Seller Profit Guard · How it works · CSV privacy

Marketplace fee anomaly checker

Compare two synthetic marketplace fee packets against dated rule snapshots. Enter the fee family, declared base, expected percentage rate, fixed amount, observed total, expected and observed line counts, currency, period, and evidence context. The browser-local checker flags amount, effective-rate, and duplicate-line differences without connecting to a marketplace or exposing private orders.

Maintained by Seller Profit Guard Editorial Team. Last reviewed: 2026-07-31.

Marketplace fee rule comparison from source snapshot and fee base through expected amount observed evidence anomaly decision and restoration
Compare observed synthetic fee evidence with one dated rule version before alleging a fee error.

Start with a dated rule snapshot

A fee anomaly is the difference between an observed synthetic fee total and the amount implied by a declared fee family, event grain, base, percentage rate, fixed amount, currency, effective date, and expected line count.

A current fee table is not valid evidence for an older event unless the effective dates match.

Keep this intent distinct

Use fee calculators to plan expected charges before a sale, payout reconciliation to bridge account movements, and this checker to compare observed fee evidence with one preserved rule version.

The checker does not estimate contribution or accounting profit.

Define one fee family

Separate listing, transaction, payment processing, third-party transaction, referral, promotion, logistics, advertising, tax-on-fee, regulatory, conversion, app, and shipping-label charges.

Combining unlike fees destroys the denominator.

Define the event grain

State whether one expected line belongs to a listing publication, auto-renewal, quantity, order, transaction, payment, payout, invoice period, app usage event, or other documented trigger.

Repeated order references do not automatically mean duplicated fee lines.

Record the fee base

Identify whether the base includes merchandise, discounts, shipping, gift wrap, tax, duties, store credit, gift cards, or only a provider-processed portion.

Never infer a base from the fee amount alone.

Record rate and fixed amount separately

Percentage and fixed components can coexist, vary by country or plan, and use different currencies or conversion conventions. Preserve both values and their source.

Do not hide a fixed charge inside an effective rate.

Record effective scope

Capture platform, country, seller location, plan, category, payment provider, program enrollment, promotion, channel, currency, and effective date.

One published rate rarely applies to every seller context.

Calculate expected fee per line

Expected fee per line equals declared base multiplied by the expected rate plus the expected fixed amount.

Rounding should follow the source or observed statement convention.

Calculate expected fee total

Expected total equals expected fee per line multiplied by the expected line count at the declared event grain.

The count is evidence, not a convenient multiplier.

Calculate the amount difference

Fee difference equals observed fee total minus expected fee total. Positive means observed is higher; negative means lower.

The sign does not identify the cause.

Calculate an effective rate

For percentage-bearing fees, remove the expected fixed amount per observed line and divide the remaining observed amount by the declared base.

An effective-rate difference can reveal a changed rate, wrong base, or mixed fee family.

Detect duplicate-line differences

Line-count difference equals observed matching lines minus expected matching lines. Positive values require duplicate or multi-trigger review; negative values require missing-line or scope review.

Do not delete a row until the platform event grain is proven.

Use amount and rate tolerances together

Amount tolerance handles currency precision and documented rounding. Rate tolerance handles small percentage differences. Neither can waive a wrong event grain, missing source, or open conflict.

Version and approve tolerances before an incident.

Version the fee-review contract

Record a real source-review date that covers the evidence month and is not in the future, a policy effective date no later than that review, scenario rule-effective and event dates in correct order, a minimum closed evidence duration, and bounded monetary and fee-line inputs. Confirm the nine synthetic-data, rule, grain, base, evidence, threshold, counterexample, restoration, and authority controls before interpreting a result.

Invalid governance controls mask every derived fee output.

Read Ready

Ready means both synthetic fee packets match their preserved rules within amount, rate, and line-count controls, with complete context and no declared conflict.

It proves the worksheet arithmetic only.

Read Review

Review means the evidence structure is complete but an amount, effective rate, or line count differs. Assign an owner to inspect the source rule and protected fee detail.

Do not call the difference overcharging.

Read Block

Block means a required value, count, context, currency, period, scope, privacy, or conflict control failed.

Repair the evidence packet before interpretation.

Investigate changed rates

Compare effective dates, seller country, plan, category, payment method, promotion, waiver, threshold, onboarding program, and rule version.

A legitimate scope change can look like an error.

Investigate duplicate lines

Compare fee type, event ID in protected storage, role, timestamp, amount, source file, reversal or credit status, and expected trigger count.

A listing renewal and a transaction fee may share context without being duplicates.

Investigate wrong bases

Reperform merchandise, discount, shipping, gift-wrap, tax, duty, store-credit, gift-card, and provider-processed portions using the official rule.

Do not retrofit the denominator to force a match.

Investigate fixed-fee and currency differences

Check fee currency, account currency, listing currency, conversion timing, converted fixed charges, minimums, thresholds, and minor-unit rounding.

A fixed USD fee can appear differently in another account currency.

Preserve first-party evidence

Record source URL, title, access date, effective date, relevant scope, screenshot or export fingerprint in protected storage, and superseded rule version.

A search snippet alone is not durable evidence.

Use synthetic public examples

Public worksheets need no buyer, customer, order, address, payment, bank, payout, credential, or raw CSV data. Use invented bases, amounts, counts, dates, and rule labels.

Keep operational identifiers in authorized systems.

Assign human authority

Name the evidence owner, independent reviewer, dispute or correction approver, monitoring owner, stop authority, and restoration owner.

Ready never authorizes a platform dispute or ledger correction.

Back up before correction

Preserve the source report, rule snapshot, mapping, event-grain definition, fee classifications, prior accepted comparison, exception log, downstream state, and tested restoration instructions.

The public tool creates no backup.

Monitor fee-rule drift

Track new platform notices, category or plan changes, effective dates, recurring residuals, line-count patterns, reversals, credits, and reviewer decisions.

Stop automation when an unreviewed rule changes.

Retest counterexamples

Test exact match, changed rate, duplicate line, missing line, wrong base, fixed-only fee, zero base, threshold boundary, malformed counts, invalid currency, short scope, and open conflict.

Every test must remain synthetic and bounded.

Keep platform claims scoped

Etsy transaction, listing, processing, advertising, regulatory, and conversion fees differ; Shopify plan and provider fees differ; TikTok fee families, categories, programs, and effective dates differ.

Verify current first-party guidance for the exact account.

Do not infer financial loss

An observed difference can come from scope, timing, conversion, tax-on-fee, refunds, credits, mixed sources, rounding, or an incomplete export.

Escalate only with reproducible protected evidence.

Release through ordered gates

Index only after distinct intent, current sources, functioning logic, counterexamples, answer-first content, metadata, schema, diagrams, links, privacy, similarity, mobile, backup, deployment, and live checks pass.

Search signals are measured later, not required for release.

Keep a change log

Record comparison version, rule source, effective scope, input totals, expected and observed result, tolerances, decision, reviewer, correction, monitoring trigger, and restoration reference.

Never overwrite the prior accepted rule without supersession evidence.

Sources and further reading

Related Seller Profit Guard tools

Use the interactive tool

Enable JavaScript to open the calculator and process browser-local inputs. The explanatory content and source links remain available without JavaScript.

Related guide: Define marketplace fee rules, event grain, expected amount, rate and line-count anomalies, evidence, decisions, monitoring, and restoration.

This tool provides operating estimates, not tax, accounting, legal, financial, or marketplace-policy advice. Verify current official sources and your own records before changing prices or operations.