When a keyword's campaigns drift out of control — often because automated bidding software silently duplicated or restructured them — the fix is manual, granular campaign management rather than a blanket spend increase.
Method:
Case example: one keyword ended up spread across 5 duplicate exact campaigns from an automation bug; another had 3 broad campaigns running alongside 1 exact/phrase campaign that was the actual proven winner. In both cases the redundant campaigns diluted performance on a keyword that was otherwise a proven winner — trimming to the single best campaign, then optimizing its placement, was the fix.
This is the tactical follow-through to Two-Step Conversion-Then-Indexing SOP's conversion-rate-first step: once SQP benchmarking flags a keyword worth more spend, this is how that spend gets concentrated rather than spread thin.
One concrete application: some sellers deliberately exclude Phrase match entirely from their campaign structure, on the reasoning that anything Phrase match would catch is already covered by Broad match. Running both just duplicates ad spend across two match types chasing the same search terms and muddies the clean per-match-type data split that siloed campaigns are meant to produce (see Automatic Campaign Four-Way Split (Close Match, Loose Match, Substitutes, Complements)).