Google ads adds multi-campaign A/B tests for budgets and ROI

56

TL;DR:

Google Ads is simplifying A/B testing for Search campaigns, letting advertisers test budgets and ROI targets across multiple campaigns in one experiment. The September rollout also expands AI Max testing, giving marketers a cleaner way to validate changes before moving more spend.

Article:

Google is simplifying A/B testing in Google Ads, allowing advertisers to test different budgets and return-on-investment targets across multiple Search campaigns within a single experiment. Rolling out in September, the change matters because it reduces the operational friction between identifying a campaign opportunity and testing whether additional spend actually improves returns.

The new workflow also extends to AI Max for Search. Advertisers using brand or location controls can now retain those safeguards while testing AI Max, making it easier to compare Google’s AI-powered search features without first loosening campaign guardrails.

Google says advertisers can now “test different budgets and ROI targets across multiple Search campaigns in a single A/B test.” Its AI Max experiments also avoid creating a separate campaign copy: Google Ads Help says the system “splits your current campaign 50/50 to get results faster.”

The timing is significant because paid-search economics leave less room for untested assumptions. WordStream’s 2026 benchmark, based on 13,474 US search advertising campaigns, reported a median cost per click of $5.42 and cost per lead of $66.69. That makes cleaner experimentation increasingly useful when marketers are deciding whether incremental budget deserves to move.

Google has also connected experimentation more closely with Performance Planner, which can model how bidding or budget changes may affect existing campaigns and then apply selected recommendations directly.

For advertisers, the practical takeaway is simple: easier testing should not mean automatic scaling. Use the new Google Ads experiments to isolate one material change, define the success metric before launch and move budget only when the measured result supports it.