How Is AI Changing Digital Marketing? What the AI Max Data Shows
Google’s AI Max for Search expands how campaigns can match queries and serve ads. For advertisers accustomed to managing keywords and other targeting controls directly, the change raises practical questions: how much additional conversion volume might automation bring, and at what cost?
Mike Ryan of Smarter Ecommerce examined more than 600 Google Ads accounts in The Ultimate Guide to AI Max, as reported by Search Engine Land. In Ryan’s analysis, the accounts saw a 13% increase in conversions and an average 16% increase in cost per acquisition (CPA). Google, by comparison, says AI Max can deliver 14% more conversions at a similar return on ad spend (ROAS). These are different claims from different sources; the figures in Ryan’s analysis should not be read as a guarantee of what any individual account will experience.
Weigh Conversion Growth Against Cost
Ryan’s reported results show why conversion volume alone is not enough to assess a change in campaign performance. A higher conversion count alongside a higher average CPA may be worthwhile for some advertisers, but the effect on profitability depends on each business’s margins, conversion values, and goals. The reported averages do not establish why CPA changed or predict the outcome for a particular account.
Before expanding use of AI Max, advertisers can compare results with an appropriate baseline and decide which outcomes matter to the business. Conversion volume, CPA, and ROAS can provide useful context, but they should be interpreted alongside conversion quality and value. These are practical evaluation steps, not findings attributed to Ryan’s analysis.
Assess Incrementality and Query Quality
A rise in reported conversions does not, by itself, show how many were incremental. Advertisers evaluating AI Max may want to examine available search-term and campaign reporting, compare performance with their existing approach, and consider whether the additional conversions represent valuable demand. The results reported by Ryan do not establish that AI Max routinely captures existing brand traffic or overlaps with other campaign types.
Negative keywords and search-term reviews may be useful tools for managing query coverage, depending on an account’s setup and available controls. They should not be treated as proof that automated campaigns are spending on irrelevant queries. A relevant resource on this topic is a guide to negative keywords in Google Ads.
Review Campaign Structure in Context
Campaign structure is another area to review when testing automation. The source findings cited here do not establish that account segmentation reduces performance, that AI Max necessarily overlaps with Performance Max or broad match, or that campaigns compete against one another in auctions and raise CPCs. Those outcomes should not be assumed without account-specific evidence.
Advertisers can instead inspect their own campaign settings, reporting, and results to determine whether their structure is helping them evaluate performance. Any consolidation or separation of campaigns should be based on the account’s objectives and observed data, rather than on a universal rule about how automated systems perform.
Test Changes Before Scaling
Ryan’s account-level analysis offers a point of reference, not a forecast for every advertiser. A measured test can help businesses understand how AI Max performs in their own circumstances before they make broader changes. The test design should make the comparison meaningful and account for factors that could affect results; a short-term change in CPA alone does not explain its cause.
Google has also described changes to AI Max reporting and controls. Advertisers can review relevant platform updates, including Google’s AI Max transparency and multilingual AI Brief controls, to understand what information and options are available. Reporting capabilities can change, so consult current Google documentation when planning a test.
Set Evaluation Criteria for AI Automation
The findings Ryan reported—13% more conversions and an average 16% higher CPA across the accounts he analyzed—suggest that volume and cost should be considered together. They do not show that every advertiser will see the same results, nor do they establish the causes of those changes.
Before adopting or expanding AI Max, define the outcomes that matter, establish a suitable comparison, and monitor performance over a period appropriate to the account. Review conversion value and quality as well as volume and cost, and use available reporting to investigate changes. This approach keeps decisions grounded in an advertiser’s own results rather than treating either Google’s stated benchmark or an external account average as a guarantee.
