Return rate
Return rate is the percentage of units sold that customers send back over a period. It is calculated as units returned divided by units sold, multiplied by 100. Ecommerce operators track it because every return claws back the revenue while the shipping, handling and payment costs of the sale stay spent.
| Units returned | Units sent back for a refund in the period, matched to the sales cohort they came from |
| Units sold | Units shipped in the same period. Keep both sides in units, or both in orders - never mix the two |
Worked example
A store sells 2,000 units in a month at an $80 average selling price (example numbers, flagged as such).
The 8% return rate does not just remove $12,800 of revenue. It leaves $2,400 of already-spent fulfilment cost on the books, before any write-down on returned units that cannot be resold as new.
What is a good return rate?
There is no universal benchmark, and any single number quoted without context ignores what actually drives returns. A sustainable return rate depends on your category (fit-risk apparel behaves nothing like consumables), your price point, how much sizing and variant complexity you carry, whether exchanges are counted, and above all your margin structure - a high-margin product can absorb a return level that would sink a thin-margin one.
The useful question is what a return costs you. Our free returns impact calculator computes the real margin cost of your own return rate from your own numbers.
Return rate vs related metrics
| Metric | What it measures | How it differs |
|---|---|---|
| Sell-through rate | Share of available stock sold in a window | Outbound velocity. Return rate measures what comes back after the sale. |
| Retention rate | Share of customers who buy again | Customers staying, not product coming back. A high return rate usually drags retention down with it. |
| Gross margin | Revenue minus cost of goods | Returns cut realised gross margin: the refunded revenue leaves while most costs stay. |
| Contribution margin | What a sale keeps after all variable costs | Return costs are variable costs, so this is where they land. Return rate is a margin metric, not a service metric. |
Common mistakes
- Dividing this month's returns by this month's sales. Returns lag sales by days or weeks. Match returns to the sales cohort that produced them, or a growing store will understate its true rate.
- Counting only the refund. Outbound shipping, pick and pack, payment fees and return processing are all still spent. The refund is the visible cost, not the whole cost.
- Blending the whole catalogue. Return rate is a per-SKU problem. One bad size curve or a misleading product page can hide inside a comfortable average.
- Treating exchanges as refunds. An exchange keeps the revenue and the customer; a refund keeps neither. Track them separately before judging a product.
- Mixing units, orders and dollars. Pick one basis and hold it. A unit return rate and a dollar return rate on the same store can differ materially when expensive items return more often.
Frequently asked questions
In Blufire, S7 Returns runs this per SKU: returns economics computed on your own reconciled numbers, plus refund-risk scoring that flags the products most likely to come back before they erode margin.
Updated July 2026