Google Search Console's New Generative AI Report: What It Shows (and What It Doesn't)

By Minel Gunesoglu, founder of Is My Brand in AI · Last updated October 10, 2026

TL;DR: Google's Search Console AI report (the generative AI performance report, launched June 3, 2026) shows how often your pages appear in AI Overviews and AI Mode. Google rolled it out to all websites worldwide on August 31, 2026. It reports impressions only, with no clicks, CTR, queries or position, and it excludes ChatGPT and Perplexity.

For two years, the biggest blind spot in AI search has been measurement: your content was showing up inside Google's AI Overviews, but Search Console lumped those appearances in with everything else, so you couldn't see them separately. That changed on June 3, 2026, when Google announced dedicated Search Generative AI performance reports in Search Console, one of the most-requested features in the tool's history. Here's exactly what it does, the real limits of this first version, and how to use it without over-reading the data.

What is the new Generative AI report?

It's a new section inside Search Console's Performance area. Google ships it as two separate reports, one for Search and one for Discover. The data reflects where your pages appear inside Google's generative experiences:

• In Search: AI Overviews (the AI-written answer box at the top) and AI Mode (Google's full conversational search). Google's documentation names those two surfaces and nothing else, and it leaves out Search Labs experiments.
• In Discover: the AI-powered features inside the Discover feed, in a separate report of its own.

In short, it isolates your presence in Google's AI surfaces so you can finally see it on its own, instead of guessing. The rollout is finished. Google's help documentation states that "as of August 31, 2026, we've rolled out these insights to all websites worldwide", and the Beta label is gone, so the report is no longer limited by country. (Checked against Google's own documentation on 19 September 2026. This page said "UK first, expanding later" until September 6, which was true at the June 3 launch and stopped being true on August 31.)

If the report shows no data for your site, Google's help page names two reasons, and your location is not one of them: your site hasn't had enough impressions in these AI features yet, or you have excluded your site from Google's generative AI features. That exclusion is a setting you control, so check it first.

One detail worth knowing: Google's documentation says the report draws on the Web search type from the Performance report, so this is a dedicated lens on search data rather than a separate collection of its own.

What it tracks (and what it doesn't)

This is the part to be clear-eyed about, because the first version is deliberately narrow. It reports impressions only (how often your pages were shown inside these AI features), broken down by pages, countries, devices and dates, with dates grouped by day, week or month. (The device dimension is available for Search only, not Discover.) You can export both the chart and the table. That's genuinely useful, but notice what's missing.

Two limits matter most. First, no clicks and no CTR: you'll see that you appeared, not whether anyone came to your site from it. Second, and bigger, no query data: you won't know which questions triggered your appearance. For a discipline built on understanding intent, that's a real gap. There is no position metric either, so you cannot tell whether your link led the answer or trailed it. Google has said it's still working with site owners on which insights are most useful and plans to add more metrics over time, so query and click data may follow. But for now the report answers "how often am I showing up?" and not much more.

Why this matters (even with the limits)

Step back from the feature details and the signal is bigger than the report itself: Google is officially measuring AI search visibility. When the world's dominant search engine builds a dedicated report for "how often you appear in AI answers," it confirms what this whole field has been arguing: that being present in AI-generated results is now a metric worth tracking, not a curiosity.

Practically, three things follow:

1. You finally get a free Google-side baseline. Before this, your AI Overview presence was invisible inside aggregate Search data. Now you can watch it as its own line.
2. It legitimizes the work. It's easier to justify investing in AI search visibility to a boss or client when Google itself reports on it.
3. It's a leading indicator. Even impressions-only data tells you whether your content is being pulled into AI answers at all, and which pages. That's a starting point for getting cited more often.

September 2026: Search Console is also where Google is testing payments

The same tool that measures your AI impressions is now where Google is testing paying for them. In a public policy post on June 18, 2026, Google wrote that "beyond news, we are piloting a new way to partner with websites whose content meaningfully contributes to the freshness and factuality of generative AI responses through the process of grounding."

That pilot now has an address. On September 14, 2026, Digiday reported that invited sites see an AI earnings widget in their Search Console dashboard, showing a monthly earnings figure, with money accruing when their content contributes significantly to answers in the Gemini app, AI Overviews and AI Mode. Google confirmed it as an early-stage learning pilot, testing how to reward high-quality content on top of the traffic and tools it already provides.

Keep the known and the unknown apart:

• Known: it is invitation-only, at least dozens of publishers have been approached, it reaches past news publishers, and the widget lives inside Search Console next to the reports above.
• Not known: how the payment is calculated, what the rates are, how many sites are in it, and whether it ever opens to everyone. Google's help documentation for it is not public, and publishers who have it describe the payout math as a black box.

There is nothing to apply for. The useful read is structural: Search Console is becoming Google's ledger for AI usage as well as its meter, and the qualifying signal Google describes, content that adds freshness and factuality to an answer, is the same thing that earns you a citation in the first place. Optimizing for one is optimizing for the other.

How to actually use it

With impressions-only data, the useful moves are simple but real:

• Find your AI-visible pages. Sort by impressions to see which pages Google's AI features surface most. Those are your strengths. Study what they do well and repeat it.
• Spot the gaps. Pages you expected to appear but don't are candidates for an AI Overview optimization pass: clearer answers up top, better structure, stronger sourcing.
• Watch the trend, not the day. A single day's impressions are noisy. Track the line over weeks to see whether your AI presence is growing as you publish and improve content.
• Segment by country and device. If you're strong in one market and absent in another, that's a content or localization signal.

This same impressions data also informs the opposite decision: whether to remove your site from these AI surfaces at all. If you are weighing that, our guide to opting out of Google AI Overviews uses exactly this report as the starting point.

Resist the urge to over-interpret. Without clicks or queries, this is a presence gauge, not a traffic or intent report. Use it to confirm direction, then do the actual visibility work elsewhere.

Where it falls short — and what fills the gap

The honest boundary of this report is that it only sees Google, and only impressions. Most of your buyers' AI questions don't happen in Google at all. They happen in ChatGPT, which OpenAI says has more than 1 billion weekly users, plus Perplexity and Gemini. Search Console will never show you any of that.

That's the gap dedicated AI-visibility tools fill: tracking your brand across every major engine, showing which sources get cited instead of you, and turning that into a to-do list. The free Search Console report and a cross-engine tracker are complements, not substitutes:

• Search Console (free): your impression presence inside Google's AI features.
• A GEO tracker: your mention and citation share across ChatGPT, Perplexity, Gemini and Google, with competitor comparison.

If you want to see the dedicated tools that cover the engines Search Console can't, we compare them in our best GEO tools for 2026 roundup, with the per-engine versions split out: ChatGPT brand monitoring tools for the engine most of your buyers use, and Gemini rank trackers for the standalone app this report explicitly excludes. And for a free first look at whether AI engines mention your brand at all, sample them by hand: ask each one your category questions and note whether your brand is named. Our guide to tracking brand mentions in ChatGPT walks through that manual baseline step by step.

What our own measurements show

Google's report is one first-party view. These figures are ours, each with its method and data published on the linked page.

• Microsoft's side counts citations, by page and by phrase. In the 91 days to 7 September 2026, Bing's AI Performance report counted about 89,000 citations of this site in answers from Microsoft Copilot and its partners. Four guides earned 94% of them, and the phrase that brought the most, "improve brand visibility in answer engines", brought 9,274. The record is in our AI visibility case study.
• Presence in one engine says little about the next. On the 40 questions where Google AI Overviews, ChatGPT, Gemini, Perplexity and Claude (API) each cited a source, all five cited the same page in 2, and at least two of them in all 40. That is round three of our Cross-Engine Citation Study.
• A source list changes between askings. Asked ten of those questions a second time, Google AI Overviews cited again 63% of the pages of its first answers (32 of 51). That count, in the same Cross-Engine Citation Study, is the measured case for watching the trend and not the day.

How to improve what shows up in the report

A measurement report only helps if you act on it. The levers that increase your impressions in Google's AI features are the same ones that earn citations everywhere:

• Lead every important page with a clear, self-contained answer in the first paragraph, the passage AI features lift.
• Structure for extraction: descriptive headings, short sections, comparison tables, a real FAQ.
• Earn genuine mentions across the web (Reddit, roundups, reviews), which correlate strongly with AI citations.
• Keep dated content fresh and your pages crawlable.

The full playbook for Google's AI answers is in our guide to AI Overview optimization, and the broader strategy across all engines is in how to rank on ChatGPT.

Frequently asked questions

Is the Generative AI report available to everyone?

Yes, worldwide, since August 31, 2026. Google's help documentation says it has rolled the report out "to all websites worldwide" and no longer labels it Beta. If yours shows no data, Google names two causes and your country is not among them: your site hasn't had enough impressions in these AI features, or you have excluded your site from those features. At the June 3, 2026 launch it was a UK-first beta, which is what this page described until September 6, 2026.

Does the report show clicks from AI Overviews?

No. This first version reports impressions only (how often your pages appeared in the AI features), not clicks, CTR or sessions. Click data may come later, but it isn't in the initial rollout.

Can I see which queries triggered my AI appearances?

No. The initial report does not include query data, only impressions broken down by page, country, device and date (day, week or month). There is no position data either. The missing query layer is the report's biggest limitation for now.

Does it track ChatGPT, Perplexity or Gemini?

No. Search Console only covers Google's own search surfaces (AI Overviews, AI Mode, and Discover's AI features). It does not cover the standalone Gemini app. To track your visibility in ChatGPT, Perplexity and Gemini, you need a dedicated cross-engine tool. See our GEO tools comparison.

Is Google paying sites for their content in AI answers?

Inside one invitation-only pilot, yes. Since September 14, 2026 it has been reported that selected publishers see an AI earnings widget in Search Console showing monthly earnings when their content contributes significantly to answers in the Gemini app, AI Overviews and AI Mode. Google calls it an early-stage pilot and has not published how the payments are calculated. There is no application process, and it is separate from the generative AI performance report described on this page.

Should I still use a paid AI-visibility tool if this is free?

For Google-only impressions, the free report is a fine baseline. But if you care about ChatGPT and the other engines (where most AI conversations happen), or you need clicks, queries, competitor share and recommendations, a dedicated tracker still does far more. They work best together.

---

The headline isn't really the feature. It's the validation. Google building a Generative AI report in Search Console is proof that AI search visibility has graduated from buzzword to a metric serious teams measure. Use the free report for your Google baseline, fix the pages that should be appearing and aren't, and pair it with cross-engine measurement for the full picture. We'll update this guide as Google expands the report (clicks and queries, hopefully). For the strategy behind moving these numbers, start with how to rank on ChatGPT.

Written by Minel Gunesoglu, founder of Is My Brand in AI. More about us.