At a glance
Order Defect Rate (ODR) is the percentage of your orders that ended in a defect: a negative seller-feedback, an A-to-z Guarantee claim, or a service chargeback. It is the single most important account-health metric on Amazon, because Amazon’s published expectation is that ODR stays under 1%. Breach it and your account is at risk of review and suspension. This is not a vanity metric. It is the line between trading and being deactivated, so it should be watched harder than almost any other card.
| What it measures | The share of orders in the window that resulted in at least one defect. Amazon’s three defect types are negative feedback (1-star or 2-star), A-to-z Guarantee claims, and service chargebacks. Expressed as a percentage of orders. |
| The threshold | Amazon’s stated target is under 1%. Sustained ODR above 1% is the classic trigger for account review and possible suspension. This card warns earlier, at 0.7%, to give you runway. |
| Why it is the most important card | Most cards affect revenue. ODR affects whether you have an account at all. A suspension stops all Amazon sales overnight, so a rising ODR is a higher-priority event than almost any sales dip. |
| What feeds it | The same operational problems behind Negative Feedback (30d), A-to-z Guarantee Claims (open), and Late Shipment Rate all roll up here. |
| FBA vs FBM | FBM sellers carry the most ODR risk because they own dispatch and packaging. For FBA orders, fulfilment defects are Amazon’s responsibility, and many are excluded or removable from your ODR. |
| Small-denominator effect | At low order volumes, a single defect can swing ODR sharply. A new seller with 40 orders sees one defect register as 2.5%, even though the underlying service is fine. Read alongside order count. |
| Time window | 30D vsP (trailing 30 days vs the prior 30 days, the window Amazon weights most) |
| Alert trigger | >0.7% warns; >1% is the suspension-risk line. Driven by the account-health detection layer at its highest priority. |
| Roles | owner, operations |
Calculation
Calculated automatically from your Amazon Seller Central data. ODR is the share of orders in the window that ended in a defect (negative feedback, A-to-z claim, or service chargeback). See the At a glance summary above and the worked example below.Worked example
A UK FBM kitchenware seller on amazon.co.uk. Period: trailing 30 days to 14 Mar 26.| Defect type | Count (this 30 days) |
|---|---|
| Negative feedback (1 or 2 star) | 5 |
| A-to-z Guarantee claims | 2 |
| Service chargebacks | 1 |
| Defective orders (deduped to order level) | 7 |
| Total orders in window | 850 |
| Order Defect Rate (this card) | 0.82% |
- 0.82% is in the warning band, not yet the danger line. It is above the 0.7% warn threshold but below the 1% suspension line. That is exactly the window to act in: there is still margin before Amazon takes action, but the trend needs reversing now.
- Defects dedupe to the order. Seven defective orders, not eight, because one order had both a negative feedback and a chargeback. ODR is per defective order, not per defect event.
- The denominator matters enormously. With 850 orders, each defect is worth about 0.12 percentage points. At 85 orders, each would be worth 1.2 points, so a single bad order would breach the line. Always read ODR next to order volume.
- Trace each defect to a root cause. The five negatives and two A-to-z claims almost certainly share a driver (a dispatch delay run, a packaging fault). Fixing the operational cause is what brings ODR down; you cannot argue the rate down.
- This is the highest-priority alert in the system. A sales card dropping costs money this month. ODR breaching 1% can cost the entire account. When this card warns, it jumps the queue.
>0.7% warning, and Vortex IQ Nerve Centre surfaces it as a high-priority account-health alert with the defect breakdown, so operations can fix the root cause before ODR crosses the 1% suspension line.
Sibling cards merchants should reference together
ODR is fed by other defect signals. These cards show you the components and the context:| Card | Why pair it with Order Defect Rate |
|---|---|
| Negative Feedback (30d) | One of the three defect inputs. A feedback spike often shows up here as a rising ODR. |
| A-to-z Guarantee Claims (open) | The most serious defect input. An open A-to-z claim is both an ODR hit and an immediate cash and reputation risk. |
| Late Shipment Rate | The most common upstream cause. Late dispatch drives negatives and claims, which drive ODR. |
| Account Health Status | The overall standing ODR is the headline contributor to. |
| Account Health Creep | The early-drift signal. Catching the creep lets you act before ODR itself moves. |
| Pre-Fulfilment Cancel Rate | A related account-health metric; oversells and cancels often share a root cause with defects. |
Reconciling against Amazon Seller Central
Where to look in Seller Central: The closest native view is:Seller Central → Performance → Account Health. The Order Defect Rate tile is shown prominently with its three components (negative feedback, A-to-z claims, chargebacks) broken out and the 1% target marked.The Account Health page is the authoritative source. This card mirrors that ODR figure and breakdown so you can watch it in Nerve Centre alongside every other channel. Timing and reporting-lag table:
| Topic | Detail |
|---|---|
| Rolling window | Amazon evaluates ODR over a rolling window (commonly framed around 60 days for enforcement, with a shorter view shown for trend). A defect rolls off as it ages out of the window, so ODR can fall without any new action. |
| Defect recognition lag | Negative feedback, claims, and chargebacks each register on their own timeline after the order. ODR updates as each defect is recognised, so today’s figure can rise as older orders attract late defects. |
| Removable defects | Some defects (FBA-caused, policy-breaching feedback) can be removed, which lowers ODR once processed. The card follows Amazon’s current ODR after removals. |
| Refresh cadence | Account Health updates close to real time. Vortex IQ reads it on each sync, so the card may differ by the sync interval. |
| Reason | Direction | Why |
|---|---|---|
| Window definition | Possible difference | If the card’s display window and Amazon’s enforcement window differ in length, the percentages differ. Treat the Account Health page as authoritative for enforcement. |
| Removal timing | Ours can lag briefly | A defect removed on Amazon drops ODR; the card reflects it on the next sync. |
| Denominator | Possible small gap | Exactly which orders count toward the denominator (and which are excluded) follows Amazon’s rules; small definitional differences move a low-volume ODR noticeably. |
| Sync interval | Ours can lag | The card reflects the last sync; Account Health is near real time. |
| Card | Expected relationship | What causes legitimate divergence |
|---|---|---|
shopify.fulfilment_issues | Shared operational root, separate metrics. A fulfilment problem in your own warehouse can drive defects on both channels, but ODR is an Amazon-specific account-health concept with no Shopify equivalent. | A carrier or 3PL failure can raise both Amazon defects and DTC complaints at once, which is a correlation to investigate, not a discrepancy. |
ebay.seller-standards-level | Peer concept on another marketplace. eBay’s seller-standards and defect-rate system is the conceptual cousin of ODR. | The thresholds, windows, and defect definitions differ between marketplaces, so the two numbers are not interchangeable. |