Google Just Gave You AI Visibility Data — Here Is What to Measure First
Google Search Console now shows AI Overview and AI Mode impressions. Here is what to measure first, where the data falls short, and the proxy metrics that actually connect AI visibility to pipeline.
On June 3, Google added dedicated Gen AI performance reports to Search Console — the first time site owners can see how their content appears inside AI Overviews and AI Mode as a metric separate from organic search. If you run marketing for a B2B company, this changes what you can measure this week. But the reports have a critical gap that most coverage has missed, and if you do not build a proxy metric around it, you will mistake visibility for value.
What the New Reports Actually Show
The Gen AI performance reports track five dimensions: impressions inside AI features, which specific URLs surfaced, geographic breakdown, device split, and date granularity across daily, weekly, and monthly views. Coverage extends to both Google Search and Discover.
That is the good news. Here is the bad news: there is no click attribution. An impression inside an AI Overview or AI Mode response does not confirm whether a user clicked through to your site. Google gives you the numerator — you appeared in the answer — but not the denominator that matters to your pipeline.
The rollout is also staged. Google released these reports to a subset of websites with no disclosed timeline for broader availability. Check your Search Console account weekly. If you do not see the Generative AI section yet, you are in the waitlist.
Why Impressions Without Clicks Are Still Useful
The instinct is to dismiss impression-only data. That is wrong. Here is what the number actually tells you:
It tells you which content Google's AI systems consider citable. That is a fundamentally different signal from traditional ranking. Research from early 2026 shows the decoupling is real — citations that came from top-10 ranked pages dropped from 76% in mid-2025 to roughly 38% by early 2026. Ranking number one no longer guarantees you appear in the AI-generated answer.
It tells you where attention is going even when clicks are not. Analysts estimate AI Mode's zero-click rate at roughly 93%. But sites cited inside an AI Overview see 35% more organic clicks than they would from a standard position-one result. Being in the overview matters more than ranking below it.
It gives you a baseline for the proxy metric you need to build. More on that below.
The Three Measurements That Matter This Week
Do not wait for Google to add click data. Build the proxy now.
1. AI impression-to-traffic ratio. Pull your top 20 pages by AI impression volume from the new reports. Compare each page's AI impressions to its direct and organic referral traffic across the same period. Pages with high AI impressions and zero corresponding traffic lift are feeding answers without driving visits. That is your content-to-pipeline leak, and you need to see it before you can fix it.
2. Content-type audit. Track which content formats generate AI impressions. Early data indicates that structured comparisons, listicles with named entities, and pages with clear answer-first formatting perform disproportionately well in AI features. If your highest-impression pages are all one format and your investment is in another, reallocate.
3. Cross-engine comparison. Google is not the only AI engine citing your content. Forrester reports AI-generated traffic represents 2–6% of total organic traffic and is growing over 40% per month. Google's new reports cover Google only. If you are not also tracking citations in ChatGPT, Perplexity, and Gemini separately, you are measuring one channel and calling it a strategy.
What This Means for Your AI Visibility Stack
The Gen AI reports give you the Google layer. But the measurement problem is bigger than one platform.
The UK CMA ordered Google to give publishers clearer attribution in AI-generated search results on the same day these reports launched — June 3. Regulatory pressure is moving toward mandatory attribution data. The direction is clear: measurement is coming whether platforms volunteer it or not.
For B2B teams specifically, the urgency is higher. LLM-referred traffic converts at 30–40% according to VentureBeat's analysis — and most enterprises are not optimizing for it. Before a buyer sees your homepage, an answer engine has already summarized your category, named a few vendors, and shaped the shortlist. The new Search Console data tells you whether Google's AI is including you in that summary. The rest of the stack — multi-engine citation monitoring, Machine Relations Index scoring, entity authority tracking — fills the gap Google left.
The Execution Checklist
If you are a CMO or growth lead reading this on a Wednesday, here is exactly what to do:
- Check access. Open Search Console. Look for the Generative AI section. If it is not there, set a weekly reminder to check.
- Pull your top 20. Sort by AI impression volume. Cross-reference against GA4 traffic for those same URLs over the same window.
- Flag the leakers. Pages with high AI impressions and flat or declining traffic are being cited without converting. These need structural changes — better CTAs inside the content, stronger entity attribution, clearer next-step architecture.
- Audit your format mix. Compare AI impression rates by content type. If structured, answer-first pages dominate and your production calendar is weighted toward long-form thought leadership, you have a misallocation problem.
- Extend beyond Google. Set up monitoring for ChatGPT, Perplexity, Gemini, and Claude citations on your top 20 pages. Google's new data is one instrument in the panel, not the whole dashboard.
FAQ
What exactly are the Gen AI performance reports in Google Search Console?
They are a new set of reports launched June 3, 2026 that show how often your URLs appear inside AI Overviews, AI Mode, and generative AI features in Discover. They track impressions, pages, countries, devices, and dates — but not clicks. Rollout is staged and not yet available to all sites.
Can I measure clicks from AI Overviews in Search Console now?
No. The Gen AI reports provide impression data only. Google has separately confirmed that Search Console includes clicks from AI Mode in its standard performance reports, but the dedicated Gen AI section does not break out click attribution. Build a proxy metric by comparing AI impression volume against traffic data for the same pages.
How should I track AI visibility beyond Google?
Google's reports cover Google only. To track visibility across ChatGPT, Perplexity, Gemini, and Claude, you need a multi-engine monitoring approach. Citation frequency, share of voice across platforms, and cross-engine divergence analysis give you the full picture. A page can be invisible to Google's AI while being cited heavily by Perplexity, or vice versa.