How to Audit Your Own Amazon PPC Account in One Afternoon
Five native reports, a spreadsheet and three focused hours. The seven checks that turn an Amazon ad account into a ranked list of where the money is leaking.
A useful Amazon PPC audit takes five downloaded reports and about three focused hours: no tool subscription, no consultant, no account access anyone has to grant you. The point of a self-audit is not to rebuild the account in an afternoon. It is to produce a ranked list of where the money is going and which of those places you would not choose to send it again. Everything below is done in a spreadsheet. If you would rather hand the list to someone, our Amazon PPC management starts with exactly this audit.
What to download before you open a spreadsheet
Four of the five reports live in the Amazon Ads console under Measurement and reporting → Sponsored ads reports, where you pick the ad product (Sponsored Products), the report type, the time unit and the period. The fifth lives in Seller Central. As of Q3 2026 that is the path; Amazon moves this menu every couple of years, so if the wording has shifted, look for the reporting section rather than the campaign manager.
| Report | Where it lives | What you use it for |
|---|---|---|
| Campaign | Ads console → Sponsored ads reports | Spend concentration |
| Search term | Ads console → Sponsored ads reports | Wasted spend, brand vs non-brand |
| Budget | Ads console → Sponsored ads reports | Budget-capped campaigns |
| Placement | Ads console → Sponsored ads reports | Placement multipliers |
| Detail Page Sales and Traffic by Child Item | Seller Central → Reports → Business Reports | ASIN-level profitability |
Pull all of them for the same period, summary view, last 60 days. Sixty rather than ninety because Amazon's search term report only retains roughly 65 days of history. Matching the window across reports means the numbers reconcile instead of arguing with each other. Do not audit a 7-day window; you will be reading noise.
The seven checks of an Amazon PPC audit, in order
The order matters. Each check narrows the surface for the next one, and the first three usually explain most of what is wrong.
1. Spend concentration
Open the campaign report, sort by spend descending, and add a running-total column as a percentage of total spend. You are looking for how few campaigns account for the bulk of the budget.
Extreme concentration means your account is really one or two campaigns wearing a costume: anything you change elsewhere is rounding error. The opposite pattern, spend spread thinly across dozens of campaigns with no single one carrying meaningful volume, is worse: no campaign has enough data for Amazon's algorithm or for you to make a decision from. Both are structural findings, and both are answered by campaign structure rather than by bid tweaks.
2. Wasted spend in the search term report
Filter the search term report to rows with clicks but zero orders, then sort by spend. Sum that column. That number, as a share of total spend, is your first hard finding.
Read the top 50 rows as language, not data. You are looking for three species: queries for a product you do not sell, queries with intent you cannot satisfy (a different size, a different format, a competitor's model number), and queries that are simply too broad to convert. In an auto or broad campaign, ASINs also appear in the search term column. Those are the detail pages your ad ran on, not typed queries, and they get judged the same way.
Anything with meaningful spend and no orders over 60 days is a negative keyword candidate. Anything with orders at a punishing cost is a bid problem, not a negative.
3. Brand vs non-brand split
Tag every search term row as brand (contains your brand name or a close misspelling) or non-brand, then compare spend, sales and ACoS between the two groups.
Brand terms almost always look spectacular (low CPC, high conversion rate) because those shoppers were coming to you anyway. If brand traffic is carrying your blended ACoS, the account is not performing well; it is reporting well. Calculate the non-brand ACoS on its own and treat that as the real number. A healthy account can tell you what share of ad sales is defensive and what share is acquisition, without having to derive it in a spreadsheet.
4. Budget-capped campaigns
The budget report gives you average time in budget per campaign, plus Amazon's estimates of missed impressions, clicks and sales for campaigns that ran out of budget, and a recommended budget.
Any campaign that is profitable and sits below 100% average time in budget is losing hours of the day it has already earned. That is the cheapest fix in the whole audit: you are not finding new performance, you are stopping the account from switching itself off at 4pm. Treat the missed-sales and recommended-budget figures with more caution: Amazon derives them by modeling against similar campaigns that did not go out of budget, and says explicitly they are estimates for reference. The constraint flag is a fact about your account; the size of the opportunity is a prediction.
5. Placement multipliers
The placement report splits performance across Top of Search, Rest of Search, Product Pages and Amazon Business. Calculate conversion rate and ACoS per placement, per campaign.
Two failure modes show up here. The first is a large Top of Search adjustment sitting on a campaign whose top-of-search conversion rate is no better than the rest: you are paying a premium for position you are not converting. The second is an untouched Product Pages placement quietly absorbing a large share of spend at a very different ACoS, in either direction. Placement adjustments on Sponsored Products run from 0% to 900% and are increase-only, so they compound with everything else; a multiplier set once during a launch and never revisited is one of the most common findings in an account that used to perform and stopped.
6. Bid strategy mismatches
For each of your top-spend campaigns, note the bidding strategy against the job that campaign is doing. New campaigns default to dynamic bids – down only; up and down lets Amazon raise your bid by up to 100% for top-of-search placements and up to 50% elsewhere; fixed disables the adjustment entirely.
The mismatch to look for is up-and-down stacked on top of an aggressive Top of Search multiplier, which multiplies rather than adds: your effective maximum CPC can be several times the number you typed into the bid field. The reverse mismatch is a rank-defense or launch campaign left on down-only, quietly refusing to bid at the moments that matter. Neither is wrong as a setting; both are wrong when nobody chose them. The mechanics of how these interact are worth a session of their own. See bidding strategies and budgets.
7. ASIN-level profitability
Finally, join the advertised-product or campaign spend to the Business Report (Reports → Business Reports → Detail Page Sales and Traffic by Child Item), which gives you sessions, unit session percentage and units ordered per child ASIN.
Now ask the question ads reporting cannot answer: after referral fee, fulfillment fee, landed cost and ad spend, which ASINs actually make money? Almost every account we look at has at least one ASIN receiving budget in proportion to its revenue rather than its margin. A low-margin variation with a strong conversion rate can absorb spend for months and look fine in the ads console, because the ads console has never heard of your COGS.
What to do with what you found
You now have seven findings. Rank them by dollars, not by how annoying they are, and expect the order to be roughly: budget caps and negatives first (fast, low risk, reversible), then placement and bid-strategy corrections (medium risk, needs two to four weeks of data before you judge them), then structure and ASIN-level allocation last, because those change what all the other numbers mean.
Change one class of thing at a time and give each change a full sales cycle before the next. An account we took over in the outdoor category went from a 44% to a 21% ACoS while growing 68%. That work was months of sequenced changes, not a weekend of edits, and the sequencing is most of why it held.
FAQ
How often should you audit an Amazon PPC account?
Quarterly for a full pass, monthly for checks 2 and 4 (search terms and budget caps), which decay fastest. Also audit after any structural event: a catalog change, a price change, a new competitor entering your main keywords, or the end of a peak season when bids and budgets were temporarily inflated.
Can you audit an Amazon PPC account without third-party software?
Yes. Every check here runs on native reports from the Amazon Ads console and Seller Central Business Reports plus a spreadsheet. Third-party tools save time on joining and visualizing the data, but they do not see anything Amazon does not expose, and none of them know your cost of goods unless you enter it.
How long does an Amazon PPC audit take?
An afternoon for an account with a few dozen campaigns: roughly thirty minutes to pull and align the reports, two hours across the seven checks, thirty minutes to rank the findings. Large catalogs with hundreds of campaigns take a day, mostly because reading search terms as language does not parallelize.
What should you fix first after a PPC audit?
Budget-capped profitable campaigns and clear wasted spend. Both are reversible, both show a result within a sales cycle, and neither disturbs the data you need for the harder decisions. Structural changes to campaign organization go last, because they reset the performance history you would otherwise be reading.
Start here Monday morning
Block three hours, pull the five reports for the last 60 days, and work the checks in order. The ranked findings list is the deliverable, not the edits. If you would rather see the same seven checks run against your account by someone who does it weekly, our Amazon growth plan covers the same ground, and the ongoing version of this work is what Amazon PPC management actually consists of.
