Order Defect Rate (ODR) is the single metric most likely to suspend an Amazon seller. Cross the threshold and your account is at immediate risk. The metric is downstream of customer feedback and claims data — by the time it shows up, the underlying issue has been brewing for days.

This guide breaks down how ODR is calculated and how to track the inputs daily.

TL;DR: Order Defect Rate is the percentage of orders with at least one of: negative feedback, A-to-z guarantee claim, or service chargeback. Threshold is 1% over rolling 60-day windows. Sustained breach risks suspension. The metric lags real-world events because feedback and claims arrive days after the order. Monitoring leading indicators — negative feedback rate, claim count, refund spikes — catches issues before ODR moves.

How ODR is calculated

ODR = Orders with a defect ÷ Total orders

An order “has a defect” if it received any of:

• Negative feedback (1- or 2-star).
• A-to-z guarantee claim.
• Service chargeback.

Same order with multiple defects counts once. Calculated over rolling 60-day windows.

Why ODR matters more than other metrics

ODR is Amazon’s primary signal of customer experience quality. The threshold is strict (1%) and the consequences are severe (Buy Box loss, listing suppression, account suspension).

Unlike Late Shipment Rate, which is operational, ODR captures customer perception. The bar to drive it down is therefore higher — you need to make customers happy, not just ship on time.

What typically drives ODR up

Quality issues

Single bad batch of product, sourcing change that affected build quality, packaging that fails to protect product.

Listing-product mismatch

Listing copy or images promise something the product does not deliver. Returns and negative feedback follow.

Late or missing deliveries

Carrier issues that cause delivery failures lead to A-to-z claims when customers cannot reach the seller.

Customer service gaps

Customers who cannot get a response from the seller often escalate to A-to-z. Slow responses are the leading cause of avoidable claims.

Refund processing delays

Customers who do not get refunded promptly file chargebacks. Often FBA-side delays you cannot control — except by responding immediately when the customer contacts you.

Where the data lives

• Customer Feedback report — negative feedback per order.
• A-to-z claim notifications — through Notifications API.
• Refunds and reimbursements — settlement reports.
• Order Returns — returns reports tied to orders.

Joining these to your orders and SKUs surfaces the patterns that drive ODR drift.

Leading indicators to track

ODR itself lags by days. The signals that move first:

• Trailing 7-day negative feedback rate. Spikes here always hit ODR.
• A-to-z claim count daily. Even one or two on a low-volume day is significant.
• Refund rate per SKU per day. Sudden spikes signal product issues.
• Customer message response time. Slower responses correlate with more claims.

How to act when ODR drifts

Identify the SKU

Most ODR spikes trace back to one or two specific SKUs. Find them.

Read the feedback and claim reasons

The text of negative feedback and claim narratives tells you the actual issue. Address it.

Pause the SKU if needed

If a sourcing or product issue is driving ODR, pausing the listing protects the broader account.

Respond to every customer

The fastest path to fewer claims is faster responses. Aim for under 24-hour response on every customer message.

The bottom line

ODR is the most expensive metric to ignore on Amazon. Tracking the leading indicators — feedback, claims, refunds — lets you intervene before ODR itself moves into warning territory.

DataDoe’s Amazon data layer joins feedback, claims, refunds and orders per SKU per marketplace so AI tools can flag ODR risk in real time, not after the warning email.