Audit Rules

ReimburseOps runs four audit rules against every row in your reimbursement data. Each rule looks for a specific pattern that suggests Amazon may owe you more — or that your cost data needs attention.

Rule A: Missing Cost

Code: missing_cost Severity: High

What it detects: A reimbursement record exists, but no sourcing cost can be found for that product.

How it works:

1. Look up the reimbursement row’s FNSKU in your cost data
2. If no FNSKU match, fall back to ASIN
3. If still no match → flag

Why it matters: Without a cost reference, you can’t verify whether Amazon’s payout is fair. This flag tells you to either add cost data for this SKU or manually review the reimbursement amount.

What to do: Add the missing product to your sourcing cost CSV and re-run the audit. Or review the payout manually against your purchase records.

Rule B: Under Reimbursed

Code: under_reimbursed Severity: Low / Medium / High (depends on gap size)

What it detects: Amazon’s reimbursement payout is significantly less than your sourcing cost.

How it works:

1. Match the reimbursement to a cost record (by FNSKU, then ASIN)
2. Calculate: reimbursement_amount vs sourcing_cost × quantity
3. If the payout is less than 90% of expected → flag

Severity thresholds:

Why it matters: This is where you find money. Amazon sometimes uses outdated or incorrect valuations when calculating reimbursements. Even a 15% shortfall across hundreds of units adds up fast.

What to do: Open a case in Seller Central referencing the specific reimbursement ID. Provide your invoice or purchase order as evidence of the correct unit cost.

Rule C: Inconsistent Valuation

Code: inconsistent_valuation Severity: Medium

What it detects: The same product was reimbursed at significantly different rates for similar events.

How it works:

1. Group reimbursements by FNSKU and event type
2. Compare the highest and lowest payout per unit within each group
3. If the ratio exceeds 1.25× → flag

Why it matters: If Amazon reimbursed the same item at $12 in January and $8 in March for the same event type, one of those is wrong. This rule surfaces those discrepancies so you can dispute the lower amount.

What to do: Compare the flagged records side by side. File a case for the lower-valued reimbursement, citing the higher one as precedent.

Rule D: Cost Definition Warning

Code: cost_definition_warning Severity: Medium

What it detects: Your sourcing cost data may include non-product costs like shipping, customs, or handling fees.

How it works:

Scans cost record notes/descriptions for keywords:

• shipping
• freight
• customs
• tariff
• handling
• landed

If any keyword is found → flag.

Why it matters: Amazon calculates reimbursements based on product cost, not landed cost. If your cost CSV includes shipping or duties, the Under Reimbursed rule may produce false positives — flagging reimbursements that are actually correct.

What to do: Review your cost data. If the flagged cost includes non-product expenses, update it to reflect the pure product cost and re-run the audit.

Severity Levels

Priority Score

Each flag receives a priority score that combines:

• Severity level — high flags rank above medium and low
• Dollar impact — larger potential recovery amounts rank higher

The flagged records table sorts by priority score by default, putting the highest-value opportunities at the top. Focus on the first page and work your way down.