Google Ads October 2026: Performance Max Asset Experiments Are Live and Every Ecommerce Brand Must Run One Before Q4

Common Thread Collective

by Common Thread Collective

Oct. 02 2026

On October 2, 2026, Google rolled out asset experiments for Performance Max campaigns to all advertisers. For ecommerce brands running PMax, which now drives 45% of all Google Ads conversions, this means you can A/B test creative assets with statistical confidence for the first time, replacing guesswork with data on which images, headlines, and videos are actually driving results.

What Does Google's Performance Max Asset Experiments Feature Mean for Your Campaigns?

Performance Max has been a creative black box since launch. Asset experiments fix that by letting you run a controlled test between your existing assets (the control arm) and a modified or new set (the treatment arm), with a clear statistical read on which drives more conversions. With PMax responsible for nearly half of all Google Ads conversions and BFCM approaching, this is the most actionable creative testing tool Google has shipped all year.

How Did the PMax Creative Black Box Problem Affect Ecommerce Brands?

Before asset experiments, Performance Max delivered blended campaign results across all asset groups with no way to isolate what was driving performance. You could see that your PMax campaign was generating conversions, but you could not determine whether the new lifestyle video was outperforming the product-white-background image, or whether a testimonial headline lifted your conversion rate. Brands running dozens of product lines across multiple asset groups had no reliable way to learn which creative direction deserved more investment.

Asset experiments close that gap by building a controlled testing structure directly into Performance Max. You now have a scientific path to creative learning inside the platform that was previously impossible without expensive third-party incrementality tools or rough geo-tests.

What Are the Exact Settings for a Google Performance Max Asset Experiment?

Asset experiments are accessed through the Experiments tab in Google Ads. Here is how the structure works:

  • Control arm: Your existing asset group, unchanged for the duration of the test
  • Treatment arm: A modified or entirely new asset group with the creative you want to measure
  • Traffic split: You set the percentage split between control and treatment (50/50 recommended for clean measurement)
  • Lock rule: Once an experiment starts, the asset group is locked. No edits, additions, or removals are permitted until the test concludes
  • Recommended duration: 4 to 6 weeks for statistically meaningful results
  • Second success metric: Coming soon, an optional second metric lets you measure broader impact alongside your primary conversion goal

All asset experiments and conversion lift studies are now consolidated in one Experiments page. Google Ads API and Manager Account (MCC) support is rolling out in the coming weeks for agencies managing multiple client accounts. Assets generated in Asset Studio can be added directly to a treatment arm, including Gemini-powered AI video assets built from your product URLs and photos.

For context on where Performance Max fits in your Q4 strategy, see how Google's Product Value Optimization works in PMax and track every Google update this year at our full Google Ads 2026 update tracker.

Which Creative Assets Should Ecommerce Brands Test First in Performance Max?

Video is the highest-leverage starting point. Google's own data shows campaigns with video assets see 25 to 40% performance improvement compared to campaigns running without video. If you are running PMax without a video asset in your asset group, your first experiment is clear: add one and measure the lift against your current creative.

For brands already using video, here is the priority ranking for asset experiments heading into Q4:

  1. Video format: Test 9:16 Shorts-optimized vertical video against your existing 16:9 or 1:1 formats for reach and CTR across placements
  2. Image style: Lifestyle photography vs. product-only vs. user-generated content to identify what converts at the asset level
  3. Headline angle: Price-forward vs. outcome-forward vs. urgency-forward messaging in your text asset group
  4. Seasonal vs. evergreen: Run BFCM-specific creative against your permanent brand assets to calibrate how much of a holiday lift you actually get from seasonal messaging

Asset Studio lets you generate AI video assets directly from your product URLs and photos using Gemini Omni. These can be included in a treatment arm without a production budget, making it possible to test AI-generated video against your existing assets before committing to a full production shoot. If you want a framework for running systematic creative tests across PMax, CTC's Google Ads team runs structured experiments for 7 to 9 figure ecommerce brands every quarter.

STAY CURRENT

Ad platforms change weekly. Your inbox shouldn't miss it.

Get CTC's weekly breakdown of the platform changes that actually matter for ecommerce brands -- no noise, just what affects your campaigns.

What Should You Do Now with Google's Performance Max Asset Experiments?

  1. Open the Experiments tab in Google Ads today and look for the asset experiment option under your Performance Max campaigns. If you do not see it yet, check back within the next few days as the rollout completes across all accounts.
  2. Start with one variable per test only. Swap an image set, add a video, or test a new headline group, but do not change multiple elements at once. Multiple changes make it impossible to identify what caused any result shift.
  3. If you have no video assets in your PMax campaigns, make your first experiment a video addition. Campaigns with video assets see 25 to 40% performance improvement. Use Asset Studio to generate a Gemini-powered video from your product URLs at no added cost if you do not have footage ready.
  4. Set a 4 to 6 week test window with a 50/50 traffic split. Shorter windows will not generate statistically reliable results, especially in accounts with lower conversion volume.
  5. Launch your BFCM seasonal creative experiment immediately. You need 4 to 6 weeks of data to make confident Black Friday budget decisions, and that window closes in mid-October.

Related Reading

Frequently Asked Questions

Can I run multiple asset experiments on the same Performance Max campaign at once?

No. The asset group being tested is locked for the duration of the experiment, which means only one experiment can run per asset group at a time. If you have multiple asset groups in one PMax campaign, you can run separate experiments on each, but they will be measured independently.

How long should a Performance Max asset experiment run for reliable results?

Google recommends 4 to 6 weeks. Shorter tests, especially on accounts with lower conversion volume, will not accumulate enough data to reach statistical significance. Ecommerce brands with high conversion volume may be able to use shorter windows, but 4 weeks is the minimum for any meaningful read.

Can I use AI-generated assets from Asset Studio in a Performance Max experiment?

Yes. Assets generated in Asset Studio, including Gemini-powered AI video assets created from your product URLs and photos, can be added directly to a treatment arm. This is one of the most practical applications: test an AI-generated lifestyle video against your existing product creative without a production shoot.

Does the asset experiment feature work in Google Ads API and Manager Accounts?

Google Ads API and Manager Account (MCC) support is rolling out in the coming weeks as of October 2, 2026. If you manage multiple client accounts through an MCC, watch for the expanded rollout. In the meantime, experiments can be set up directly in individual Google Ads accounts through the Experiments tab.

What happens to my asset group during a Performance Max experiment?

The asset group is locked for the full duration of the test. You cannot add, remove, or edit any assets until the experiment ends. This is by design: changes during the test would contaminate the results. Plan your creative before starting and do not launch an experiment if you expect to need asset updates in the near term.

CTC's performance team runs structured creative experiments across Google Ads for 7 to 9 figure ecommerce brands every quarter. If you want a systematic approach to Q4 creative testing across Performance Max, talk to our team.


Common Thread Collective

Common Thread Collective is the leading source of strategy and insight serving DTC ecommerce businesses. From agency services to educational resources for eccomerce leaders and marketers, CTC is committed to helping you do your job better.

For more content like this, sign up for our newsletter, listen to our podcast, or follow us on YouTube or Twitter.

Cashmas in July — Turn stale inventory into cash before Q4. Apply Now →