Skip to main content
Browse topics
Back to Docs

Ad Platform Playbook

Match Type Strategy

Post-BMM match type landscape, automated bidding interaction, and negative keyword dependency

For · Operators choosing keyword match types and managing negatives
Time · 6 min read
Next · Learning period and strategy transitions

Google retired broad match modifier (BMM) in 2021, and phrase match absorbed its behavior. The match type landscape today has three options: broad, phrase, and exact. Each interacts differently with automated bidding, and your choice of match type directly affects how much you depend on negative keywords.

The Short Version

  • Broad match modifier is gone. Three match types remain, and phrase match absorbed BMM’s behavior.
  • Match type and bid strategy are one decision, not two. Broad match is designed to run alongside Smart Bidding, and it needs both conversion volume and well-defined conversion goals to work.
  • Your negative keyword obligation scales with how broad your match types are. Broad match with a thin negative list is the risk case, and Trellis flags it.
  • Do not pick a match type from published CPC trends. Your own CPC and conversion data are the only inputs that should decide it.

The Current Landscape

Match typeWhat triggers itReach vs. controlChoose this when
ExactSearches that share the same meaning as your keywordMost control, least reachYou want a controlled, high-converting core and are willing to cap volume
PhraseSearches that include your keyword’s meaning, with words before or after. Respects word order when order changes meaningMiddle groundYou want expansion beyond exact without opening up to unrelated queries
BroadSearches Google considers related, even when none of your keyword’s words appearWidest net, least controlYou have the conversion volume to train Smart Bidding and the negative-keyword discipline to contain it

Match Types and Automated Bidding

Google has designed broad match to work as part of a system alongside Smart Bidding. When broad match is paired with an automated strategy like Target CPA, the bidding can adjust in real time, bidding higher on queries it predicts are likely to convert and lower on those it doesn’t.

That said, this system depends on two things: sufficient conversion volume to train the bidding, and well-defined conversion goals so the bidding optimizes for the right outcomes. Without both, broad match can produce spending on low-intent traffic.

The Negative Keyword Dependency

Broader match types surface more search queries, including irrelevant ones. The broader your match types, the more critical your negative keyword coverage becomes.

With exact match, you’re unlikely to trigger on unrelated searches. With broad match, your ads may appear on queries you didn’t anticipate. Regular search term review and proactive negative keyword management can help control this.

Cost Dynamics

Industry data from mid-2025 showed phrase match CPCs rising 43% compared to 29% for broad match over a two-year period. This doesn’t mean broad match is automatically cheaper for your account; your own CPC and conversion data should drive that decision. Still, it’s a trend worth monitoring.

Trellis Perspective

Trellis audits segment performance by match type and flag cases where one match type is significantly outperforming or underperforming others. The audit also checks negative keyword coverage relative to the match types in use. Broader match strategies with thin negative lists are flagged as a risk.

Sources