Auditing the True Cost of Free Returns

November 8, 2026

Free returns cost the outbound shipping, the return shipping, the inspection and restocking labor, the markdown on the resold item, and the fraud that rides along with them. Add those up per order and most stores find that free returns cost 15 to 30 percent of the order value, which is more than the margin on the sale.

The number everyone quotes, and the number nobody does

Retailers made returns easy on purpose. Free returns, long windows, and refunds issued at the carrier's scan help win orders, and they also remove the point where anyone inspects the item before the refund is paid. The result is a returns bill that most stores never fully count, because the costs are spread across five budgets and no one owns the total.

The headline number is the fraud. Fraudulent returns cost retailers about $101 billion in 2024, with 15.14 percent of all returns deemed fraudulent, and total returns reached $890 billion, or 16.9 percent of all retail sales. Those are the numbers from the National Retail Federation's returns report, and they are the ones that get quoted. But fraud is not the biggest cost. The biggest cost is the legitimate return, processed badly, at scale.

The audit below adds up the five costs of a free return, and the total surprises most merchants, because the shipping alone is often more than the margin on the item that came back.

The five costs, added up

First, outbound shipping. The store paid to ship the item to the customer, and that cost is gone whether the item comes back or not. On a free-shipping order, the outbound cost is usually $5 to $12, and it is the first line of the loss.

Second, return shipping. A prepaid return label costs the store another $5 to $12, and the carrier charges it whether the item is resellable or not. Third, inspection and restocking labor. A returns desk worker needs five to ten minutes to open, inspect, photograph, grade, and restock a return, and at a loaded labor cost of $20 an hour, that is $2 to $4 per item, before the system time and the exception handling.

Fourth, the markdown. A returned item that cannot be resold as new goes to a resale channel, an outlet, or a liquidator, and the recovery is often 30 to 50 percent of the original price. On a $100 item, that is a $50 to $70 loss, and it is the largest single line in the audit. Fifth, the fraud and abuse that rides along: wardrobing, bracketing, empty-box claims, and tag switching, which the wardrobing detection guide covers in detail.

Add them up for a typical $100 order: $8 outbound, $8 return shipping, $3 labor, $50 markdown, and a share of the fraud cost. The total is $70 or more, on a sale that made perhaps $30 of margin. Free returns, done badly, cost more than the sale was worth.

How to run the audit on your own store

Start with the data you already have. Pull the last 90 days of returns and calculate five numbers: the return rate by category, the average cost per return including shipping both ways, the inspection time per return, the resale recovery rate, and the fraud rate by condition code. Most stores can get the first two from their carrier and their order system. The last three need the returns desk to record condition codes consistently, and that is the part most stores skip.

Then segment by category. Occasion wear, footwear, and high-value items will have a higher cost per return than basics, because the markdown is larger and the inspection takes longer. A category with a 30 percent return rate and a $70 cost per return is a different problem from a category with a 10 percent return rate and a $20 cost, and the fix is different too.

Finally, compare the cost to the margin. A store with a 40 percent gross margin can afford a higher return cost than a store with a 20 percent margin, and the audit should end with a number: the maximum return cost per order the store can afford, and the categories that exceed it. Those categories are where the policy changes go: shorter windows, restocking fees, inspection before refund, or a paid return label.

What to do with the answer

The answer is rarely to end free returns. Free returns win orders, and a store that charges for returns will lose sales to one that does not. The answer is to make free returns cheaper: inspect before refund, so the worn items do not get a full refund; shorten the window on the categories where the cost is highest; and publish better fit information, so the customer buys one size instead of three.

The other answer is to fix the product page. A return that happens because the size chart was wrong, the photo was misleading, or the description promised something the product does not do is a return the store caused, and the cheapest fix is to fix the page. The denim returns audit shows how the same logic works on a single category, and the return benchmarks guide gives the category numbers to compare against.

Frequently asked questions

How much do free returns actually cost?

Add up outbound shipping, return shipping, inspection and restocking labor, the markdown on the resold item, and the fraud that rides along. On a typical $100 order, the total is often $70 or more, which is more than the margin on the sale. Fraudulent returns alone cost retailers about $101 billion in 2024.

What is the biggest cost in a free return?

The markdown. A returned item that cannot be resold as new goes to a resale channel or a liquidator, and the recovery is often 30 to 50 percent of the original price. On a $100 item, that is a $50 to $70 loss, and it is the largest single line in the audit.

Should my store stop offering free returns?

Usually not. Free returns win orders, and a store that charges for them will lose sales to one that does not. The better answer is to make free returns cheaper: inspect before refund, shorten the window on the highest-cost categories, and publish better fit information so customers buy one size instead of three.

How do I audit my return costs?

Pull the last 90 days of returns and calculate the return rate by category, the average cost per return including shipping both ways, the inspection time per return, the resale recovery rate, and the fraud rate by condition code. Then compare the cost per return to the margin per order, and target the categories that exceed it.