Guide · Google Search Console

What Google Search Console's AI report shows, and what it doesn't

Google Search Console's AI report shows where your pages were displayed inside Google's AI features, not whether your business is winning across AI assistants.

By Published 11 min read

Direct Answer

Google Search Console's AI report shows where your pages were displayed inside Google's AI features, not whether your business is winning across AI assistants. Since 31 August 2026, Google has reported Generative AI impressions by page, country, date and device inside Search Console, covering Google AI Overview, Google's AI answer box, and AI Mode, but it does not show clicks, does not show search queries, and does not separate those two features, according to Google Search Console documentation and product rollout notes from August 2026. That makes it useful for spotting which pages are gaining visibility in Google's AI answers, but not for proving business impact on its own.

Search Console's AI report is a Google visibility report, not a full AI visibility report

If you want to know whether your content appears in Google's AI answers, this report matters. If you want to know how visible you are across AI assistants, it is only one part of the picture.

Google rolled out the Search Console Generative AI performance report worldwide on 31 August 2026. Google says it reports impressions inside AI Overviews and AI Mode by page, country, date and device. Google also says the report does not include clicks, does not show search queries, and does not split AI Overviews from AI Mode.

In plain English, it tells you this: your page was shown somewhere inside Google's AI answer experience. It does not tell you which exact question triggered that appearance. It does not tell you how many people clicked through. It does not tell you whether the user saw a summary, a link card, or another presentation.

That is why this report is best used as an early visibility signal. It helps you see whether Google's AI systems are surfacing your pages at all. It does not replace your wider measurement plan.

The report is useful because it shows page-level AI visibility inside Google

The biggest gain is simple. You can finally see which pages are getting shown in Google's AI features.

You can identify which content Google is surfacing

The report breaks out impressions by page. That means you can spot which buying guides, product pages, category pages, comparison pages, local pages, and help articles are appearing in Google's AI answers.

For a marketing lead, this helps answer a practical question: which parts of the site are actually earning AI exposure.

If your pricing page starts to collect AI impressions, that matters. If your alternatives page does, that matters too. If only blog posts appear, but your commercial pages do not, that tells you where the gap is.

You can compare countries and devices

Google includes country and device filters. That helps if you sell in more than one market or if mobile matters more than desktop for your customers.

A merchant can check whether product-related pages are surfacing more often in Ireland than in the UK. A B2B software firm can compare desktop-heavy visibility against mobile. A local business can watch whether service pages are gaining AI impressions in its target area.

You can track change over time

Date filtering lets you see whether visibility is rising, falling, or staying flat. That makes the report useful after a content update, a site migration, or a new page launch.

It is especially helpful when paired with a monthly review. If your AI impressions rise after you publish stronger answer-led pages, add named authors, improve your review presence, or launch honest comparison content, you have a directional sign that Google is surfacing more of your site.

That does not prove causation on its own. It does help you decide where to look next.

The report has hard limits, and those limits matter for business decisions

The report is useful, but it leaves out several things decision-makers usually want to know.

It does not show clicks

Google's Search Console Generative AI report does not report clicks, according to Google's August 2026 documentation. So you cannot use it to say how much traffic Google's AI features sent to your site.

This matters because AI visibility and website visits are not the same thing. A page can be shown often and still drive little traffic. Another page can be shown less often but influence higher-value visits later through branded search, direct visits, or sales conversations.

It does not show the search queries

You do not get the exact prompts or searches that caused the impression. That means you cannot open the report and see, for example, whether your page appeared for "best payroll software for Irish SMEs" or "how to choose a fire door supplier".

Without query data, you lose context. You know a page appeared, but not the wording, intent, or buying stage behind it.

It does not separate AI Overview from AI Mode

Google combines both features in one report. So you cannot tell whether a page's visibility came from Google's AI answer box or from AI Mode.

That matters because these experiences can behave differently. One may expose your brand to a broad audience. The other may show up deeper in a research journey. If they are blended together, you need to be careful about over-interpreting trends.

It does not cover other AI assistants

This is the most common mistake. A rise in Google AI impressions does not mean you are also visible in ChatGPT, Copilot, Perplexity, Claude, or other AI assistants.

That gap matters more now because ChatGPT search has run on OpenAI's own index since 4 September 2026, with Bing used mainly in Deep Research, according to HelloWorld's reviewed methodology based on OpenAI and industry reporting from September 2026. In plain terms, being in Bing's phone book no longer guarantees visibility in ChatGPT search.

If you want the wider picture, you also need Bing Webmaster Tools' AI Performance report, which entered public preview on 1 February 2026 and added query-to-page mapping on 23 March 2026. Microsoft says it reports total citations, average cited pages, grounding queries, and page-level citations. It is a sample, not a click report.

For a fuller measurement framework, see How to measure your AI search citation performance.

The report is best used as a page-priority tool

The strongest use of this report is not vanity reporting. It is page prioritisation.

Use it to find pages that deserve more investment

If a page already earns AI impressions, Google has given you a signal. The page is relevant enough to surface. That often means it is worth improving first.

You might add clearer answer sections. You might tighten product facts. You might add pricing detail, delivery detail, returns detail, or named expert input. You might strengthen the page with current reviews, comparisons, or supporting proof.

The goal is not to rewrite copy "for AI". Google's own generative AI optimisation guidance says that rewriting content specifically for generative AI is unnecessary, and that content chunking is not required because its systems understand multi-topic pages. The job is to make the page more useful, more direct, and easier to trust.

Use it to spot missing commercial pages

If your informational pages appear but your money pages do not, that is a business issue. It often means your site answers questions, but does not yet support recommendation and comparison use cases strongly enough.

For many established firms, the missing pages are predictable. Pricing pages are weak. Alternatives pages do not exist. Comparison pages are too sales-led to be credible. Product or service pages lack plain-English facts. Author pages are missing. Reviews are thin.

These are exactly the assets that help AI assistants understand who you are, what you sell, and why someone should consider you.

Use it alongside your review and comparison footprint

HelloWorld's current methodology treats honest comparison pages, review platform presence, roundup placements, YouTube explainers, and business profiles as high-weight levers. Search Console will not tell you whether those off-site signals improved your standing. It may show that certain pages gained AI impressions after your wider visibility work improved.

That is why the report is best read in context, not in isolation.

If you are weighing in-house work against specialist support, see AI SEO agency vs DIY: which is right for your business?.

You should not treat AI impressions as your main KPI

Impressions are a signal. They are not the final score.

A page that appears often in Google's AI features may still produce little pipeline, few enquiries, or low-quality traffic. Another page may have modest AI exposure but influence high-intent buyers.

The better approach is to stack your evidence.

Pair the report with analytics and sales signals

Google's report tells you visibility inside its AI features. Your analytics platform tells you what happened on your site afterwards. Your CRM tells you whether leads improved.

Google Analytics 4 added a default AI Assistant channel on 13 May 2026 for referrals from ChatGPT, Gemini, DeepSeek, Copilot and Grok, according to Google Analytics release notes from May 2026. Google also says this channel excludes Google AI Overviews and AI Mode, is not retroactive, and does not include every assistant on the published list. So even that number is a floor, not a complete answer.

For Google specifically, the Search Console report is the visibility layer. Your business metrics still sit above it.

Watch for page-level patterns, not single-day swings

AI features change often. Presentation changes. query mix changes. User behaviour changes. A small rise or dip on one day is rarely meaningful.

Look for repeated patterns over weeks and months. Which page types keep appearing. Which countries are improving. Which commercial pages remain absent. Which updates are followed by sustained movement.

The report does not remove the need for technical checks

Even though the report is about visibility, access still matters. If Google's AI readers cannot reach or trust your pages, your reporting will stay thin.

Your AI welcome sign still matters

Your AI welcome sign, robots.txt, is the small file that tells AI readers whether they are invited in. For Google, this remains a basic hygiene check. For other assistants, access can also be affected by hosting or security settings.

That is especially important because Cloudflare started blocking its Training and Agent crawler categories by default on ad-bearing pages for new zones on 15 September 2026, while Search stayed allowed, according to Cloudflare product documentation from September 2026. ChatGPT-User is classed as an Agent. In plain terms, a site can look open in its welcome sign but still return a blocked response at the network level for some AI readers.

This change does not directly alter Google's Search Console report, but it is a good reminder that visibility reports only reflect what systems could access and use.

Your machine-readable business card is still hygiene, not the main lever

Your machine-readable business card, schema markup, is still worth keeping tidy because it is cheap and checkable. But HelloWorld does not treat it as a proven growth lever for AI visibility on its own.

If you need the detail, read What is schema markup, and does it matter for AI search? and Does schema markup help with AI search visibility?.

IndexNow helps Bing and Copilot, not ChatGPT search

IndexNow is a fast update signal for Bing and services that rely on Bing. It is not a ChatGPT lever now that ChatGPT search runs its own index, according to HelloWorld's reviewed methodology updated 4 September 2026.

That matters because some teams still assume one submission method covers every assistant. It does not. For the background, see What is IndexNow, and which AI assistants does it help?.

A practical way to use the report each month

You will get more value from this report if you review it on a fixed rhythm and combine it with a few other checks.

  • Open the Search Console Generative AI report once a month and export page-level impressions.
  • Highlight the pages with the biggest gains and losses over the last 28 days and the last 90 days.
  • Group those pages by type, such as product, category, pricing, comparison, local service, help article, or blog.
  • Check whether your commercial pages are appearing, not just your informational pages.
  • Compare country and device patterns if you sell in more than one market or rely heavily on mobile.
  • Review whether pages with AI impressions have clear answers near the top, current facts, named authors, and honest dates.
  • Check your review footprint on the platforms your buyers actually use, such as G2, Capterra, Google Business Profile, or merchant review systems.
  • Review Bing Webmaster Tools' AI Performance report for citation and grounding-query clues beyond Google.
  • Check your AI welcome sign and hosting or security settings so important AI readers are not blocked by accident.
  • Record business outcomes beside visibility data, such as branded search lift, qualified leads, demo requests, or revenue from the pages that gained exposure.

The right conclusion is modest but useful

Google Search Console's AI report is worth using. It gives you first-party evidence that your pages are appearing inside Google's AI experiences. That is valuable.

But it is not a scoreboard for all AI assistants. It is not a click report. It is not a query report. It is not a revenue report.

Use it to answer three grounded questions. Which pages is Google surfacing. Which important pages are missing. Which page types deserve your next round of work.

If you keep the interpretation that simple, the report becomes genuinely useful. If you ask it to do more than it can, it will mislead you.

Does the Search Console AI report show traffic from Google's AI answers?

No. Google says the Generative AI report shows impressions, not clicks. It tells you that your page was shown inside AI Overviews or AI Mode, but not how many visits followed.

Can I see which search queries triggered my page in the AI report?

No. Google does not provide query data in this report. You can see the page, country, date, and device, but not the exact searches or prompts behind the impression.

Does this report tell me how visible I am in ChatGPT or Perplexity?

No. This is a Google-only report. It does not measure visibility in other AI assistants. For broader tracking, you need other sources such as Bing Webmaster Tools' AI Performance report, analytics, and manual testing.

Should AI impressions be my main KPI?

No. Treat them as an early visibility signal. Your main KPI should still connect to business outcomes, such as qualified traffic, leads, sales conversations, or revenue influenced by the pages gaining exposure.

What should I do if only blog posts appear in the report?

That usually means your site is answering questions but not yet supporting recommendation and buying use cases strongly enough. Review your pricing, alternatives, comparison, product, category, and service pages first. Make them clearer, more factual, and easier to trust.

Next step

Find out where you stand in AI answers

The AI Visibility Audit tests your buyers’ real questions across five AI assistants, shows who gets cited instead of you, and ends with a phased 90-day plan.