A Simple Model for What Your Return Policy Is Actually Costing You

Return rate gets tracked religiously. Return cost almost never does, at least not accurately. Most merchants can tell you their return percentage instantly and have no real number for what each of those returns actually costs once every line item is added up.

The four costs that make up a return

A single return carries four separate costs that rarely get added together in one place: the return shipping label, processing labor for inspection and restocking, the restock or write-down cost if the item can't be resold at full price, and the lost revenue if the item can't be resold at all and ends up written off entirely.

Return shipping alone typically runs several dollars per package depending on size and distance. Processing labor adds a comparable amount on top of that, covering the time someone spends opening, inspecting, and deciding what happens to the item next. Once an item has to be marked down or written off entirely, the real cost of that single return can land well above what most merchants budget for it when they only think about the shipping label.

A worked example

Take a $60 item with a 20% return rate. Out of every 100 orders, 20 come back. If each return costs $18 all-in (label, processing, and a partial markdown on resale), that's $360 in return costs against the revenue from those 100 orders, a cost most merchants have never subtracted from their reported margin because it lives in a different line of the P&L than the original sale.

Now add a merchant-controlled return fee of $6 per return, set specifically to offset processing cost rather than punish the customer for returning something that didn't work out. Across those same 20 returns, that recovers $120, cutting the net return cost from $360 to $240, a 33% reduction without changing the return rate itself at all, meaning no change in customer behavior was required to see the improvement.

Where the model changes based on your setup

The specific numbers shift depending on category and return reason, so a single blended number can hide more than it reveals. Apparel returns driven by fit issues recover better through exchanges than refunds, since an exchange keeps revenue in the store instead of sending it back out entirely and starting the sales cycle over from zero.

Damaged or defective returns cost more per unit, since resale value often drops closer to zero and the item may not be sellable again at any price. Run the model separately for your top return reasons if you track them at that level of detail, because a single blended return-cost number tends to hide which specific return type is actually driving most of the cost, making it harder to target the fix where it matters most.

Setting a fee without triggering the complaint backlash

Merchant-controlled fees carry real risk if set carelessly. Recent industry data shows that a meaningful share of retailers who added return fees saw more customer complaints as a direct result, which means the $6-per-return recovery in the example above isn't free money, it has to be sized and communicated in a way that doesn't cost more in customer goodwill than it recovers in cash.

The fee that avoids backlash tends to be transparent about what it covers, disclosed before purchase rather than surfaced as a surprise at the return step, and modest relative to item value rather than a flat number that hits a $20 item as hard as a $200 one. A fee framed as "this covers return processing" and shown at checkout, not hidden until the customer initiates a return, reads very differently to a customer than the same dollar amount appearing unexpectedly.

Where the real lever is

The most overlooked variable in this whole model is monthly software fees charged by returns platforms. A returns platform charging a flat monthly fee on top of these per-return costs adds a fixed cost that doesn't scale down in slow months and doesn't get smaller no matter how well a merchant manages return rate through better sizing guides or product descriptions.

Run this comparison directly against whatever returns platform you're using today, monthly fee included as a per-return cost by dividing it across your actual monthly return volume. A platform charging $299 a month against 50 returns is adding roughly $6 per return before a single label or hour of labor is counted, which changes the worked example above meaningfully once that fixed cost gets folded in alongside the variable ones.

ShipAid's Smart Returns removes that fixed cost entirely, with no monthly software fee, discounted return labels that lower the per-return shipping line in this model directly, and merchant-controlled fees built to recover cost without becoming a customer-trust problem that drives shoppers away before they even buy. Run your own numbers with your real return rate and label cost, then run them again without the monthly fee, and see what the model actually says about where your returns budget is going and how much of it is avoidable.