A factory that shipped the same defect thousands of times would stop the line. Online brands ship the same wrong size chart and misleading photo every week, with the reason sitting in a database nobody opens. The cheapest return is the one you prevent, and the reason code tells you how.

Most returns aren’t defects

Petersen and Kumar note that only about 5% of products are returned because they’re defective Published. The rest are gaps between what the customer expected and what arrived. The “defect” is usually in the information, so the product page is the first place to look.

Two studies show how the gap gets made. Alec Minnema and colleagues found, at an electronics and furniture retailer, that reviews more glowing than a product’s long-term average raised purchases and returns, by setting expectations the product couldn’t meet Published. Edlira Shehu, Dominik Papies and Scott Neslin found that free-shipping promotions pushed customers toward riskier items and raised returns enough that, in their simulation, the promotions lost money Published. The First Offer covers shipping as price.

A reason list that can be read

Vague codes (“didn’t like it”) can’t be acted on. A good list is specific, short and tailored by category. For apparel:

Keep “other” below 10% of reasons. If it’s higher, the list is missing something your customers keep typing.

Ordering several sizes to keep one, known as bracketing, means the customer didn’t trust your sizing. In the 2024 NRF and Happy Returns survey, 51% of Gen Z shoppers said they bracket Reported. Better size guidance reduces it more than any fee.

A case: Revolve

Revolve, the online fashion retailer, sells in a category where returns are a large share of sales. For the fourth quarter of 2024, management said the return rate fell by more than two points year over year, credited size and fit initiatives and AI, and said lower returns helped cut selling and distribution costs to 16.5% of net sales, down 129 basis points Reported. The work isn’t described in detail, but a public company told investors that return prevention showed up in the P&L.

The best return policy is a product page that tells the truth.

The weekly defect review

1. Pull the twenty products with the most returnsLast four weeks, with return rate and the top two reasons for each.
2. Flag what changedAny product whose return rate or top reason shifted since last month.
3. Assign one fix per flagged productAn owner from the table above and a date. Copy, photo, size chart, supplier call.
4. Check last month’s fixesDid the product’s return rate for that reason fall? If not, the fix was wrong.

Compare each product with itself before and after, and remember that promotions and seasons move return rates too.

Do this

• Rewrite your reason list using the table, and put the first weekly defect review on the calendar for next Monday with the people who own each fix.