The Search terms and ad combinations report for AI Max puts four things on one row: the query that triggered your ad, the headline and description shown, the landing page the user reached, and where the match came from. Its job is to let you spot mismatches, such as a cheap query landing on a premium offer, and decide whether to add a negative keyword or URL exclusion. As of 14 October 2026, Search Engine Land says it has appeared in some accounts, so check yours.
Key Takeaways
- The view links search terms, headlines and descriptions, landing pages, match sources and asset sources in one place, per Search Engine Land (13 October 2026).
- Google Ads Help describes it as the default view of the search terms report for AI Max, with other views toggled at top right.
- Match source tells you whether traffic came from AI Max expanded matches or from landing page and asset-based matching.
- Judge on conversions, cost per qualified lead and customer quality, not clicks.
- Google's guidance: wait at least two weeks before big changes, review every one to two weeks, and use negatives sparingly.
- Text customisation assets are labelled "Google AI" in asset reports, and cannot be removed in bulk in the interface.
What the report is and where it came from
Search Engine Land reported on 13 October 2026 that Google Ads had begun showing a Search terms and ad combinations report for AI Max for Search campaigns. It was first spotted by Google Ads specialist Natasha Kaurra, who shared it on LinkedIn. The article notes it had been seen in at least one account and that broader availability was unconfirmed.
It was not a complete surprise. On 23 September 2026, Google's Ads and Commerce blog announced a new reporting feature giving "a single, unified view of the Search ads journey": which search terms triggered ads, which assets the user saw, and where they landed, with more details and availability promised later in the year. The report you may now see is consistent with that description, although Google's post does not name it.
Google's reporting help page now documents the view. It says the Search terms and ad combinations view lists search terms with the headlines, landing pages, campaigns and ad groups in the customer journey, and is the default landing view, with other views available via a toggle.
One row, four questions: what did they search, what did we show, where did they land, why did we match?
The columns, one by one
| Column or view | What it tells you | What to check |
|---|---|---|
| Search term | The actual query | Is it a query you would pay for? |
| Match type / source | Whether AI Max expanded matching or landing-page and asset matching triggered it | Which mechanism produces your worst traffic? |
| Headline and description | The copy served for that query | Does it promise what the query asked for? |
| Landing page ("Selected by") | Which URL, and whether you or AI Max picked it | Is the page right for the intent and price point? |
| Asset source | Advertiser-supplied or Google AI generated | Are generated lines accurate? |
Per Google's help page, the keywords report also carries two aggregate rows at the bottom: "Total: AI Max expanded matches" (broad match keywords AI Max created from your keywords) and "Total: AI Max landing page matches" (queries matched through your landing pages or assets, outside your keywords). Those two rows are the quickest answer to "how much of my traffic is AI Max reach, and how does it perform?"
One caveat Google gives: filtering the match type for "AI Max" may show lower numbers, and the filter excludes "Other search terms," so do not treat a filtered total as the full picture.
For API users
Google's Ads API reporting guide documents search_term_view with a segments.search_term_match_source field (values include AI_MAX_KEYWORDLESS, AI_MAX_BROAD_MATCH and VERTICAL_ADS_DATA_FEED), an ai_max_search_term_ad_combination_view exposing search term, headline and landing page together, plus expanded_landing_page_view and final_url_expansion_asset_view. If you build dashboards, these are the resources to pull; check the page for current field names before you code, because they can change.
How to spot a mismatch
Search Engine Land's example is a basic car-wash query triggering an ad for a $699 ceramic coating package. The same pattern shows up in other verticals:
- A "free course" query landing on a paid programme page.
- A generic "shoes" query hitting a premium line.
- A job-seeker query matched to a hiring-services page.
- A brand-comparison query landing on a page that does not mention the competitor.
For each row, ask three questions in order:
- Intent fit: Would this person plausibly buy what the landing page sells?
- Message fit: Does the headline answer the query or contradict it?
- Economic fit: Could this query ever convert at your target CPA given the price on the page?
If the answer to the first is no, add a negative keyword. If the intent fits but the page is wrong, add a URL exclusion or an inclusion for a better page. If the message is off, fix the source assets or switch off text customisation for that campaign.
Mismatch hunting: the value of the report is catching the premium offer shown to a bargain query.
A weekly review routine
Google's help guidance suggests waiting at least two weeks after enabling AI Max before making changes such as adding negatives so the campaign can learn, then reviewing search terms and performance every one to two weeks, and using negatives only for terms that consistently underperform across campaigns. My routine fits that.
Week 0-2 (observe)
- Export the report. Do not add negatives yet unless a term is clearly harmful (wrong brand, offensive, wrong country).
- Note the share of spend coming from expanded matches versus landing-page matches.
Week 3 onward (act weekly, 30 minutes)
- Sort search terms by cost; look at the top 30 spenders.
- Mark each as keep, negative, or investigate.
- Sort landing pages by cost; check "Selected by" and conversion rate.
- Review asset reports for "Google AI" assets with spend; read each line.
- Log changes with dates in a change log.
Monthly
- Compare AI Max traffic with your keyword traffic on cost per qualified lead.
- Consider a one-click AI Max experiment if you have not proved the lift: Google says traffic and budget are split between a control with AI Max off and a trial with it on, without copying the campaign. Some campaigns are ineligible, such as those with Display network targeting or shared budgets.
Decision rules I use
| Pattern in the report | Action |
|---|---|
| High-cost term, zero conversions after meaningful spend, off-intent | Add as negative |
| Good term, wrong landing page | Add URL exclusion for the bad page; check inclusions for the right one |
| Good traffic, weak copy | Edit source assets; review text customisation |
| Expanded matches convert at or near target | Leave alone; resist over-pruning |
| Landing-page matches burn spend on pages you did not mean to advertise | Add URL exclusions at campaign level |
| Low-volume odd terms | Do nothing; wait for more data |
"Meaningful spend" depends on your CPA. A rule of thumb I use is two to three times target CPA before calling a term a loser, but that is my own heuristic, not a Google rule.
Cautions
- Pinned assets: Google notes that pinned RSA assets may not be respected when final URL expansion is on. If compliance requires fixed wording, check this first.
- Tracking templates: Google says to use {lpurl} and test dynamic landing pages, since incompatible templates can cause errors.
- Bulk removal: Google says you cannot remove Google customised text assets in bulk in the interface, but you can through the Ads API and Ads Editor.
- Keyword prioritisation: Moving to AI Max or turning off broad match changes how broad keywords are prioritised; add exact copies if you rely on that behaviour.
- Budget-limited campaigns: Google says AI Max will not be effective if campaigns are limited by budget.
Details are in Google's AI Max explainer. For background on the ongoing migration, Google's DSA to AI Max announcement says automatic upgrades for ACA and campaign-level broad match campaigns begin in September 2026 and the DSA sunset moved to February 2027.
What this means for SEO and content
The report is a free window into real query language and which of your pages Google considers relevant. I use it three ways: as a source of new content ideas, as a check on which landing pages Google selects for broad intents (if it picks a thin page, organic is likely to as well), and as a copy test bed. For the paid-organic overlap in Google's newer AI surfaces, see my notes on AI Mode ad slots and AI Max with Conversion Lift.
FAQ
What is the Search terms and ad combinations report?
It is a view in Google Ads for AI Max for Search campaigns that links search terms to the headlines, descriptions and landing pages shown. Google's help page says it is the default view in the search terms report, with other views available at top right.
Is the report available in every account?
Not confirmed. Search Engine Land reported on 13 October 2026 that it started appearing in some accounts and had been spotted in at least one. Google's reporting help page documents it, but check your own account.
How do I know if a match came from AI Max?
Use the match type and source information. Google's help says you can add the Source column to the search terms view, and the keywords report shows aggregate rows for AI Max expanded matches and landing page matches.
Should I add negative keywords immediately?
Not usually. Google's guidance suggests waiting at least two weeks after enabling AI Max, and using negatives sparingly for terms that consistently underperform. Exceptions are clearly harmful terms.
What is "Selected by" on the landing pages report?
It shows whether AI Max chose the URL. Use it to confirm Google-selected URLs fit your needs, then set URL inclusions or exclusions accordingly.
Can I delete AI-generated headlines?
Individually, yes, by removing the asset. Google says you cannot remove Google customised text assets in bulk in the interface, but you can through the Google Ads API and Google Ads Editor.
What metrics should I judge AI Max on?
Conversions, cost per qualified lead and customer quality, rather than clicks. Search Engine Land makes this point in its report, and it matches Google's emphasis on checking relevance and value.
Does the report work for Shopping campaigns?
Google has a separate help page for reporting in AI Max for Shopping campaigns. This post covers AI Max for Search; check Google's Shopping page for its specifics.
How often should I review it?
Every one to two weeks, per Google's guidance, with a deeper monthly comparison against your keyword-based traffic.
Can I test AI Max safely before committing?
Yes. Google describes AI Max experiments that split traffic and budget between a control and a trial inside an existing campaign. Check eligibility first.
Need a second opinion on your account?
If you would like help turning this report into a routine, you can see my work and contact me through the form at younusfardeen.in. I have 4+ years of marketing experience across SEO, AEO and paid-search-adjacent content work, and I am glad to review your review process.