Google Ads is giving advertisers more room to test campaign changes before committing to them account-wide. A new round of updates covers Search campaign experiments, AI Max testing flexibility, and a faster path from Performance Planner forecasts to live changes. Some of it is live now; a broader multi-campaign testing option starts rolling out in September.
Testing budget and ROI changes across multiple campaigns at once
The headline addition is multi-campaign Search experiments, arriving in September. Today, testing a budget or target ROI change typically means running an A/B experiment on one campaign at a time. The new capability lets advertisers bundle a group of Search campaigns into a single experiment and test the same budget or ROI target adjustment across all of them simultaneously, while keeping a control group for comparison.
This builds on the one-click experiment framework Google previously shipped for AI Max, but extends it beyond single-campaign changes. Consider an advertiser weighing a meaningful budget increase across a portfolio of campaigns rather than one at a time — instead of rolling that change out campaign by campaign and hoping the pattern holds, they can now test it as a coordinated group and measure the aggregate effect on performance before deciding whether to scale it further. For accounts where budget and bidding decisions already get made at a portfolio level rather than a per-campaign level, this closes a real gap between how spend gets managed and how it gets tested.
AI Max experiments can now preserve brand and location settings
The second update addresses a specific friction point in testing AI Max for Search: advertisers can now run AI Max experiments while keeping brand controls and location settings active. Previously, those guardrails could constrain how an AI Max test was structured, which meant advertisers sometimes had to choose between preserving the campaign setup they actually intended to run long-term and getting a clean read on how AI Max would perform.
That trade-off mattered most for advertisers who rely on brand exclusions to avoid poor query matches, or on geographic restrictions to keep spend inside serviceable areas. Stripping those controls out just to run a valid experiment risked producing results that didn’t reflect real-world performance, since the test conditions differed from what the campaign would actually look like post-rollout. Now, advertisers can test AI Max with the same brand and location guardrails they’d keep in production, which should make experiment results a more reliable predictor of what happens after the change goes live.
Performance Planner adds one-click implementation
The third change shortens the gap between forecasting a campaign change and actually making it. Performance Planner can now show the projected effect of bidding or budget target changes on existing campaigns, and advertisers can apply the suggested changes directly with one click, without leaving the planning workflow to rebuild them manually inside each campaign.
Before applying anything, advertisers can review the proposed changes at the campaign level and deselect any campaigns they don’t want touched — so a portfolio-wide forecast doesn’t have to become an all-or-nothing action. Once changes are applied, they show up in the Bulk Actions section of Google Ads, where they can be monitored and undone if the results don’t hold up. The upside is obvious: less manual rebuilding of forecasted scenarios inside live campaigns. The catch is that one-click implementation raises the stakes on reviewing the forecast carefully first, since it’s now just as easy to apply a flawed recommendation as a good one.
What this means for how advertisers manage risk
All three updates point in the same direction: Google Ads keeps expanding what gets automated, and in response, it’s also expanding the tools advertisers have to validate automated decisions before trusting them at scale. Multi-campaign experiments let advertisers test larger, portfolio-level strategy changes with statistical rigor instead of inferring patterns from one-off campaign tests. Preserving brand and location controls during AI Max experiments means those tests better represent how a campaign will actually run once live. And a faster planner-to-live-campaign pipeline removes friction from acting on a validated forecast — while also making pre-application review more important, not less.
For accounts that manage spend across many campaigns rather than optimizing each one in isolation, the multi-campaign testing option in particular is worth planning around ahead of its September rollout. It’s a meaningfully different workflow than testing individual campaigns and extrapolating: a coordinated test with a real control group produces a cleaner signal for whether a budget or target change is worth adopting broadly, rather than a guess based on how one campaign happened to respond.
The practical advice heading into these changes is straightforward: treat the new one-click Performance Planner actions the same way you’d treat any bulk change to a live account — review every campaign in the proposed set individually before applying, watch the Bulk Actions log afterward, and don’t assume a forecast holds just because it’s now one click away from becoming real. The tools are getting faster to act on; the judgment calls around when to act on them haven’t gotten any less important.
Frequently asked questions
When does multi-campaign Search experimentation launch?
Google says the new multi-campaign testing option for Search campaigns begins rolling out in September, letting advertisers test budget and ROI target changes across a group of campaigns in a single A/B experiment.
What changed for AI Max experiments?
Advertisers can now run AI Max experiments for Search campaigns while keeping brand controls and location settings enabled, rather than having to remove those guardrails to run a valid test.
What does the Performance Planner update actually do?
It lets advertisers apply forecasted bidding or budget changes directly to live campaigns with one click, after reviewing and optionally deselecting individual campaigns, with applied changes tracked and reversible through Bulk Actions.
Why does testing across multiple campaigns matter more than single-campaign tests?
Portfolio-level budget and bidding decisions don’t always behave the same way in aggregate as they do on one campaign. Testing a group of campaigns together with a shared control group gives a more reliable signal before rolling a change out account-wide.
Does the one-click Performance Planner feature increase risk?
It reduces the manual effort of implementing a validated forecast, but it also makes it easier to apply a flawed recommendation quickly, so reviewing proposed changes carefully before applying them matters more, not less.
For more on how automated bidding and testing decisions play out in practice, see where to spend your time on Google Ads bidding strategies in 2026, how to test a new bid strategy in Google Ads, and Google Ads’ journey-aware bidding and budget pacing updates. For background on AI Max itself, see how Google is replacing Dynamic Search Ads with AI Max and Google’s launch of AI Max for Shopping and Travel campaigns.