Match type controls the range of queries a keyword can reach; it does not guarantee a conversion rate or ACoS. Choose broad, phrase or exact according to the exploration you want, then inspect the actual search terms and economics.

Understand the positive match types

Amazon's match-type guidance describes broad, phrase and exact, including applicable variations. Treat exact as a targeting rule rather than a promise that only identical characters can match. Amazon match types.

Type	Operating use	Review priority
Broad	Explore related query demand	Product fit and query mix
Phrase	Explore around a more constrained phrase	Longer queries and their specific intent
Exact	Give a supported query more direct bid control	Actual matched variants, CPC and contribution

A high-performing broad campaign may simply contain strong branded demand. A high ACoS exact target can still be too expensive or point at an unsuitable offer. Compare like-for-like intent groups rather than declaring a match type universally better.

Pick the KPI for the decision

For discovery, track relevant new terms, spend consumed and the quality of evidence. For established traffic, track mature sales and contribution after ads. CTR describes click response; it cannot establish profitability. CVR needs a consistent order definition and adequate observations.

Hypothetical comparison: at £30 revenue per order and 30% contribution before ads, the product has £9 contribution per order. One traffic group costs £0.45 per click and converts at 5%, so expected ad cost per order is £9. Another costs £0.70 and converts at 10%, so expected cost is £7. The second is economically stronger under these assumptions despite its higher CPC. These groups are not claimed benchmarks for any match type.

Promote queries deliberately

When a discovery query shows relevant, mature evidence, create a manual target if separate control will help. Check existing targets first and retain its source history. An exact negative in the discovery source can route traffic, but only after confirming the destination is eligible and has budget. It is not mandatory to negate every promoted term.

Keep automatic or broad discovery when it continues to find useful traffic within its limit. Pause or reduce it when economics, stock constraints or lack of useful discovery justify doing so.

Use the correct negative type and scope

Amazon keyword negatives use negative phrase and negative exact, not negative broad. Phrase negatives can suppress useful longer queries, so require a reason that applies across the phrase. Use the campaign scope only if the exclusion applies to all affected ad groups; otherwise use a narrower supported scope.

Negative product targeting is a separate mechanism for product opportunities. It does not pause the advertised product itself. See negative keyword design.

Budget and review cadence

Set a discovery allowance and a loss limit from your account's economics. No fixed exact/phrase/broad percentage is optimal for every catalog. Avoid changing every match type at once; preserve a comparison where practical.

Use comparable mature reporting periods. A fixed number of clicks is not a guarantee of stable conversion, and the latest sales can be incomplete. Amazon conversion metrics.

Should exact always have the highest bid?

No. Bid according to expected value and the effective campaign settings. The starting CPC guide shows the calculation, while the auto-to-manual guide covers the handoff.

Related Arctavia resources

Use these supporting pages to compare Amazon PPC operating models and implementation choices.

• Amazon Ad Rejections: Check Claims, Images and Destinations
• Amazon Ads ACoS Suddenly Increased: What to Check First
• Amazon Ads ML Bid Review: Guardrails Before Aggression
• Amazon PPC Automation: Data, Decisions and Operating Controls
• Amazon PPC Automation Migration Checklist
• Amazon PPC Automation vs Agency: Which Operating Model Fits Better?