The Amazon PPC audit that finds the 30% of ad spend doing nothing

Most Amazon ad accounts are running on a number that flatters them. ACOS looks fine, spend looks controlled, and nobody goes looking any further. In our own audits of client accounts at VendLab, we typically find that 30% to 40% of a seller’s Sponsored Products budget has landed on search terms that have never earned a single order. Not a rounding error. Real money, sitting on the wrong keywords, every month.

Why a healthy ACOS does not mean healthy spend

ACOS is an average across the whole account, not a report on where the money actually went. A campaign can post a perfectly respectable ACOS while a third of its budget sits on clicks nobody ever bought from. The good terms carry the bad ones, and the summary number never tells you which is which. You only see it once you stop reading the top line and start reading the search terms underneath it.

There is a second cost that rarely gets counted. Every pound spent on a dead search term is a pound not spent on one that converts, and Amazon’s own ranking signals respond to sales, not to spend. A keyword that never converts does nothing for your organic rank. A keyword that does convert is quietly building it every time it fires. So the wasted third of the budget is not neutral. It is actively slower than putting the same money behind what already works.

Amazon’s own targeting tools make this easy to fall into in the first place. Broad match and auto campaigns are built to cast wide, and Amazon will keep buying a click that loosely matches your product long after it has proved it never converts. Amazon is fairly direct about the fix in its own advertiser guidance: negative keyword targeting “prevents ads and content from appearing in search engines for specific search terms,” and its own advice is to sit down periodically and “review which keywords and search terms are performing best” before trimming the list down (Amazon Ads, Keyword Targeting guide, checked 12 September 2026). That review is the whole audit. Most sellers run it once, at launch, and never open the report again.

In my experience, this rarely happens because a seller was careless setting the campaign up. It happens because the account has kept running and sales have kept coming in, so nobody had a reason to think the Search Term Report needed a second look. The leak sits quietly underneath an ACOS number that never flagged it, for months, sometimes for years.

The five checks we run on every account

This is the version we actually run on a new client account, in the order we run it. It takes about an hour on a mid-sized account and it usually pays for itself in the first week.

1. Pull the Search Term Report in Campaign Manager and sort it by spend, high to low. This is the report Amazon itself points sellers toward, and it is the one most accounts open once and never again.
2. Flag anything with real spend and zero orders in the last 60 days. That is the leak, not a guess. A handful of terms usually account for most of it.
3. Judge by placement, not just keyword. A term can convert well at the top of search and lose money in the rest of search or on product pages, so the same keyword can be a good spend in one placement and a bad one in another.
4. Check every ad is pointing at a listing that can actually win the sale. An out of stock or suppressed ASIN still spends the budget behind it for nothing.
5. Add the confirmed non-converters as negative exact matches and move that freed budget onto the terms already converting. Then set a reminder to repeat the check monthly, so the same waste does not simply grow back.

What this is actually worth

Say an account spends £6,000 a month on Sponsored Products. A typical audit at that level finds close to £2,000 of that sitting on search terms that never converted. Cut those terms and put the same £2,000 behind the keywords already proven to sell, and you have not spent a penny more. You have simply stopped paying for clicks that were never going to buy anything, and started paying for more of the ones that do.

The maths scales with the account rather than staying fixed at one figure. A seller spending £2,000 a month is typically looking at somewhere around £700 sitting on non-converting terms. A seller spending £20,000 a month is often looking at something closer to £7,000. These figures are illustrative, drawn from the pattern we see across audits rather than a guaranteed percentage for any one account, but the shape holds broadly regardless of size.

Do it before Q4, not after

This is worth doing now rather than in November. Every keyword gets more expensive as Q4 approaches, so a leak that costs you relatively little in September gets considerably more expensive once the whole market is bidding for the same clicks. Running this audit in good time also means the budget you free up is already working for you through Black Friday and Cyber Monday, rather than sitting on dead terms during the busiest weeks of the year.

An Amazon ad account that has been running for six months without this check has almost certainly built up a layer of dead weight in its Search Term Report. That is not a sign anyone did anything wrong. It is simply what an account looks like when nobody has gone back to read the report since launch. Fix it, and every pound you already spend does more, before you ever raise the budget at all.

This is the audit we run on every new VendLab client’s account in their first week. If you would like a second pair of eyes on yours, get in touch.