One SKU. 13.3% Return Rate. Six Things to Check Before Deciding Anything.
When one child SKU showed a 13.3% return rate and a customer said "the bottle is leaking," the question wasn't what to do. It was what else was true first.
A high return rate on a single SKU combined with a customer complaint feels like an open-and-shut case: defective product, pull the listing. That's almost always premature. A single review does not explain eight returns. Eight returns across 60 units is an outlier that warrants investigation — but the investigation must separate physical defects from listing expectations, compatibility confusion, and fulfilment damage. Jumping to a solution before checking the other five factors usually wastes money or fixes the wrong problem.
In our returns diagnostic across a UAE marketplace account, one child SKU immediately caught everyone's eye: 60 units sold, 8 returned, a 13.3% return rate. More than double the account average.
Then someone found the review: "The bottle is leaking. The whole purpose is defeated." Case closed, right? Manufacturing defect. Contact the supplier, stop selling the SKU.
Not so fast. Here is why structured investigation matters more than quick reactions.
The eight returns didn't all say "leaking"
When we pulled the actual return reason codes from Amazon Seller Central for those eight units, a different picture emerged:
| Recorded reason | Count | % of returns | What this actually suggests |
|---|---|---|---|
| Defective | 2 | 25% | Consistent with the leakage review — points toward a physical defect or seal failure. |
| Unknown | 2 | 25% | Customer gave no reason or selected a generic option. Cannot be attributed to leakage. |
| Quality Unacceptable | 1 | 12.5% | Expectation gap — product didn't match the customer's quality perception from the listing. |
| Not Compatible | 1 | 12.5% | Customer bought it expecting it to fit or work with something it doesn't. A listing problem. |
| Other reasons | 2 | 25% | Buyer preference and miscellaneous codes. Standard marketplace return noise. |
Look at that breakdown. Only two returns were classified as "Defective." One was "Quality Unacceptable." One was "Not Compatible" — which has nothing to do with leaking. Two were completely unknown, and two were general preference returns.
If we had assumed the leakage review explained all eight returns, we would have misdiagnosed at least half of them. That is the danger of anchoring to a single vivid data point.
The framework: Signal → Pattern → Evidence
To avoid the trap of jumping from a single metric or a single review to an operational decision, we use a three-level framework:
Six things to check before deciding anything
When a SKU return rate spikes, work through these six checks in order. Don't skip to number six until you've answered numbers one through five.
Different diagnoses lead to completely different actions
The danger of a quick reaction is that every return looks like a reason to kill the product. In reality, each diagnosis points to a different team and a different fix:
| Diagnosis | Key signal | Correct action | Wrong reaction |
|---|---|---|---|
| Physical defect | Defective codes + specific reviews (leakage, breakage) | Inspect inventory, quarantine batch, contact supplier for credit / redesign | Immediately delisting without checking if it's batch-specific |
| Listing expectation gap | Quality Unacceptable + "smaller than expected" reviews | Update images with scale references, clarify materials, revise copy | Blaming the supplier when the listing made promises the product can't keep |
| Compatibility issue | Not Compatible codes + fitment complaints | Add compatibility chart, update bullet points, add fit guide to A+ content | Treating compatibility returns as quality defects |
| Fulfilment damage | Carrier Damaged / Customer Damaged disposition | Improve packaging, add bubble wrap or inner box, file carrier claims | Redesigning the product when the box was crushed in transit |
| Buyer remorse / trial | Unwanted Item + Sellable disposition | Factor return cost into SKU pricing; restock units promptly | Panicking over returns that were restocked at near-zero loss |
If you're running PPC ads to a SKU with a 13.3% return rate, every ad dollar is amplifying the problem. But pausing ads completely may kill momentum on a product whose only issue is an unclear listing image. Fix the diagnosis first, then adjust the ad spend to match.
Action checklist: what to do this week
-
Pull the SKU-level return report with reason codes Don't settle for the headline return percentage. Download the full report with customer return reason codes and disposition status.
-
Cross-reference reason codes against customer reviews Look for whether the reason codes in the report match what customers are saying publicly. Consistent overlap points toward a real issue.
-
Check whether returned units went back to sellable inventory Calculate the true economic cost: return processing fees + non-sellable unit loss. It's often lower than the headline return count implies.
-
Audit the product listing for expectation mismatches Check dimensions, material descriptions, compatibility claims, and color accuracy against the physical product.
-
Review PPC campaign spend on outlier SKUs Verify whether ad spend is actively scaling a SKU with unresolved return patterns. Coordinate between ads and operations.
-
Document the hypothesis before taking action Write down what you believe the root cause is, what action you're taking, and what metric should change. Test, don't guess.
Frequently Asked Questions
What is a high return rate for a single SKU? +
How many returns does it take before investigating a SKU? +
What does "Defective" mean as an Amazon return reason? +
Should I delist a SKU because of a high return rate? +
What is the Signal → Pattern → Evidence framework for returns? +
Have an outlier SKU you need to diagnose?
We help UAE brands dig into return data, connect it to advertising and product economics, and make decisions based on evidence rather than panic.
Request an account review