At a glance
An alert that fires when your rate of new negative seller feedback jumps sharply week on week, the earliest visible symptom that something has gone wrong in fulfilment, product quality, or customer service. Negative feedback on Amazon is distinct from product reviews: it rates you, the seller, and it feeds the customer-service signals behind Account Health. A spike here often precedes an Order Defect Rate rise and, left unchecked, an account-health dip.
| What it counts | The count of new negative seller-feedback entries in the period, with a week-on-week comparison to detect a sudden jump. Negative feedback is the 1-to-2-star end of the seller-feedback scale. |
| Feedback vs reviews | Seller feedback rates the transaction and seller (shipping speed, packaging, service). Product reviews rate the product. This card tracks seller feedback, which is what affects your seller metrics. |
| Why it matters | Negative feedback is a leading indicator. A spike usually means a fulfilment slip, a defective batch, or a service breakdown, the same root causes that push up Order Defect Rate and A-to-z claims a week or two later. |
| Fulfilment scope | Mostly FBM-driven, because the seller owns delivery and service. FBA-related feedback that is genuinely about Amazon’s fulfilment can sometimes be removed, see the FAQs. |
| Removal | Some negative feedback is eligible for removal (for example, feedback that is actually a product review, contains profanity, or concerns an FBA delivery issue). Acting on a spike includes requesting removal where eligible. |
| Relationship to Account Health | Negative feedback contributes to the customer-service picture. A sustained spike feeds the metrics behind Account Health Status. |
| Time window | 7D (seven-day rolling, compared week on week) |
| Alert trigger | +50% WoW. A jump of more than 50% in new negative feedback versus the prior week flips the card and notifies owner, operations, and marketing. |
| Roles | owner, operations, marketing |
Calculation
Calculated automatically from your Amazon Seller Central data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.Worked example
A US apparel seller, mostly FBM on amazon.com. Two weekly snapshots, ending 14 Mar 26.| Week ending | New negative feedback | WoW change | Dominant theme |
|---|---|---|---|
| 07 Mar 26 | 6 | baseline | mixed, no pattern |
| 14 Mar 26 | 11 | +83% | “arrived late” and “wrong size sent” |
- An 83% jump fired the alert. The absolute numbers are small, 6 to 11, but the rate of change is what matters. A doubling of negative feedback in a week is a clear signal that something changed in the operation, even at low volume.
- The themes point straight at the root cause. “Arrived late” is a fulfilment problem; “wrong size sent” is a pick-and-pack error. Two themes, two fixable causes. Reading the feedback text turns a number into an action list.
- This is a leading indicator. Negative feedback shows up before the slower metrics. Expect Order Defect Rate and A-to-z Guarantee Claims (open) to rise over the next week or two if the cause is not fixed.
- Some of these may be removable. If any of the “arrived late” entries are FBA orders, or if any are actually product complaints rather than seller-service complaints, they may qualify for removal. Acting on the spike includes a removal review, not just a fix.
Sibling cards merchants should reference together
Negative feedback is an upstream warning. Pair it with the downstream account-health metrics it predicts:| Card | Why pair it with Negative Feedback Spike |
|---|---|
| Negative Feedback (30d) | The 30-day level behind the spike. This card flags the sudden change; that card shows the standing total. |
| Order Defect Rate | The lagging metric a feedback spike usually predicts. A spike now often becomes an ODR rise in a week or two. |
| A-to-z Guarantee Claims (open) | Buyers who leave negative feedback often file claims too. Correlated movement confirms a shared root cause. |
| Account Health Status | The composite that negative feedback ultimately feeds. A sustained spike threatens the status. |
| Late Shipment Rate | ”Arrived late” is the most common negative-feedback theme. A rising LSR explains many spikes. |
| Return Rate | Product-quality spikes show up as both negative feedback and returns. Read them together to separate service from product issues. |
Reconciling against Amazon Seller Central
Where to look in Seller Central:Performance → Feedback (the Feedback Manager). It lists every feedback entry with its star rating, date, and order, and shows your feedback rate over recent periods. Filter to negative entries in the matching date range to reconcile the count.The Feedback Manager is also where you request removal of eligible entries and where you can leave a public reply to a negative entry. Timing and reporting-lag table:
| Topic | Detail |
|---|---|
| Timezone | The Feedback Manager uses the marketplace’s local timezone. The card’s weekly comparison is aligned to consistent boundaries; an entry near midnight can fall in a different week between the two views. |
| Feedback delay | Buyers can leave feedback for a window after delivery, so a spike can reflect orders shipped a week or more earlier. The themes in the feedback text point to when the underlying problem actually occurred. |
| Removal lag | Once Amazon removes an eligible entry, it drops out of the count. There can be a short delay between your removal request and the count updating. |
| Refresh cadence | New feedback is polled on a regular cycle, so the count stays close to current; the week-on-week comparison is recomputed each cycle. |
| Reason | Direction | Why |
|---|---|---|
| Removal timing | Ours briefly higher | An entry you have just had removed can still count until Amazon finalises the removal and our next poll reflects it. |
| Window alignment | Either direction | The exact week boundary affects which entries fall in each week, which changes the computed percentage jump. |
| Negative definition | Either direction | If you filter Seller Central to a narrower star band than the card uses, the counts can differ. Confirm both are counting the same star ratings. |
| Card | Expected relationship | What causes legitimate divergence |
|---|---|---|
ebay.feedback-score | Marketplace peer. eBay feedback is the closest analogue, buyer feedback on the seller, with a different scoring model. Used for cross-marketplace service comparison. | Independent populations; the same fulfilment problem can hit both channels and spike both. |
shopify.return-rate | Shared-cause signal. A defective batch can spike Amazon feedback and Shopify returns simultaneously. | Separate channels; correlated only when product quality or fulfilment is the shared cause. |