Order Defect Rate

Order Defect Rate rolls up the most serious negative outcomes an order can have. Three things typically count as a defect: negative customer feedback (a one- or two-star seller rating), an A-to-z Guarantee claim (where a buyer escalates a problem to Amazon), and a service chargeback (where the buyer disputes the charge with their card issuer for a service-related reason). If any of these attaches to an order, that order is counted as defective.

The rate is then the number of orders with at least one defect divided by the total number of orders in the measurement window, expressed as a percentage. Because it is a ratio, low-volume sellers feel each defect more sharply — a single bad order represents a larger share of a small order count.

ODR is measured over a rolling period rather than a single day, so it reflects recent performance rather than a one-off event. That smoothing cuts both ways: a good run gradually improves it, but a cluster of problems can keep it elevated for weeks even after you fix the underlying issue.

ODR is one of the headline metrics Amazon uses to judge whether a seller is providing an acceptable customer experience. Amazon publishes a widely cited target of keeping ODR below 1%. Crossing that threshold can trigger a performance review, warnings, or in serious cases suspension of selling privileges — which is why ODR is watched so closely.

Because the consequences are severe, treating ODR as an early-warning signal rather than a problem to fix after the fact is the safer approach. A rising ODR usually points to an underlying issue — a product quality problem, a fulfilment bottleneck, or a misleading listing — that is generating unhappy customers before it shows up as an account threat.

The practical levers for ODR are consistent. Ship on time and reliably, because late or failed deliveries are a leading cause of negative feedback and claims. Keep listings accurate, so customers receive what the listing promised and have no reason to complain. Respond quickly and helpfully to buyer messages, since many disputes can be resolved before they escalate into a defect.

For multichannel sellers, the same discipline applies on every marketplace — most operate an equivalent seller-performance regime, even if the exact metric names differ. Accurate cross-channel inventory prevents overselling and the cancellations and late shipments that follow, which protects defect-style metrics everywhere you sell at once.