The cost of returns starts on the product page

A return shows up in the warehouse, so it gets treated as a warehouse problem: policies, carriers, refund times, staff. All necessary work, all of it after the fact. Meanwhile the decision that produced that return was made weeks earlier, on a product page that did not say enough.

Why things come back

The reasons for a return are few and almost always the same.

It was the wrong size. Dominant in fashion and footwear, and it covers both sizing and fit: two products in the same nominal size can fit completely differently, and the page usually does not say so.

It was not what it looked like. Colour different from the photo, material different from what was imagined, dimensions misjudged because there was no reference. This is the expectation return, and it nearly always comes from an image with no context.

It was not compatible. Dominant in technical goods and spare parts. The product was right, but not for that model, that fitting, that size. The page did not carry compatibility, so the customer guessed.

They ordered three to keep one. The return planned from the start, which is the rational answer for somebody with no way to choose beforehand. That is not a dishonest customer: it is a page that does not allow a decision.

Three of those four are solved by writing. The fourth shrinks when you solve the first.

The bill nobody adds up in full

The cost of a return is not the return shipping. It is the outbound shipping already paid, the return leg, the time of whoever opens and checks it, reconditioning where needed, the capital tied up, and the share of goods that does not come back sellable at full price.

Added up, in many categories a return costs more than the margin on a sale. Which means two sales and a return can close at a loss while looking like a good month on the dashboard.

What to put on the page, concretely

Real measurements, not the size label. A size is a manufacturer's convention; a centimetre is not. For clothing and footwear, garment measurements in every size are the single piece of information that cuts returns most.

How it fits against the standard. One line: runs small, runs generous, take a size up. It is information your support team already gives by voice twenty times a day.

A scale reference in the photos. A known object next to the product, or the model's height with the size worn. It removes an entire category of misunderstanding.

Compatibility stated, not implied. Which models it works with, and which it does not. The "not compatible with" line is worth as much as the positive one, often more.

What is in the box. Including the parts that seem obvious. Many returns start from an accessory the customer assumed was included.

One figure, as an example

At a womenswear brand that moved size advice into the conversation, before the purchase rather than after, returns fell 18% and repeat questions about size and fit fell 55%.

Read it for what it is: there, returns came from sizing, so acting on sizing reduced them. In a category where returns come from something else, the same intervention will not produce the same effect. The transferable lesson is not the number, it is the method: find out what your returns come from before deciding what to write.

Where to start

With the return reason, if you collect it. If you do not, start there: a mandatory field with five options at the moment of the request is worth more than any analysis.

Then cross the reasons with the products. You will almost always find returns are not evenly spread: a few items generate a disproportionate share, and those few nearly always have the same missing information.

Frequently asked questions

Is it not simpler to make returns harder?

It reduces returns and it reduces orders too, because a clear returns policy is one of the things people check before buying from a shop they do not know. Tightening the policy moves the cost from the warehouse to lost revenue, where you cannot see it.

How far can a return rate actually fall?

Entirely dependent on category. Where returns come from missing information, the room to improve is wide. Where they come from buying behaviour, like ordering three sizes to choose, it shrinks but does not disappear. Be wary of anyone promising a percentage without having looked at your return reasons.

Do reviews help with sizing?

A lot, and they are free. If customers write "runs small" in reviews, that information is already on your site but at the bottom of the page and mixed in with everything else. Bringing it up top, as data rather than as one user's opinion, is one of the highest-return jobs available.

Do I need a size recommender?

Not before real measurements in a table. A tool estimating size from a page that does not contain the garment's measurements is guessing with extra steps. Data first, tool afterwards if at all.

Does this apply to technical products?

Even more so, because there returns are nearly always about compatibility, and that is the most avoidable kind of return there is: it depends on data that exists, that the manufacturer publishes, and that simply was not carried into the page.