A negative keyword tells Amazon not to show your ad for search terms that match it. Amazon offers two negative match types: negative exact, which blocks only that query and close variants such as plurals, and negative phrase, which blocks any query containing that sequence of words. There is no negative broad match. Negatives can be set at campaign level or ad group level, and the same idea applies to ASINs through negative product targeting.
Why it matters
Negatives are the main tool for cutting wasted spend without cutting the terms that sell. A broad-match keyword or an automatic campaign will always attract queries you did not intend: the wrong size, a competitor's brand, an accessory, a use case your product does not fit. Without negatives you pay for those clicks every week. With them, the budget concentrates on queries that convert, which lowers ACoS and makes the campaign's data cleaner.
Negatives also stop your own campaigns competing with each other. When a search term is promoted to its own exact-match campaign, adding it as a negative exact in the discovery campaign it came from prevents both campaigns bidding on the same query.
What moves it
- The threshold you apply. Negating after two clicks is too aggressive for a high-priced product; waiting until $50 of spend is too slow for a $12 one. The rule should scale with price and conversion rate.
- Phrase versus exact. A negative phrase on "kids" removes every query with that word, which is right if you never sell to children and wrong if some do convert.
- Reviewing negatives later. A term negated in January can be worth testing again once the listing changes.
How we use it
Negatives are added weekly from the Search Term report, using a spend-without-order threshold tied to each product's price and margin. Every promoted term is negated in its source campaign. We keep a negative list of irrelevant brand and category words that new campaigns inherit on day one, so they do not repeat the learning cost.