Fee Audit

When to use this

• FBA fees feel high and the seller does not know which line is wrong.
• A product is near a size-tier boundary.
• High refund rate and not sure the referral fee was credited back.
• Routine quarterly audit to catch silent fee creep.

The framework. The Five Fee Anomalies

1. Size-tier misclassification. The unit is billed at a higher size tier than its measurements warrant. Cross-check listed dimensions vs the FBA tier table.
2. Missing referral refund. A refund was processed but the referral fee was not returned proportionally. Cross-check refund row vs the original sale.
3. Duplicate fulfillment fee. Two FBA fees for the same order. usually a system error.
4. Storage discrepancy. Monthly storage billed for units already removed or units in transit.
5. Reimbursement gap. A reimbursement issued at the old (lower) policy rate that should have been higher under 2025 cost-basis policy.

Step by step

1. Collect inputs. A Payments report (date range, transactions) plus product dimensions for the SKUs in scope.
2. Build the expected fees. For each SKU and order, what the fees should be by the current FBA fee tables.
3. Compare row by row. Flag every transaction where actual exceeds expected, or where a refund did not produce a matching referral credit.
4. Compute the recovery. Total dollars across all flagged transactions.
5. Open the fee reimbursement case under Seller Support > Fulfillment by Amazon

   FBA Fee Discrepancies, one case per anomaly type, with the report rows as evidence and the calculated delta. (SAFE-T is a separate process for FBM/SFP A-to-z refunds, do not file fee disputes through it.)

6. Plan the structural fix. A size-tier issue is recurring. The fix is updating the listed dimensions, not just claiming back.
7. Run the quality check, then deliver.

Output format

## Fee Audit. [account / period]

### Anomalies found
1. Size-tier misclassification . [SKU] . [units affected] . [$ delta]
2. Missing referral refunds . [orders] . [$ delta]
...

### Total recovery available: [$]

### Cases to open (Seller Support > FBA > Fees)
Case 1. [anomaly] . [evidence rows] . [case narrative]
...

### Structural fixes
[dimensions to update, processes to add]

Worked example

Quality check

• Every anomaly is one of the five named types.
• Expected fees are computed from the current FBA tables, not assumed.
• Each finding has report rows as evidence, not just a calculation.
• The structural fix is included so the leak does not recur after the recovery.

Common mistakes

• Auditing only when it feels expensive. Most fee leaks creep quietly.
• Claiming the dollars without fixing the source. Same anomaly returns next month.
• No expected-fee baseline. Without 'what should it have been', there is nothing to detect.
• Ignoring referral refunds. Easy to miss on high-refund SKUs and small per-item.

Built by Jay GPT Pro