What £12,000 of Monthly Amazon Ad Spend Actually Buys

Analysis · Verified against Amazon Ads documentation, 27 August 2026

Drift almost never concentrates in one bad search term. It spreads across
hundreds that individually look like noise, which is exactly why it survives a manual review.

Start with what Amazon says broad match does

Amazon’s own example, from its targeting guide. The keyword is sneakers. Amazon says
it can match:

“canvas sneakers,” “sneaker,” “basketball shoes,” “athletic shoes,” “cleats,”
“trainers,” or “foam runners”

“Cleats” and “foam runners” share no words at all with “sneakers.” That’s Amazon
documenting its own behaviour. Broad match treats your keyword as a topic and decides what else
belongs to it.

The modelled account

£12,000 monthly spend, 38 campaigns, a £1.05 average CPC and a 9% conversion rate. Roughly 11,400
clicks, spread across about 4,200 distinct search terms.

Why a manual review finds nothing

At that spread, the median search term gets fewer than three clicks a month. Scan
the report and every row looks defensible — three clicks and no order isn’t evidence of anything.

So a reviewer negates the twenty worst offenders, recovers maybe £400, and concludes the account is
clean. The other waste is still there, distributed too thinly to see.

What n-gram analysis does instead

Aggregate on the shared words rather than the whole query:

1. Download the search term report
2. Split every term into monograms, bigrams and trigrams
3. Aggregate impressions, clicks, cost, orders and sales across every search term containing
   that gram
4. Rank grams by spend with zero or negligible conversions
5. Negate at gram level

In the modelled account, four grams surface:

£936 a month, invisible at search-term level because no single row exceeded five clicks. Annualised,
that’s over £11,000 on a £144,000 spend.

The negation threshold that beats a flat rule

You’ll see flat rules published — negate at £35 spend with no conversions, or 20 to 30 clicks. Fine
as a starting point, but they ignore your conversion rate, which is the only thing determining how
many clicks a converting term should need.

Expected clicks per order = 1 ÷ conversion rate. Allow 2–3× that before
negating.

• At 10% conversion, one order takes ten clicks — negate at 20–30
• At 4%, one order takes 25 clicks — negating at 20 cuts terms before they’ve had a fair test
• At 20%, waiting for 30 wastes money on terms you could have called at 12

Three constraints on the report itself

Before you trust the numbers, know what Amazon publishes about them:

• “Search term reports only include impressions that resulted in at least one ad click.”
  Impressions with no clicks are invisible — you’re seeing the spending part of drift, not its full
  extent.
• Product page placements show an asterisk. Real spend you cannot attribute to a
  query.
• There is no Sponsored Display search term report at all. SD waste is inferred,
  never measured.

And the timing. Amazon publishes a 12-hour P99 for impressions and clicks to become available, but
up to 72 hours for invalidation — Amazon reserves three days to strip invalid traffic.
That’s the hard number behind “don’t optimise on fresh data.” Work on data at least 72 hours old, and
preferably a fortnight, so the attribution window has closed.

Your negative keyword capacity is not the constraint

Widely cited guidance says 1,000 negatives per campaign and per ad group. Amazon’s live
limits page says 10,000 for both. Most published advice is running on a 2023 figure.

Which means a short negative list is a process problem, not a platform one. Note also that
there is no negative broad match — only negative exact and negative phrase exist. That
asymmetry is the whole problem: broad match expands into synonyms and related terms, and your
negatives cannot.

A figure we won’t quote

You’ll see it stated as fact that the average Amazon seller wastes 28–40% of monthly ad spend. We
traced it: one agency cites another, which attributes it to “PPC audit data from industry analysts”
with no study, no sample size and no methodology.

No published study with a disclosed methodology quantifies this. Run your own
n-gram analysis and you’ll have a number for your account that’s worth more than a figure nobody can
source.

Prepared by Ecom Enable. Match-type definitions, campaign limits and reporting
constraints verified against Amazon Ads documentation on 27 August 2026. The account figures are a
model, not a client result.