Google Adds Asset Testing Tools to Performance Max

Google rolled out new asset experimentation capabilities for Performance Max campaigns on 8 June 2026, giving advertisers a more structured way to test creative decisions before applying them at scale.

For teams running PMax campaigns, this addresses one of the most persistent frustrations with the campaign type: the difficulty of knowing whether a creative change helped or hurt performance without risking the entire campaign budget in the process.

What the New Asset Experiments Allow

The new experiments enable marketers to compare entirely new asset groups, evaluate the impact of adding individual assets, or measure the performance of seasonal creative against evergreen content.

Advertisers will also be able to test assets generated through Google’s Asset Studio, which means human-created, seasonal, evergreen, and AI-generated assets can now be evaluated side by side within a controlled experiment.

This is a meaningful shift in how creative decisions can be made inside Performance Max. Previously, testing a new set of assets meant accepting performance variance across the entire campaign with limited ability to attribute changes to specific creative inputs.

The Second Success Metric

One of the more practically significant additions is the ability to measure experiments against a second success metric.

Rather than declaring a winner based on a single KPI, marketers will be able to evaluate how changes affect broader campaign performance. For advertisers balancing competing objectives, such as maximising conversion volume while maintaining a target cost per acquisition, this gives a more complete picture of whether a creative change is genuinely improving results or trading one metric against another.

This matters because Performance Max campaigns often serve multiple business goals simultaneously. A single conversion metric can obscure the full effect of a creative test, and decisions made on that basis can be misleading.

Consolidated Experiments Page

Conversion lift studies and experiments are being brought together under one Experiments page.

This consolidation reduces the operational overhead of managing multiple test types across different areas of the Google Ads interface. Having all experiments in one place makes it easier to track what is running, what has concluded, and how individual tests are performing relative to each other.

API and MCC Support

Expanded support for manager accounts (MCCs) and the Google Ads API is expected to begin rolling out in the coming weeks.

For agencies and teams managing multiple accounts, MCC-level access to these experiments will be significant. It removes the requirement to manage asset testing account by account and opens the door to more consistent experimentation practice across a client portfolio.

Why This Update Matters for PMax Advertisers

Performance Max has automated a large portion of campaign optimisation across Google’s inventory since its wider rollout. The trade-off, from the advertiser’s perspective, has been reduced visibility into what is actually driving performance and limited control over how creative changes affect results.

Creative remains one of the biggest levers available to Performance Max advertisers, yet testing new assets has often involved risk. The new experimentation tools provide a structured way to validate creative decisions with data before fully committing budget.

The direction Google is moving is clear. As automated bidding and AI-generated creative become more deeply embedded in PMax, the ability to test and measure the creative layer becomes more important. Advertisers who develop a systematic approach to asset experimentation now will be better positioned to use these tools effectively as Google continues to expand them.

Additional experiment and measurement capabilities are planned for future releases.