Match Types and Negatives: Steering Broad Without Bleeding
Match type debates usually get framed as control versus reach. That framing is a decade out of date. Modern matching is semantic, exact is no longer literal, and broad match behaves less like a wildcard and more like a recommendation engine that reads your conversion data. The practical question is not which match type is safest, but whether your account has the signal quality to steer a loose one.
What each type is for
| Match type | Best use | Requirement |
|---|---|---|
| Exact | Proven, high-value queries you must win | Enough volume to justify the isolation |
| Phrase | Themes with a consistent core intent | Regular search term review |
| Broad | Discovery and scaling with automated bidding | Reliable conversion signal and active negatives |
Broad match has a precondition
Broad match with manual bidding and weak conversion tracking is how accounts burn money on irrelevant queries. Broad match with clean conversion data and a value-based target is how accounts find demand no keyword researcher would have listed. The difference is entirely in the signal, not the match type.
Give the system a loose target and a clear objective, or a tight target and a vague objective - never both loose.
The weekly search terms ritual
Set aside thirty minutes a week. Sort the search terms report by cost, read the top hundred, and sort each into three buckets: irrelevant becomes a negative, relevant but off-theme becomes a candidate for its own ad group, and relevant and converting stays. This single habit does more for account efficiency than most structural rebuilds.
Negative lists as shared infrastructure
Negatives added campaign by campaign drift out of sync. Maintain shared lists instead: a universal list for things you never sell, a competitor list, a jobs and careers list, a free-and-cheap list for products with no free tier. Apply them at account level and add to them centrally, so a lesson learned in one campaign protects every other.
Keep brand separate
Brand queries convert cheaply because the person already decided. Mixed into the same campaign as non-brand, they drag the blended cost per acquisition down and hide whether prospecting is working at all. Isolate brand, report it separately, and judge non-brand on its own numbers - that is the only view that tells you whether paid search is growing the business or billing you for demand you already had.
Related articles
Retargeting Works Until It Annoys: Setting Frequency Honestly
Retargeting does not create demand, it closes it. Budget it like a closing tool and cap it like one.
Meta Ads Account Structure in 2026: Consolidate, But Not Blindly
Fewer campaigns usually means faster learning. But consolidation fails predictably in three situations, and knowing them is worth more than any template.
Video Ads That Earn the First Three Seconds
The first three seconds decide whether the other twenty-seven exist. Most budgets are spent producing the part nobody reaches.