Ecommerce returns, and the ones actually worth preventing

What a return actually costs

The refund is the visible part and the smallest part of the problem. Against a single returned order you have already spent the outbound carriage, and you are about to spend the return carriage if you fund it, the labour to receive and inspect the item, and the packaging to send it out again. Payment processing fees are rarely refunded in full. Anything that cannot go back on the shelf as new gets written down.

On a healthy-margin product that is an irritation. On a low-margin one, the true cost of a single return can exceed the margin on two completed sales, which is the calculation that turns this from an operations question into a commercial one. It is also the reason returns are worth attention before another round of traffic acquisition: you are already paying to acquire the orders that come back.

Why stock comes back

Almost every return resolves to one of three causes, and they need completely different fixes.

It did not match the expectation. The colour was different, the fabric felt cheaper, the scale was wrong. This is a product-page problem and it is the one worth fixing first, because the fix improves conversion at the same time rather than trading against it.

It did not fit. Genuine in apparel and footwear, and frequently mistaken for the first cause. The tell is whether customers reorder a different size, which is an exchange dressed as a return.

It was bought on the assumption it would go back. Bracketing: two sizes ordered, one kept. A free, frictionless returns policy actively encourages it, which is a legitimate trade rather than a mistake, but it should be a decision somebody made rather than a surprise in the numbers.

Measure yours first, and ignore the industry figure

Most writing on this opens with a headline return rate for the industry. We are not going to give you one, because the average across categories with nothing in common is not information, and we do not hold first-party returns data across a client base that would let us publish our own honestly.

What is worth having instead: your return rate by product and by category rather than as a single store-wide figure, the stated reason captured at the point of request rather than inferred later, and the rate tracked over enough months to separate a trend from a bad January. A store-wide percentage tells you almost nothing. Three products accounting for most of the volume tells you exactly where to start.

Reducing returns without cutting conversion

The trap in this whole subject is that returns are trivially easy to reduce. Make the policy strict, make the process awkward, and the line improves. Orders fall at the same time, and because the two numbers usually live in different reports, the improvement gets reported and the cost does not.

The changes that reduce ecommerce returns without costing you orders are the ones that improve both numbers at once, and there are more of them than the policy lever suggests. Better photography at the scale a customer actually needs, on a plain reference rather than a styled set. Measurements written into the page instead of a size chart behind a modal. Material and weight described honestly, including the parts that are less flattering. Reviews that mention fit surfaced near the size selector rather than at the bottom of the page. Every one of those reduces the first cause, and every one of them helps somebody decide to buy.

This is why we treat it as conversion work. It is the same discipline, applied to the half of the funnel that happens after the order confirmation.

The returns policy itself

An ecommerce returns policy does two jobs that pull against each other: it reassures somebody who has not bought yet, and it governs somebody who wants to send something back. Written only for the second, it costs you orders. Written only for the first, it costs you margin.

The practical version states the window, who pays carriage, what condition is required, and how long a refund takes, in plain terms and in a place a hesitant buyer can find before checkout rather than only in the footer. Vagueness here does not protect you. It generates support tickets and disputes, both of which cost more than the clarity would have.

The online returns process

The operational half is what most people mean by ecommerce returns management: how the request is raised, how the label is issued, how the item is tracked back, what happens on receipt, and what triggers the refund or the exchange.

Plenty of stores run the whole online returns process by email far longer than they should, and that is a defensible position while the volume is low. It stops being defensible at the point where the labour is a real cost, or where the delay between an item arriving back and a refund landing is generating the support load itself. That is the threshold to watch, rather than a particular order count.

Where exchanges are the right outcome, make them the easy path. A size exchange that is simpler than a refund keeps the revenue and the customer, and the store that buries it is choosing a refund it did not have to give.

Common questions