and the size worn.
• The fit note sits next to the size picker, not at the bottom of the page.
• The same data goes into the knowledge base used by your team and the AI assistant.

You can add measurement tables as knowledge in Meet AI: as a Help Center article, a CSV or XLSX file, or store pages. The assistant and Copilot suggestions then rely on that data; without a reliable passage, the assistant does not invent measurements but offers to pass the question to your team. Our chatbot knowledge base guide explains how to write such answers.

Meet AI is not an automatic size advisor and does not connect to your product catalogue. Sources are refreshed manually (a full website refresh at most once every seven days), so a corrected chart on your site does not change answers immediately. Keep measurements in one source and update it with the product page.

Heatmaps and recordings: does anyone see the size chart

Frequent size questions do not always mean the chart is missing. It may sit in a tab, behind a tiny link or, on mobile, below the reviews. Before rewriting anything, check the product page's click and scroll maps to see whether shoppers reach it.

• Few clicks on the chart linkThe link is hard to spot or vaguely named. Check its position relative to the size picker and its label, for example "Measurements of this style" rather than "Sizes".
• Chart below the scroll reachThe scroll map shows that on mobile most visitors stop scrolling before the measurements. Move them closer to the size picker or into an expandable panel beside it.
• A click, then a chat questionThe chart is noticed but does not answer the question. Usually the garment measurements or the method of measuring are missing.

Review maps separately for mobile and desktop. In Meetlead, a click map point leads to related session recordings when they are retained in the selected range. Watch what shoppers do after opening the chart: close it, scroll sideways or return to the photos. See also how to read heatmaps and analysing session recordings.

How to check whether the fix worked

Do not judge a change on one week of sales; too many factors affect orders. A faster measure is the number of "size" conversations for that product before and after the change, over a comparable period. If your store records return reasons, compare the share of size-related returns too, but only after a full return cycle.

Zalando states that in European online fashion return rates can reach around 50%, with size and fit behind up to half of those returns. These are market figures from a large platform with its own sizing tools; they say nothing about your store's returns or how far you can cut them. Zalando: technology and finding the right size.

Log the date of each change and what changed: added measurements, a moved link, a new fit note. Without that log, nobody will know two months later which fix worked.

Common questions

Should you recommend a size or just give measurements? Give both, with a reason, for example: "with these numbers I would go for L, because the cut is slim". Numbers alone leave doubt, and a bare "L" asks for trust the shopper does not have yet.

Can an AI assistant give size advice on its own? It can share a style's measurements from a published chart or explain how to measure. It does not know the shopper's body, so it cannot replace a size recommendation system. For "I am between sizes", a person should reply, with Copilot preparing a draft.