Google Ads AI Max: What I’d Check Before Turning It On

If you run an ecommerce business, you probably don’t need another Google Ads feature asking for more freedom with your budget.

You need to know what it changes, what you can control, and whether it will help you sell more products profitably.

That’s where I’d start with Google Ads AI Max.

Google is changing how Search campaigns match queries, create ad copy, and choose landing pages. It’s also moving certain older features into AI Max, including Dynamic Search Ads beginning in February 2027. Meanwhile, ads are appearing inside AI Overviews, and Google continues testing formats in AI Mode.

There’s a lot happening. There’s also a lot of advice that boils down to “turn on the AI thing.”

I’d like a little more information before handing over the budget.

AI Max can help an existing Search campaign reach searches your keyword list misses. But broader reach only earns its place in your account if it helps the business. Before I recommend switching it on, I want to understand your tracking, your landing pages, your brand controls, and how we’ll judge the test.

Let’s work through those decisions.

Key takeaways

• AI Max for Search is a suite of features inside an existing Search campaign: search term matching, text customization, and final URL expansion.
• You don’t need AI Max specifically to qualify for ads within AI Overviews. Broad-match Search, Shopping, Performance Max, and DSA can also provide eligible targeting.
• Campaigns using legacy automatically created assets or campaign-level broad match began migrating in September 2026. DSA auto-migration is scheduled for February 2027.
• Google’s Search performance benchmarks come from non-retail advertisers. Its separate Shopping benchmark isn’t a forecast for your ecommerce Search campaigns.
• I’d test AI Max against a stable baseline and evaluate total business results before expanding it.

What AI Max for Search actually changes

First, a useful distinction: AI Max isn’t another campaign type alongside Search and Performance Max.

It lives inside Search campaigns. Google introduced it in May 2025 and announced that it was moving out of beta in April 2026.

There are three main capabilities to understand:

Search term matching expands how Google finds relevant searches. It uses broad match and keywordless technology informed by your existing keywords, creative, and URLs. Turning on AI Max enables search term matching, which you can disable at the ad group level.

Text customization generates headlines and descriptions. This was previously called automatically created assets. Google uses information from your website and campaign to create copy relevant to the search.

Final URL expansion lets Google choose another relevant landing page on your domain. It requires text customization so the ad copy can match the page Google selects.

When you enable the full suite, you’re giving Google more discretion over the search, the message, and the destination. You can adjust the individual settings; you don’t have to treat the suite as one permanent, all-or-nothing decision.

My first question is which of those capabilities your campaign actually needs.

If your keyword coverage is strong but your landing pages need work, more automation won’t fix the underlying problem. If valuable searches are falling through the gaps, expanded matching may be worth testing.

Do you need AI Max to advertise in AI Overviews?

No. And I want to make that clear before we go any further.

Google says text and Shopping ads from eligible Search, Shopping, and Performance Max campaigns can appear within AI Overviews. To match those placements, it requires AI-powered targeting such as broad match in Search or the keywordless technology available through AI Max, Performance Max, Shopping, and DSA.

AI Max is one route to eligibility. A Search campaign using broad match can qualify without it.

Also, eligibility doesn’t guarantee an appearance. Google considers the query, the content of the AI Overview, relevance, and the auction. Certain sensitive verticals are excluded.

There are two placement situations to keep separate:

Placement	What Google currently supports
Above or below an AI Overview	Ads are eligible in the 200-plus markets where AI Overviews are available, under the existing auction system.
Within an AI Overview	Eligible text and Shopping ads, in English, on mobile and desktop in 12 countries.

Those 12 countries are Australia, Canada, India, Indonesia, Kenya, Malaysia, New Zealand, Nigeria, Pakistan, the Philippines, Singapore, and the United States.

Here’s the measurement catch: Google doesn’t offer separate reporting for ads within AI Overviews. You also can’t target only those placements or opt out of them.

So if someone tells you AI Max improved your AI Overview ROAS, I’d ask how they isolated that placement. The standard reports don’t give you that answer.

What about ads in AI Mode?

AI Mode has its own rollout and testing history. I wouldn’t treat it as interchangeable with AI Overviews.

Google announced testing ads below and integrated into AI Mode responses in May 2025. At Google Marketing Live in May 2026, it introduced tests for Conversational Discovery ads and Highlighted Answers, along with an expansion of the Direct Offers pilot that began in January 2026.

Google has identified Performance Max, Shopping, and broad-match Search campaigns, including AI Max, as eligible routes for its AI search advertising experiences. But these announcements don’t mean every advertiser can activate every format in every market.

My recommendation is to review the opportunities available in your account. I wouldn’t rebuild a profitable Search strategy around a placement you can’t independently measure.

Read Google’s performance numbers before using them in a forecast

Google’s headline AI Max statistics sound appealing: typically 14% more conversions or conversion value at a similar CPA or ROAS, rising to 27% for campaigns that relied heavily on exact and phrase match.

The footnote matters. Those figures come from Google’s internal data for non-retail advertisers. The 27% figure refers to campaigns where more than 70% of conversions or conversion value came from exact or phrase match keywords.

Google later reported an average 7% improvement for the full AI Max Search suite compared with search term matching alone, also using non-retail data. That’s a different comparison from turning AI Max on versus your existing setup.

For retailers, Google separately reports a typical 5% increase in conversions or conversion value at a similar CPA or ROAS for AI Max for Shopping. Google announced that product in April 2026, and its help documentation describes it as a beta. Check availability in your account.

These figures describe different products, populations, and comparisons. I wouldn’t use any of them to promise an ecommerce Search campaign a specific lift.

They give us a reason to investigate. Your test gives us a reason to invest.

The controls I’d review before launch

Before touching the toggle, I’d confirm that purchase tracking and conversion values are accurate. If Google is optimizing toward duplicate purchases or inflated revenue, giving it more reach makes that mistake more expensive.

Then I’d review these settings:

Control	What I’d check
Brand exclusions	Whether prospecting should exclude your own brand, competitors, or other unwanted branded searches.
Brand inclusions	Whether branded campaigns are restricted to the intended brands.
URL exclusions	Whether final URL expansion could send shoppers to unsuitable pages, such as careers pages or irrelevant content.
URL inclusions	Whether ad groups need tighter direction toward specific collections or product categories.
Locations of interest	Whether the geographic intent of searches fits the ad group. This is separate from campaign location targeting.
Negative keywords	Whether known irrelevant searches are excluded without unnecessarily blocking useful discovery.

I’d pay particular attention to your own brand. If the purpose of a test is to find new demand, I want to know whether the reported improvement came from reaching new shoppers or capturing more people already looking for you.

Google documents these controls in its AI Max setup guide.

There are a few dependencies worth knowing before you start:

• Manual CPC limits the feature set. Search term matching doesn’t work with Manual CPC. Google recommends conversion- or value-based bidding to use AI Max’s full capabilities.
• URL exclusions require text customization and final URL expansion. They control expanded destinations; they aren’t a substitute for reviewing the URLs you supplied yourself.
• Pinned copy needs special attention. Google says pinning isn’t respected with the relevant final URL expansion and URL inclusion settings. If approved wording must always appear, review those settings before launch.
• Turning off AI Max disables its brand exclusions. Recheck your safeguards when changing the main toggle.
• Tracking templates need to accommodate dynamic destinations. Review Google’s guidance on using {lpurl} rather than assuming a hardcoded destination will work with final URL expansion.

Your product feed deserves attention, too

Even if we’re discussing Search, I’d look at the Merchant Center feed as part of the broader ecommerce account review.

Start with accurate prices, availability, descriptions, shipping, returns, and useful product imagery. Then consider whether conversational attributes would help Google understand details that basic titles and descriptions miss.

Google supports attributes including question_and_answer, document_link, related_product, and popularity_rank. Its conversational attributes guide explains how to use them.

Google also reported that, during a lululemon test, submitted conversational attributes were incorporated in 50% of relevant product recommendations in AI Mode. That tells us the information was used. It doesn’t establish a conversion lift from those attributes.

Separately, Google says merchants adopting core feed best practices saw an average 5% increase in conversions the following month. I’d keep that distinction clear rather than combining the two findings into a bigger claim.

How I’d test whether AI Max is worth keeping

I’d define success before launch.

What does the campaign contribute today? What would count as a worthwhile improvement? How much room do we have to test without disrupting revenue we depend on?

Where appropriate, I’d use a Google Ads experiment against the existing setup. I’d keep other major changes limited so we have a useful comparison, and allow for the campaign’s learning period and conversion delay before drawing conclusions.

Then I’d review three things.

First, what traffic did AI Max add? The search terms report provides an AI Max match type and a Source column. The keywords report has two useful summary rows: “Total: AI Max expanded matches” and “Total: AI Max landing page matches.”

Second, what did shoppers see and where did they land? Review search terms and ad combinations, generated assets, and landing pages. The landing pages report’s “Selected by” column identifies URLs selected by AI Max.

Third, did the business improve? I’d look at backend revenue, total ad spend, MER, and new-customer results where reliable data is available. If the scale justifies it, a properly designed holdout can help assess incrementality.

Those questions are related, but they don’t answer the same thing. An AI Max reporting label tells us how traffic matched. It doesn’t tell us the ad appeared within an AI Overview, and it doesn’t prove the sale wouldn’t have happened otherwise.

This is consistent with how we approach measurement at Solutions 8. Our published guidance warns that remarketing and returning customers can make in-platform results look strong while the business loses momentum.

I want the account to look healthy because the business is growing, with enough context to explain why.

What I’d do with existing DSA campaigns

If you use Dynamic Search Ads, there is a deadline to plan around.

Google’s developer schedule says automatic migration runs February 1–28, 2027. It also says the ability to create new DSA ad groups will be removed during that transition. January 15 is a reminder notification date, rather than the creation cutoff.

Campaigns using legacy automatically created assets or campaign-level broad match began migrating separately in September 2026.

Before that DSA transition, I’d review the job each campaign performs. Is it finding profitable searches your keyword campaigns miss? Does it depend on particular pages? Is it adding revenue, or mostly collecting credit for demand captured elsewhere?

Two of our older case studies show why I wouldn’t make the same decision for every store.

For a spiritual jewelry business, we built a blog-targeted DSA campaign alongside a US Performance Max campaign and geographic testing. The published case study reports overall Google Ads ROAS above 250% and achievement of a $29 target CPA. It separately reports nearly three times the Google Ads attributed revenue for the DSA campaign. Those overall results shouldn’t all be credited to DSA.

For a cremation urn retailer, we shifted roughly 80% of spend toward Performance Max and kept approximately 20% in Search because PMax was the stronger-performing channel in that account. From June 3 through December 31, 2022, the account recorded $1,133,212 in conversion value on $209,612 in spend, a 5.41 ROAS, compared with 3.14 in the five months before the engagement.

Neither case tests AI Max. They show why the account’s results should guide the decision.

For a productive DSA campaign, I’d evaluate an AI Max migration with careful attention to targeting and landing pages. If another campaign structure serves the business better, I’d test that alternative. If the campaign contributes little, migration is a good moment to reconsider the spend.

What I wouldn’t do is assume the upgrade preserves performance simply because it preserves the campaign’s place in the account.

AI Max FAQ

Is AI Max a new campaign type?

No. AI Max for Search adds targeting and creative features to an existing Search campaign. You can adjust its individual settings.

Do I need AI Max for ads within AI Overviews?

No. Google also identifies broad-match Search, Shopping, Performance Max, and DSA targeting as eligible routes. Relevance, auction outcomes, market availability, and policy restrictions still apply.

Can I opt out of AI Overview placements?

Google says you can’t opt out or directly target only those placements. It also doesn’t currently provide segmented reporting for ads within AI Overviews.

Does AI Max work with Manual CPC?

Its search term matching feature doesn’t. Google recommends conversion- or value-based automated bidding for the full capabilities.

What happens to my DSA campaigns?

Google schedules automatic migration to AI Max for February 2027. Dynamic ad groups become standard ad groups; Google says the automatic upgrade enables all three core AI Max features and preserves legacy URL controls. I’d still review the resulting settings, ads, and destinations.

Let Solutions 8 help you decide what belongs in your account

My advice is to give AI Max a defined job before giving it more budget.

Check the tracking. Set the controls. Establish a baseline. Then test whether the additional reach produces results worth paying for.

At Solutions 8, we manage Search, Shopping, and Performance Max for ecommerce businesses. We can review your campaign and feed eligibility for Google’s AI placements, help plan your DSA transition, and build a test around your business goals.

If you’re unsure what AI Max should change in your account, start with a conversation about our Google Ads management. I’d rather make that decision with your numbers in front of us.