Google Ads for AI Search With AI Max and Performance Max

To prepare Google Ads for AI search, establish reliable conversion signals, choose a campaign approach that fits your goal, and test expansion against business outcomes. AI Max, broad match and Performance Max offer different ways to reach demand. Evaluate them using qualified leads or sales value, with clear budget limits and enough time for conversions to mature.

Google reports that the average AI Mode search is three times as long as a traditional Search query. Longer questions give buyers more room to describe their constraints and compare options. Evaluate how those search patterns relate to your offer and the buyers your campaigns already reach. [1]

A buyer might search for “CRM software for a 20-person services firm with onboarding support and monthly billing.” Your campaign and landing page need to address the buyer's requirements. More reach will have limited value if the offer or measurement cannot distinguish a suitable prospect.

How ads appear in AI Overviews and AI Mode

Google's Gemini app, AI Overviews and AI Mode are distinct experiences. The placement rules discussed here apply to AI-powered experiences in Google Search.

Eligible Search, Shopping and Performance Max campaigns can participate in AI Overviews and AI Mode. Google distinguishes direct query matching from contextual matching that considers the query and AI response. Exact and phrase match can qualify for direct commercial queries. AI-powered targeting, including broad match and AI Max, can also qualify for contextual matching. Availability varies by market and language. [2]

For advertisers, the practical decision is how much additional matching flexibility the business can measure and support.

Exact match still has a role in an AI search strategy

Exact match can match searches with the same meaning or intent as a keyword. It is not limited to an identical string of words. Broad match reaches a wider set of related searches, and Google recommends pairing broad match with Smart Bidding. [3]

Rebar's recommendation is to preserve useful controls while testing additional reach. Keep campaigns where targeting and economics are working. Investigate expansion when the account has reliable conversion data and a specific opportunity to test.

AI Max versus Performance Max and broad match

AI Max is a set of features within Search campaigns. It includes expanded search-term matching, text customization and final URL expansion. Performance Max is a separate campaign type that can run across Google inventory, including Search, YouTube, Display and other channels. Broad match is a keyword matching option. [4] [5] [3]

The starting points below are Rebar recommendations. The appropriate choice depends on the account's goals, data and constraints.

Choose the settings deliberately. Google documents that enabling AI Max turns on several features, some of which can be controlled separately. Record the actual configuration so the result can be interpreted. [4]

Build conversion signals that reflect lead quality

Choose the business outcome used for bidding

Review the campaign's conversion goals before adding targeting flexibility. Confirm what each action means and whether the campaign is optimizing toward the intended outcome.

For a lead generation account, map the journey from form submission to qualified lead, opportunity and sale. Select a stage that is meaningful, consistently recorded and available soon enough to guide bidding. Keep other stages visible for analysis without accidentally counting several milestones as independent wins.

Connect offline outcomes with enhanced conversions for leads

Enhanced conversions for leads supplements offline conversion measurement with hashed first-party customer information. The website and offline records help Google connect later outcomes to ad interactions. It improves matching; it does not decide whether a lead is qualified. Your business must define and record that outcome. [6]

Have the implementation owner document the CRM stage, conversion time, identifiers, deduplication method, values, currency and applicable consent handling. Validate a completed lead journey and monitor import diagnostics before using the new signal to judge campaign success.

Use value based bidding when the values are meaningful

Value-based bidding optimizes toward the conversion values you supply. Those values may represent revenue, profit or a defensible estimate of customer value. It does not infer reliable lifetime value from a form submission alone. Google's guidance emphasizes meaningful values, measurement readiness and regular data uploads. [7]

If qualified leads have similar value, start by measuring that outcome consistently. When lead values differ, use observed differences in close rate and deal economics to develop a value model. Test whether the model predicts actual results.

Unsure which signal your campaigns are learning from? Request a free Rebar Google Ads assessment to identify the account priorities and next test. Website tagging and CRM validation are separate scopes when needed.

A practical example of assigning lead value

Suppose two lead segments have different historical close rates. The figures below illustrate a calculation, not a Rebar client outcome or a recommended value for your account.

Expected revenue per qualified lead = observed close rate × expected revenue per sale.

On those assumptions, B has 2.5 times A's expected revenue per qualified lead. If margins differ, a profit-based estimate may be more useful. Check cohort maturity and sample size before using either estimate.

Label expected values as estimates in reporting. Keep the value basis consistent, and avoid counting both the full expected lead value and full realized sale value as separate revenue from the same customer. A forecast is useful only while its assumptions remain credible.

Run an experiment your team can interpret

Define the test before changing the campaign. A useful hypothesis might be: “Expanding Search matching will increase qualified leads within our acceptable cost per qualified lead.”

1. Choose one meaningful change. If several settings must change together, record the bundle and evaluate the combined result.
2. Set the commercial measure. Use qualified-lead cost, accepted pipeline or an appropriate value measure. Track raw leads as supporting evidence.
3. Set the budget and stop conditions. Define the exposure the business can support and the conditions requiring intervention.
4. Allow for learning and conversion delay. Use mature outcomes and sufficient data. Avoid declaring a winner from a short period or a few conversions.
5. Use a suitable comparison. Prefer an eligible campaign experiment when available. A simple before-and-after comparison can be affected by seasonality, demand and other changes.

For Performance Max, review brand demand and changes in other campaigns alongside total business results. A change in attributed conversions alone does not establish additional demand. If the result remains inconclusive, record that finding and the evidence needed for the next decision.

Get your next Google Ads test into focus

Rebar Performance supports your team or agency through Build, Diagnose and Improve: campaign planning, account analysis and measured experiments.

Active advertisers can start with a free review of up to 90 days of available account data. Receive a scorecard, up to three supported priorities and one test to consider. Provide read-only access or exports; the assessment does not change campaigns. Website tagging and CRM validation can be scoped separately. [8]

Request my free Google Ads assessment

Common questions about Google Ads and AI search

Do I need Performance Max to advertise in AI search

No. Google's eligibility guidance includes Search and Shopping as well as Performance Max. The available matching behavior depends on the targeting approach, with geographic and language limits. [2]

Can I target only AI Overviews or AI Mode

No. Google currently does not offer exclusive targeting or a separate opt-out for those Search experiences. Their ad performance is included in standard aggregate reporting. [2]

Will enhanced conversions for leads eliminate poor leads

No. Enhanced conversions supports measurement and matching. Lead quality also depends on your targeting, offer, qualification process and the outcomes supplied to bidding. [6]

Can value based bidding optimize for lifetime value

It can use supplied values based on a credible lifetime-value estimate. The estimate needs supporting customer data and ongoing validation. Calling a fixed lead value “LTV” does not make it a reliable prediction. [7]

Start with the account evidence in our Google Ads audit guide with a Gemini prompt and export checklist.