Refund Rate
Refund rate is the share of orders returned for a refund — a direct margin hit and one of the cleanest signals that your product pages are over-promising.
Refund Rate
The percentage of orders refunded over a given period — a margin killer and a leading indicator of PDP-product mismatch.
Refund rate is the share of fulfilled orders that end in a refund, calculated over a fixed window (usually 30, 60, or 90 days from order date). It captures both full returns and partial refunds for damage, sizing, or buyer's-remorse reasons.
Most teams treat it as a finance metric, but it's just as useful upstream. A rising refund rate almost always points to a gap between what the product page promises and what arrives in the box — wrong sizing charts, optimistic photography, missing material detail. That makes it one of the most underused leading indicators in the broader set of ecommerce metrics.
On a 60% gross-margin product, a single refund wipes out roughly 1.6 net sales. That's why a two-point shift in refund rate hits contribution margin harder than a two-point shift in conversion rate — the loss includes reverse logistics, restocking, and often unsellable inventory.
Track it per SKU, per collection, and per acquisition channel. Aggregate store-level refund rate hides the 3-4 SKUs doing the damage, and paid social traffic tends to refund at a higher rate than email or organic because expectations are set by a 6-second ad rather than a considered visit.
Refund Rate = (Refunded Orders / Total Orders) × 100
Refunded Orders
Refunded orders
Orders refunded in full or in part within the measurement window.
Total Orders
Total orders
All fulfilled orders placed in the same window (use order date, not refund date, for the denominator).
A Shopify apparel store ships 4,200 orders in Q1 and processes 546 refunds against those orders by the end of Q2.
Refunded orders: 546
Total orders: 4200
→ 13.0%
Right on the apparel median. Not a fire, but a five-point gap from best-in-class — worth a sizing-chart and size-recommender review before the next paid push.
Pick a cohort window and stick to it. Mixing a 30-day refund window with a 90-day denominator inflates the rate as more refunds trickle in; comparing months only works if both are mature. For monthly reporting, lag the metric by 30-45 days so refunds have time to land.
Typical refund rate ranges by vertical
| Vertical | Excellent | Median | Needs work |
|---|---|---|---|
| Apparel & footwear | < 8% | 10-15% | > 18% |
| Beauty & skincare | < 2% | 3-5% | > 7% |
| Consumer electronics | < 5% | 6-9% | > 12% |
| Home & furniture | < 4% | 5-8% | > 10% |
| Food & supplements | < 1% | 1-3% | > 5% |
| Jewelry & accessories | < 3% | 4-7% | > 9% |
Use the benchmark as a sanity check, not a target. The fastest wins usually come from looking at the worst-offender SKUs: if 12 products generate 40% of your refunds, fixing their PDPs (better fit photography, video, size recommender, clearer material copy) moves the store-level number more than any sitewide change.
Refund rate FAQ
It depends entirely on vertical. Apparel runs 10-15% as a healthy median; beauty and supplements should sit under 5%; electronics around 6-9%. Compare against your category, not a sitewide average.
Return rate counts every item shipped back, including exchanges and store credit. Refund rate only counts the orders where money actually left your account. Refund rate is the stricter margin metric; return rate is the better operations metric.
Divide refunded orders by total orders in the same cohort window, then multiply by 100. Anchor both numbers to order date, not refund date, so the denominator doesn't shift while you're measuring.
Refunds are the clearest signal that the product page set the wrong expectation. Mismatched sizing, optimistic photography, or missing material detail all show up as refunds. Treating it as a CRO leading indicator lets you fix the PDP before the next paid push amplifies the leak.
Shopify's analytics show refunds against the order date by default, which is what you want for cohort accuracy. Just confirm whether partial refunds (shipping-only, damage adjustments) are included in your dashboard before benchmarking.
Start with sizing tools, fit photography on real bodies, and unboxing video on top-refunded SKUs. These typically reduce refunds without dragging conversion. Aggressive 'final sale' language or restocking fees lower refunds but usually cost more in lost conversion than they save.
Yes. Paid social traffic typically refunds 1.5-2x higher than email or organic because expectations are set by short-form ads. Segmenting by channel exposes whether a creative or audience is dragging your margin down.
30 days covers most apparel and beauty refunds; 60-90 days is safer for electronics and furniture where buyers take longer to decide. Pick one window per category and lag your reporting by the window length so refunds have time to land.
On a 60% gross-margin product, every refund erases roughly 1.6 net sales (the refunded order plus the reverse logistics and unsellable inventory cost). A two-point refund-rate shift moves contribution margin harder than the same shift in conversion rate.
Yes — test sizing-chart variants, fit-finder widgets, and PDP photography against refund rate as the primary KPI, not just conversion. The test takes longer to call because refunds trail orders by 30-60 days, but the margin impact justifies the wait.
Get an AI expert review of your site
Paste your URL — Metricuno's AI runs the same heuristic checks a senior CRO consultant would, scoring your page and prioritising the fixes that'll move conversion fastest.