Google Ads AI Max September 2026 Upgrade: Agency Checklist

Direct answer

Google's updated AI Max reporting can connect a triggering search term with the served headline and landing page in one journey view. Agencies should preserve that row beside migration time, controls, cost, conversions and client outcomes, then review relevance and compliance before changing exclusions. The report improves observability; it does not prove incrementality, profitability or a competitor's private setup.

What changed in Google's September 23 AI Max update?

Google announced a unified Search-ads journey view for AI Max. Its current Help documentation names the view Search terms and ad combinations and says it can show the search term that triggered the ad, the headline that served and the landing page selected.

The announcement also expands the closed beta of AI Brief to Dutch, French, German, Italian, Japanese, Portuguese and Spanish. AI Brief lets an advertiser describe its business, audience and messaging in natural language. Google says more reporting details and availability information will come later in 2026.

For Indian agencies, the reporting view is the stronger operational development. Google does not announce Hindi or other Indian-language AI Brief support in this update, and AI Brief remains a closed beta. Do not promise client access. Verify the live account and preserve the interface label and review date.

What does the unified Search journey reveal?

Google's Help page says the Search terms report can be switched to Search terms and ad combinations, formerly called Search terms and landing pages from AI Max. That view can show:

• the user's search term;
• the headline that served;
• the selected landing page;
• campaign and ad-group context; and
• performance metrics available in the account.

This joins evidence that previously required separate reviews. An agency can inspect whether the query, generated or supplied headline, and destination tell a coherent story. It can also identify mismatches such as a research query routed to a hard-sell page, a regulated claim paired with the wrong disclaimer, or an expanded URL that breaks a tracking template.

The view is not a full causal explanation. It does not expose every auction signal, model decision or counterfactual. Google's Help page also warns that filtering the search terms report to match type AI Max may show lower counts because that filter does not include “Other search terms.” Record the filter and denominator before presenting coverage as complete.

Original artifact: AI Max journey evidence card

Use one card per material search-term, headline and landing-page combination. Preserve the review window and filters so another operator can reproduce the finding.

The card separates platform observation from business proof. A relevant search-term journey can still be unprofitable. A profitable cohort can still contain an inaccurate headline that requires removal.

How do the v25.1 migration timestamps improve the audit?

Google Ads API v25.1 added two output-only Campaign fields:

• aca_migration_date_time, for the automatically created assets migration; and
• broad_match_migration_date_time, for the campaign-level broad-match migration.

Google documents the values in the customer's timezone using yyyy-MM-dd HH:mm:ss. Agencies with authorized API reporting should preserve the timestamp and timezone beside the first unified-journey review. Teams without API access should retain the account notice, change history and dated settings export.

The timestamp establishes when Google recorded the migration. It does not prove when a performance effect began, whether every feature stayed enabled or why a result changed. Keep bid, budget, conversion, creative and website changes in the same change log.

Which reports and controls should an agency reconcile?

The unified journey view is one part of the AI Max evidence stack.

Search terms report: review terms, match type, source, headlines and landing pages. Google says an agency can add a negative keyword or negative URL from supported report rows.

Keywords report: inspect the aggregate rows for AI Max expanded matches and AI Max landing page matches. These summarize traffic sources; they do not identify a single causal mechanism for performance.

Asset reports: distinguish advertiser-supplied assets from customized text marked Google AI in the Added by column. Review associations, performance and current status. Google says unfavorable assets can be removed, while bulk removal of customized text requires Google Ads Editor or the API.

Landing pages report: review advertiser-provided and Final URL expansion destinations. The Selected by column can indicate an AI Max-selected URL. Check URL inclusions, exclusions, status codes, tracking templates, canonical handling and mobile rendering.

Campaign controls: reconcile brand inclusions, brand exclusions, negative keywords, location controls, URL rules, text customization and Final URL expansion. Google says Final URL expansion requires text customization, and RSA pinning is not respected when a more relevant expanded URL is chosen.

When should an agency act on a reported journey?

Google recommends waiting at least two weeks after enabling AI Max before routine changes such as adding negative keywords. Treat that as a learning-period guideline, not permission to leave a broken URL, inaccurate claim or material policy risk live.

Use three action lanes:

1. Immediate correction: broken destinations, tracking failures, false claims, wrong brands, regulated-content problems or material customer harm.
2. Evidence review: relevance concerns with limited volume, incomplete lag or unclear business quality. Monitor through a complete window.
3. Routine optimization: repeated underperformance with stable tracking, adequate volume and client-outcome evidence. Apply the smallest defensible exclusion or asset change.

Record the action owner and expected recheck date. Avoid blanket negative-keyword or URL exclusions that collapse useful reach based on one row.

How should Indian agencies handle multilingual journeys?

Google's separate September language-targeting change removes campaign-level language targeting from Search and from the Search Network portion of Performance Max. The India language-targeting checklist covers that migration in detail.

For this AI Max audit, add three multilingual checks:

• does the search-term language match a headline the client can substantiate;
• does the selected page provide the promised language, price, geography and offer; and
• can the agency review the claim accurately, including transliteration and mixed-language queries?

AI Brief's newly listed languages do not include Hindi. Do not infer support for an unlisted language or confuse AI Brief availability with the broader language-matching behavior of Search ads.

How does this differ from AI Max for Shopping?

The unified Search journey connects Search terms, headlines and landing pages. The AI Max for Shopping agency checklist covers a retail-specific stack: Merchant Center feeds, original and generated product titles, product IDs, Final URL expansion and Target ROAS requirements.

Keep the evidence registers separate. A Search headline-to-page check does not validate a product feed. A correct product title does not validate a Search campaign's keywordless match or generated text.

How can public competitor-ad research contribute?

Public Meta and Google ads can help an agency build a pre-campaign message and destination hypothesis. Record observable claims, offers, formats, advertiser identity, served dates and landing pages using the Google Ads Transparency Center workflow and the multi-platform ad-intelligence framework.

Then test the client's own journeys in the authorized Google Ads account. Advertisng's public ad-intelligence workspace can preserve public references and research notes for that hypothesis. It does not connect a competitor's search terms to headlines, selected landing pages or private performance.

What can public ad data not prove?

Public ad data cannot prove a competitor's AI Max status, migration timestamp, search terms, match type, source column, served headline combination, Final URL expansion setting, URL or brand controls, negative keywords, bids, budgets, targeting, spend, conversions, revenue, margin, profitability or incrementality. A public ad and destination can support an observable research record, but they cannot reproduce the private unified-journey report or explain why Google's system selected an outcome.

Frequently asked questions

Is the unified AI Max journey report available in every account?

Google's Help page documents the current Search terms and ad combinations view, while the September 23 announcement says additional details and availability will be shared later in 2026. Verify the account rather than promising universal access.

Does the report show every AI Max search term?

Do not assume it does. Google warns that filtering match type to AI Max excludes “Other search terms” from that filtered count. Preserve filters and compare totals before making a coverage claim.

Does a relevant headline and landing page prove good performance?

No. Relevance is a quality check. Performance needs a complete measurement window, conversion-lag treatment and client-owned business outcomes. Incrementality needs a suitable causal design.

Can agencies remove a generated asset?

Google's current Help documentation says customized text assets marked Google AI can be removed from asset reports. It says bulk removal is available through Google Ads Editor and the API rather than the Google Ads interface.

Does the migration timestamp prove that AI Max caused a result?

No. It records a platform migration time for the relevant legacy setting. Concurrent bid, budget, conversion, creative, seasonal and website changes still need to be controlled or disclosed.

A restrained next step

Export ten material Search-term, headline and landing-page combinations from one eligible client campaign. Complete the journey evidence card, correct only urgent defects immediately, and review routine changes after a complete learning and conversion-lag window. Use public competitor ads to sharpen the hypothesis, not to manufacture private AI Max evidence.

Primary sources

1. Google Ads & Commerce: New AI Max features: Build ad campaigns more efficiently, unified Search journey reporting, AI Brief language expansion and future availability caveat; published September 23 and reviewed September 24, 2026.
2. Google Ads Help: About reporting in AI Max for Search campaigns, Search terms and ad combinations view, reporting surfaces, filters, asset removal and two-week guidance; reviewed September 24, 2026.
3. Google Ads Help: How AI Max for Search campaigns works, controls, reporting fields, text customization, Final URL expansion and tracking guidance; reviewed September 24, 2026.
4. Google Ads API: Release notes, v25.1 migration fields; reviewed September 24, 2026.
5. Google Ads API v25: Campaign resource, output-only migration timestamp definitions and timezone format; reviewed September 24, 2026.
6. Google Ads & Commerce: Google's Dynamic Search Ads are upgrading to AI Max, September legacy-setting upgrades and February 2027 DSA timetable; reviewed September 24, 2026.
7. Advertisng: Live blog inventory, 25 RSS entries and 40 sitemap URLs reviewed September 24, 2026.