Return rate is the number of returns divided by the number of orders (or units sold) over a period, expressed as a percentage. If you sold 1,000 units and 60 were returned, your return rate is 6 percent. You can pull returns from Seller Central's returns reports and compare them to your order volume for the same window. Track it per SKU, because one problem product can hide inside a healthy overall average.
What is a normal return rate on Amazon?
It depends heavily on category. Apparel and shoes see high return rates because of fit; consumables and simple hard goods see much lower ones. Rather than chase a universal number, compare each SKU against your own history and its category norm. A rate that is rising over time, or well above similar products, is the signal that matters more than any single benchmark figure.
T. Brian Jones is co-founder and CTO of Inventory Hero. He leads the engineering behind its Amazon data pipeline, demand forecasting, and the AI platform that lets sellers talk to their live inventory, sales, and supplier data in plain language.
Attack the reasons customers give. Improve listing accuracy so the product matches expectations, add clear sizing or specification guidance, fix quality issues the reviews point to, and improve packaging so items arrive undamaged. Read the return-reason data and negative reviews to find the top cause per SKU, then fix that. Most return-rate improvement comes from removing the specific mismatch driving returns, not generic tweaks.
Read article
Your Amazon return rate is returns divided by orders over a period, expressed as a percentage. The short version: what counts as normal depends heavily on your category, the levers that move it are listing accuracy, sizing, quality, and packaging, and returns cost you well beyond the refund. Track it per SKU so one bad product does not hide in the average. Below is how to measure it, what is normal, and how to bring it down.
Return rate = returns / orders (or units sold) x 100
Returns come from the FBA Customer Returns report (Reports > Fulfillment > Returns > FBA Customer Returns); orders from your sales data over the same window. There is no ready-made per-SKU return-rate dashboard, so pull the returns report, pivot by ASIN, and divide by that SKU's units sold yourself.
If you sold 1,000 units and 60 came back, your return rate is 6 percent.
The one rule that makes it useful: compute it per SKU, not just blended. A single high-return product drags your economics and your account health, and it is invisible in the overall number. Trend each SKU over time, because a rate that is climbing is a problem forming.
There is no single benchmark, because category dominates:
Apparel, shoes, and anything with fit run high return rates; customers order variations and send back what does not fit.
Consumables, simple hard goods, and commodities run much lower.
Electronics and complex products sit in between, driven by expectation mismatches and defects.
So the useful comparison is not a universal figure but two references: your own SKU's history (is it rising?) and its category norm (is it an outlier?). As a directional tier (operator rules of thumb, not published figures): apparel and shoes commonly run into the high teens or higher; electronics and complex products sit in a middle band; consumables and simple hard goods are usually low single digits. Judge each SKU against its tier, not the average. To dig into the why, pull the return-reason data at Reports > Fulfillment > Returns, filter by ASIN and reason code, and read what customers actually say. A stable rate in line with your category is fine; a rising one, or one well above similar products, is the signal to act.
The refund is only the visible part of a return's cost:
Return processing fees. In affected categories (notably apparel and shoes), Amazon charges a returns processing fee, often a few dollars per returned unit, on top of the refund.
Unsellable units. Many returns come back unfulfillable and must be reimbursed, or removed or disposed of, so you lose the unit, not just the sale.
Reduced margin. Every return erodes the net margin on the sales that stuck, since your costs to acquire and fulfill are not recovered.
Account health. High returns and the complaints that come with them can pressure your metrics.
Work it roughly. On a 30-dollar product with an 8-dollar margin, a sale that gets returned does not just erase the 8 dollars of profit; you refund the 30, eat return processing, and often cannot resell the unit, so you are out the landed cost plus fees on that unit. One return can wipe out the profit from three or four good sales of the same SKU. Put together, a high-return SKU can be unprofitable even when its headline margin looks fine, which is why return rate belongs in your real per-unit economics.
Return-rate improvement comes from removing the specific mismatch driving returns, which the data will tell you:
Listing accuracy. If the product does not match the images, title, or bullets, customers return it as "not as described." Make the listing tell the exact truth.
Sizing and specification guidance. For fit- or spec-sensitive products, a clear size chart or spec table prevents the wrong-choice return.
Product quality. If reviews and return reasons cite defects, the fix is upstream quality control, not copy.
Packaging. Items that arrive damaged get returned; better protective packaging removes that whole bucket.
Read the return-reason breakdown and the negative reviews per SKU, find the top cause, and fix that one thing. Generic tweaks do little; targeting the actual reason does.
Review return reasons monthly for your highest-volume and highest-return SKUs.
Fix the top reason per problem SKU, then watch the rate over the next cycle.
Factor returns into reorder decisions and your restock quantities, because a high-return product ties up units in the unsellable pile and distorts your true sell-through.
Return rate is a feedback loop: it tells you where the product or listing is failing customers, and lowering it lifts both margin and account health at once.
Amazon return rate is returns divided by orders, best tracked per SKU against your own trend and your category norm rather than a universal benchmark. It costs far more than the refund once you count unsellable units and lost margin, and it comes down when you fix the specific reason customers return each product: accuracy, sizing, quality, or packaging. Treat it as profit data, not just a service metric. For the unsellable side, see FBA reimbursements; for the wider economics, net profit margin.