Amazon Negative Keywords: When to Add Them (2026)

Negative keywords are usually described as housekeeping — something to tidy up when you get around to it.

On a product with a $12 margin, that's fair. On one paying a $2.02 royalty, it isn't. A single irrelevant search term pulling twenty clicks a month costs you four sales' worth of profit, and it will keep doing it quietly for as long as you leave it alone.

This is what to negate, when, and the mistake that costs more than the waste it prevents.

What a negative keyword actually does

A negative keyword tells Amazon: do not show this ad for this search term. It doesn't lower a bid or reduce priority — it removes you from that auction entirely.

There are two kinds, and the difference matters more than most guides suggest:

Negative exact blocks only that precise phrase. Add dog collar as negative exact and you still show for dog collar leather. Negative phrase blocks any search containing that sequence. Add dog collar as negative phrase and large dog collar, dog collar leather and red dog collar blue all stop.

Phrase is the bigger hammer and the one that causes accidents. It's the right tool for a whole category you never want to appear in; exact is the right tool for a single term that spent money and didn't convert.

There are also negative product targets — blocking a specific ASIN your ad keeps appearing next to. Under-used and often the biggest single source of waste on product-page placements.

When a term deserves negating

Here is where most sellers guess, and guessing costs money in both directions.

The question is: if this term converted at my normal rate, how surprised should I be to see zero orders?

P(zero orders) = (1 − conversion rate)^clicks

At a 10% conversion rate:

Clicks with 0 orders	Chance it's just luck
10	35%
20	12%
30	4%
40	1.5%
50	0.5%

So 10 clicks and no sale means nothing. 40 clicks and no sale is a real finding.

If your conversion rate is 5% rather than 10%, double every click figure in that table. Run it with your own number before you cut — this is a two-minute calculation that stops you killing terms that were about to work.

The exception is intent. Some terms are obviously wrong regardless of volume: free, used, a competitor's brand name you can't legally target, a product type you don't sell. Those you negate on sight, at one click.

What to negate on thin margins

If your break-even ACoS is 11% rather than 40%, the threshold for "wasteful" moves a long way down. Three categories worth a standing rule:

Terms that convert but not profitably. A term at 25% ACoS is a success on private label and a loss on a $2.02 royalty. These are harder to spot than zero-converters because they look like they're working. Sort by spend, not by ACoS, and check each against your break-even rather than a benchmark. Free and research intent. free t-shirt design, how to make, template, svg, png, mockup — for print-on-demand these pull people looking to make a shirt, not buy one. Negative phrase on free, svg, png and template is one of the highest-return five minutes available to a Merch seller. Wrong-format terms. For KDP: audiobook when you have no audio edition, hardcover when you sell paperback only, pdf at all. Each is a phrase negative that pays for itself indefinitely.

The auto-to-manual negative, which is not optional

This is the one people miss, and it costs the most.

The standard workflow is: run an automatic campaign to discover converting search terms, then promote the winners into a manual campaign where you control the bid. Fine so far.

If you stop there, your two campaigns now bid against each other for the same term. Amazon doesn't let you compete with yourself in the literal sense, but the auto campaign keeps serving that term at its own bid, and you lose the control you built the manual campaign to gain.

Every term you promote to manual must be added as a negative exact in the auto campaign in the same session. Not later. The promotion and the negative are one action with two parts, and doing only the first half is worse than doing neither.

When negatives go wrong

Over-negating with phrase match. Blocking shirt because shirt template wasted money removes you from every search containing "shirt". Check what a phrase negative actually blocks before you add it. Negating too early. Covered above, but worth repeating because it's the most common error: pausing terms at 10 clicks feels productive and destroys perfectly good terms roughly a third of the time. Negating at the wrong level. A negative added at campaign level applies to every ad group in it. If one ad group legitimately wants that term, you've just blocked it there too. Ad-group level is the safer default; campaign level is for terms you never want anywhere. Forgetting they're permanent. Negatives don't expire. A term you blocked during a stockout stays blocked after you restock. Worth re-reading your negative lists once a quarter.

Doing it at scale

For a handful of campaigns, the search term report in the console is enough.

Past that, the bulk file is the only sane route — it carries every campaign, ad group, keyword and target in one sheet, so you can sort the whole account by spend, find every zero-converting term above your click threshold at once, and write the negatives back in a single upload. What takes an afternoon in the console takes about ten minutes in a spreadsheet.

That is most of what PPC Optimizer Pro automates: it reads the bulk file, applies your click and spend thresholds against your real break-even — royalty-based, with returns subtracted — and hands back a file with the negatives already written.

The short version

Negate on 40+ clicks with zero orders, not 10 — unless intent is obviously wrong.
Use exact for single terms, phrase for whole categories, and check what phrase actually blocks.
On royalty products, also negate terms that convert above your break-even, not just ones that don't convert.
Always add a negative exact in the auto campaign when you promote a term to manual.
Re-read your negative lists quarterly.

Negative Keywords on Amazon: What to Add When Your Margin Is $2