Fashion has a return problem no other e-commerce category has. Apparel return rates sit far above the e-commerce average, and most of those returns come down to one thing: fit. "Too small", "too big", "didn't look like that on me". Each of those is a question the customer had before buying and could not get answered.
That makes sizing a double tax on a fashion store. A shopper who cannot get a fit answer either leaves (a lost sale) or buys two sizes to try (a return you pay for). This guide is about closing that gap with AI chat that answers fit, fabric and styling questions from your WooCommerce catalog before the order is placed.
Why sizing hurts twice
| Customer behaviour | What it costs you |
|---|---|
| Unsure on size → leaves | Lost conversion, no signal why |
| Unsure on size → buys two sizes to try | Guaranteed return + reverse shipping |
| Unsure on fabric/fit → buys → disappointed | Return + a refund + maybe a bad review |
All three rows have the same cause: a question that went unanswered at the moment of decision. Solve that and you lift conversion and cut returns at the same time. That is rare in e-commerce, where most changes improve one number at the cost of another.
What AI chat does with a fit question
Customer: is the linen blazer true to size? i'm usually a M but between sizes on top
AI: The Linen Blazer runs slightly relaxed, so most customers stay with their usual size. Since you're between sizes and it's a layering piece, an M will give you room to wear it open over a shirt; size down to S only if you want a fitted look. The fabric is 100% linen. It is breathable but it does crease, which is part of the look. Want me to show it next to the structured blazer, which holds shape better?
Customer: show me both
AI: [compares the two] …
In one exchange the assistant answered the fit question, described the fabric and suggested an alternative, which is what a good in-store associate does. It takes the answers from your product data and size guides. Because the search is semantic, it understands phrases like "between sizes" and "true to size" and picks up what the customer wants the piece for.
This is also cart recovery
Sizing doubt is one of the top reasons a full cart never checks out. Answering it during the session is the same in-the-moment recovery we describe in recovering abandoned carts with AI chat. Confidence before purchase is the best version of it: a sale made with the right size in the bag stays sold.
Setting it up on WooCommerce
- Sync your catalog and your size-guide content so the assistant knows the fit, fabric and care of each item. Setup takes about 15 minutes.
- Add custom instructions with your brand's fit notes, for example: "Our outerwear runs relaxed; recommend usual size unless the customer wants a fitted look. Always mention fabric and care for linen and wool."
- Enable human handoff for styling-heavy or VIP conversations you would rather a person took.
The bigger picture
For apparel, the return rate is the margin conversation, and most of it is preventable because most of it is a fit question nobody answered. ChatAxon answers those questions at the moment they would otherwise cost you a sale or a return. Hesitant shoppers get their answer and buy, and fewer of those orders come back.
"Does this run true to size?" is one of the most common questions a fashion store gets in Instagram DMs. ChatAxon answers it there too. See how it works.
See the full breakdown for apparel stores on the fashion & apparel use-case page →, or book a demo. You can also try Starter, Pro or Scale free for 14 days.