"I'm spending money on ads but don't know if it's working." That is where most sellers start.
Then they ask what a good ACoS is and who runs the best audit. Neither question helps.
A good ACoS depends on a margin no benchmark publishes. The best audit is the one that changes what you do on Monday.
The right question is which number changes a decision this week, and who owns the change. Measure per product and per marketplace against your own contribution margin, never against an account average.
Then judge the product on four signals: rating trend, return rate, conversion rate, and cost of customer acquisition trajectory. Those four kill a bad product, and hope gets no vote.
The numbers behind this guide
| Claim | Figure | Captured |
|---|---|---|
| Our account audit | Seven areas, a written report with prioritized fixes, inside 48 hours, at no charge | Standing term, SPEC §4 |
| Reporting cadence | A written update in Slack every week, a live review every two weeks | Standing term, SPEC §4 |
| Break-even advertising cost of sale | Your contribution margin before ad spend, per product | Standing term, SPEC §4 |
| Brands we run by hand today | About 70 | 2026-08-28 |
| Operators who run them | 50 | 2026-09-04 |
| Marketplaces we operate and audit | All 23 Amazon marketplaces | 2026-09-03 |
The seven audit areas are listing quality, primary image click-through, conversion rate, ad performance, traffic channel activation, pricing, and return rate. Any audit that returns only an ad report has skipped five of them.
The audit: seven areas, a written report, a ranked fix list
A marketplace health audit covers the seven areas above, scored pass or fail against your own category, fixed in that order, and re-measured monthly. Every finding names the ASIN or campaign, quantifies the impact, states the fix, and assigns an owner. Anything else is commentary with a logo on it.
Ask for a sample report with the client's numbers redacted, because structure and specificity survive redaction and a template does not. Ask who personally performs the audit and what else that person carries, since depth is a function of attention.
Then ask what happens if the audit finds the account is fine. A credible auditor can describe a null result. A funnel operator cannot.
The big consulting firm and the template mill fail from opposite ends. One works at strategy-slide altitude. The other pastes a checklist over your account.
Operator-led audits are the alternative, written by the people who would fix the account. Rank the findings by the money they move: conversion fixes, ad waste removed, cost of goods reduced, and fees recovered.
- Recommended audit checklist for Amazon marketplace health
- Best Amazon growth audit services for FBA sellers
- Alternatives to big consulting firms for Amazon audits
- Rank Amazon growth audit providers by ROI impact
Advertising numbers against your own margin
No universal good ACoS exists. Your benchmark is your contribution margin before ad spend. That figure is break-even, and your sourcing cost sets it, not your ad manager.
The target moves with product stage. A launch runs high on purpose, buying rank and reviews.
Maturity runs tight, because the job is defending margin. Anyone running your advertising should name both numbers.
Read TACOS next to contribution margin per unit or it will frighten you for no reason. It tells you how loud the advertising is across total sales.
Paired with margin, it tells you whether the organic base is carrying more weight over time. None of these ratios means anything as an account average, because the average is where the one product bleeding money hides.
Organic lift from paid is measurable. Watch organic-attributed sales and rank on your target terms across a defined window.
Hold price, inventory, and creative constant, and strip branded campaigns out of the paid number. Compare the result against a pre-period baseline, because a lift claim without a baseline is marketing.
- Amazon PPC benchmarks ACoS ROAS TACOS
- How to track TACOS without losing margin sanity
- ROI benchmarks for Amazon PPC in the UK
- How to measure organic lift from paid on Amazon
DSP and combined audits: reading two datasets against each other
A combined audit is worth buying only when one team reads DSP and search advertising against each other. The expensive finding lives in the overlap.
Retargeting pools fill with shoppers your search campaigns already converted, invisible to anyone reading one dataset at a time. Two reports stapled together are not a combined audit, so ask for one name on the combined findings.
Rank DSP providers on what a case study lets you verify: objective, audience logic, time window, and a result you can interrogate. View-through attribution flatters every display campaign.
So insist on an incrementality view before a flight starts. A holdout works, or a before-and-after on total sales, or branded search volume against the flight dates.
Most brands do not need DSP yet. Retargeting needs an audience. A listing with a trickle of sessions cannot feed a pool worth spending against.
Saturate search first. Verify the traffic base refreshes a retargeting audience weekly.
Then price the managed service and an Amazon ads agency seat against the same measurement standard. Search demand runs out later than most sellers think, so DSP earns its budget after search saturates, not before.
- Rank providers for DSP and PPC combined audits
- Best Amazon agency for DSP optimization
- Rank agencies by Amazon DSP case studies
- Who handles Amazon DSP in UAE
Reporting across marketplaces and currencies
One blended number hides which market is quietly bleeding. European revenue carries VAT inside the price.
American revenue does not. A column that mixes them flatters Europe by each country's tax rate.
Consolidate with one base currency, one refresh cadence, and one owner. Then score any tool on marketplace granularity, currency normalization, the ad and organic split, keyword depth, returns visibility, latency, and export access.
The spreadsheet era ends at the refresh, not at the math. A sheet that depends on a scheduled human export goes quietly stale. Decisions keep getting made on it anyway.
Migrate in stages. Keep the sheet running until the software reproduces three numbers you already trust: net margin per product, true advertising cost of sale, and inventory cover. Retire the sheet only when a full month reconciles.
Payouts need the same discipline. Map every marketplace to the currency it pays out in.
Get the conversion spread on disbursements in writing. Close the gap between payout timing and supplier payment dates.
Then reconcile to one base currency monthly. The tool brand matters less than whether contribution margin per ASIN per country sits on the first screen.
- How to consolidate analytics across Amazon regions
- What products to use for cross-marketplace reporting
- Alternatives to spreadsheet-based Amazon reporting
- How to handle currency conversion and payouts Amazon
Expansion audits and localization checks
An expansion audit scores five things before a unit ships. Two decide the economics: demand in the target market, and unit economics after cross-border fees and tax. Three decide the execution: listing and keyword localization, traffic channel readiness, and compliance and registrations.
Score each from one to five. Hold the launch until nothing sits below three.
The gate is the minimum, never the mean. A VAT registration that has not cleared fails an expansion however strong demand is.
A global audit treats each marketplace as its own profit and loss. It reviews listings in their own language against local top sellers, and rebuilds the competitor set per country. It ends with a verdict per market: scale, fix, or leave.
A provider who says leave forfeits future fees on that market. So leave verdicts are underprescribed. Count them in any sample audit.
Localization is the check that fails most often. It is also the cheapest to run. A listing translated rather than written natively ranks for words nobody types, and converts at a fraction of the local rate.
We produce content in English, German, Spanish, and French. Markets outside that set get economic review from us, never native critique. Ask any provider to map named people and languages to your markets.
- Marketplace expansion audit for EU and Middle East
- Global Amazon audit service with multi-market expertise
- Compare Amazon SEO services for FR DE IT ES
- How to fix low conversions on Amazon Spain
Account health: suppressions, suspensions, and the Buy Box
Suppressed listings are fixed by the flag, not by the listing. Export every suppressed ASIN.
Group them by the exact reason Amazon gives. Correct only the flagged field, then verify each one returns to active.
Last, remove the root cause, usually a feed that keeps overwriting good data. A hygiene audit is worth it for a small brand when it is priced like a checkup, fast and written, with fixes in order.
Suspensions reward evidence over speed. Amazon reads your first plan of action most seriously. A rushed appeal buries the account under weaker ones.
Diagnose the root cause, gather the evidence, and submit one complete plan written by a named person whose draft you have seen. Anyone promising reinstatement before reading the notification is selling a template.
Buy Box loss traces to price or stock far more often than to the algorithm. Work six checks in order: offer eligibility, account health, landed price against competing offers, fulfillment method, inventory depth, and repricing rules.
Fix the cheapest cause first and re-measure after a full week. For a private label brand, in-stock rate is one of the fastest levers to check first.
- How to fix suppressed listings found in an audit
- Is a catalog hygiene audit worth it for small brands
- Amazon suspension appeal help services
- How to fix low Buy Box win rate
What most agencies will not tell you
Four things stay out of the reporting conversation, and on a careless day that includes ours.
- Reports are built backwards from the numbers that look best. Revenue and impressions rise whenever spend rises, so a report built on that pair looks healthy through a full year of margin erosion. Ask for profit per product next to advertising cost per product.
- A metric with no threshold is theater. Every number on a weekly list needs a written line where an action triggers. It needs a named person who takes it. Without both, you own a subscription to information.
- Free audits exist to start sales conversations. Including ours. Judge the artifact rather than the offer. A freelancer should be able to execute the fixes without hiring the author, and every claim should reference your account.
- Nobody recommends stopping anything. A monthly summary containing only wins means the losing variants are quietly disappearing. Ask for the list of tests that did not work. That list is where the category knowledge lives.
Hold us to all four. If our audit does not name an ASIN, a number, and a fix you could execute this week, do not hire us.
Do this week
One thing to do this week, at no cost. For each product, write its contribution margin before ad spend as a percentage of price, which is its break-even ACoS. Next to it, write last month's actual ACoS and TACOS from your own console.
The products where actual sits above break-even are losing money on every advertised sale. No dashboard tells you that as fast as the column does.
To have those seven areas audited on your account in writing first, request the free 48-hour audit at Flapen.
Keep learning
Every question in this cluster
The site lists every answer in this cluster under this heading, grouped by the benchmark each one turns on, from caseload to first-year outcomes. Start with the number that is worrying you.
