GSC Generative AI Reports: What to Measure

Why This Matters: Google Just Gave Us AI Visibility Data

On June 3, 2026, Google announced something the SEO industry has been requesting since AI Overviews launched: an official way to see how your content performs inside AI-powered search features. The new Search Generative AI performance report in Google Search Console shows impressions, page-level data, countries, devices, and date granularity for content appearing in AI Overviews, AI Mode, and AI Overviews in Discover.

This matters because until now, site owners had zero visibility into whether their pages were being cited or surfaced in AI-generated answers. Third-party tools like Semrush and Ahrefs could show when AI Overviews appeared for tracked keywords, but they could not tell you if your specific URLs were the ones being referenced. Google’s new report closes that gap — partially.

The scale of the audience is staggering. Google confirmed that AI Overviews now has over 2.5 billion monthly active users and AI Mode has surpassed 1 billion monthly users. These are no longer experimental features. They are mainstream search surfaces that rival traditional results in reach.

What the New AI Performance Report Shows

The report lives inside Google Search Console alongside existing performance reports for web, image, and video search. Here is what the data includes:

• Impressions: How often URLs from your site appeared in generative AI features across Search and Discover. This tells you whether your content is being surfaced, but not how many people acted on it.
• Pages: Which specific URLs appeared within AI features. This is the most actionable metric — it tells you exactly which product pages, blog posts, or landing pages Google’s AI is citing or referencing.
• Countries: Geographic breakdown of your AI visibility. Useful if you sell internationally and want to know where AI answers are picking up your content.
• Devices: Desktop vs. mobile split for Search results. AI Mode, which is conversational, may skew differently than traditional search.
• Dates: Hourly, daily, weekly, and monthly granularity so you can track trends over time. The data appears to start from May 18, 2026.

The report covers three surfaces: AI Overviews in standard Search results, AI Mode (the standalone conversational search experience), and AI Overviews in Discover.

What Is Missing — and Why It Matters

The most notable gap is the absence of click data. Google is not telling you how many people clicked from an AI answer to your website. When asked about this, a Google spokesperson told Search Engine Roundtable: “We’re continuing to work with website owners to understand what insights will be most helpful to inform their strategies, and we’ll introduce additional metrics over time.”

This is a critical limitation. Impressions tell you your content appears in AI answers, but without clicks, you cannot measure whether AI visibility translates to website traffic or conversions. For ecommerce sellers, this means you can see which product pages Google’s AI surfaces — but you still cannot measure the ROI of those appearances directly through Search Console.

The workaround is to pair Search Console AI impression data with your own analytics. If you see a product page gaining AI impressions and a corresponding lift in direct or organic traffic around the same dates, you have a reasonable proxy for AI-driven visits.

The AI Opt-Out Toggle: Should You Block Your Content?

Alongside the performance report, Google introduced a toggle in Search Console that lets site owners decide whether their content can appear in generative AI Search features — AI Overviews, AI Mode, and AI Overviews in Discover.

Key details about the toggle:

• Sites that opt out will not receive traffic or impressions from generative AI features.
• Opting out does not affect traditional search rankings. Google explicitly stated the control “will not be used as a ranking signal for search results outside of these generative AI Search features.”
• For early UK testers, the setting takes effect on June 17, 2026. Before that date, configuration changes have no impact.
• The feature was developed in consultation with the UK Competition and Markets Authority, which required Google to give publishers control over how their content is used in AI features.

When Barry Schwartz polled SEOs about whether they would use such a control, 33 percent said they would block Google from using their content in AI features. However, for ecommerce and product sellers, opting out is almost certainly the wrong move. With 2.5 billion monthly AI Overviews users, removing your products from AI answers means removing yourself from the largest new product discovery channel since social media.

How to Use the Report for Your GEO Strategy

Once the report is available in your Search Console account, here is how to put it to work.

Identify your most AI-visible pages. Sort the Pages report by impressions to see which URLs Google’s AI features surface most often. These are your strongest AI citation candidates. Compare them against your top organic traffic pages — if they are different pages, your AI visibility strategy needs adjustment.

Track which product pages appear in AI answers. For ecommerce sellers, this is the most valuable data point. If your flagship products are not appearing in AI Overviews or AI Mode, you need to strengthen their structured data, entity signals, and citation-readiness. If they are appearing, double down on what is working.

Monitor trends over time. Use the date granularity to see if your AI impressions are growing, flat, or declining. Pair this with any changes you make to structured data, content, or schema markup. This creates a feedback loop for your GEO optimization efforts.

Compare country performance. If you sell in multiple markets, the country breakdown tells you where AI answers are picking up your content and where you have gaps. This can inform your internationalization and localization priorities.

Correlate with your own traffic data. Since Search Console does not provide clicks, match AI impression spikes against your Google Analytics or Shopify analytics data. A rise in AI impressions followed by a traffic lift — even if the source appears as “direct” — is a strong signal that AI answers are driving real visits.

Updated Google Guidance for AI Visibility

Google also released updated best practices for improving visibility in generative AI Search features. The key recommendations are:

• Provide unique, non-commodity content. Google emphasizes that AI features look for original information that adds value beyond what is available on dozens of other sites. For product sellers, this means detailed comparison content, original testing data, and experiential descriptions — not manufacturer copy.
• Organize your content clearly. Well-structured pages with clear headings, logical sections, and easy-to-parse content are more likely to be extracted by AI systems. This aligns with the citation-ready content approach that FirstShelf has been advocating.
• Create a good page experience. Fast loading, mobile-friendly pages with clean design continue to matter for AI visibility, just as they do for traditional rankings.
• Use high-quality images and video. Google’s AI features increasingly incorporate visual content, and pages with strong visual assets have an advantage in being selected for AI answers.

What This Means for the Future of AI Search Analytics

The Search Console AI performance report is a first step — and an important one. It signals that Google is treating AI search visibility as a legitimate analytics category, not just an experimental overlay on traditional search results. The UK-first rollout and CMA involvement suggest that regulatory pressure will continue to expand transparency for site owners.

For brands investing in GEO, this report finally provides a baseline metric: AI impressions. While it does not solve the attribution problem — you still cannot precisely measure how much revenue AI citations drive — it gives you a way to track whether your optimization efforts are increasing your presence in AI answers.

Expect additional metrics over time. Google has committed to expanding the report, and competition from Bing Webmaster Tools, which already has its own AI reporting features with more on the way, will create pressure to add click data and deeper insights.

Action Steps

1. Check your Search Console account for the new Generative AI performance report. If you do not see it yet, you are likely outside the initial UK rollout — but it will expand globally.
2. Once available, pull your first AI impressions report and identify which pages Google’s AI features surface most often.
3. Compare your AI-visible pages against your top organic traffic pages to find gaps between traditional SEO performance and AI search visibility.
4. Do not opt out of AI features unless you have a specific strategic reason. With billions of users across AI Overviews and AI Mode, the visibility opportunity far outweighs the risks for most brands.
5. Pair AI impression data with your own analytics to build a proxy for AI-driven traffic and conversions.
6. Update your product pages with unique content, structured data, and clear organization to increase your chances of being cited in AI answers.

Turn the report into a seller workflow

An impression is a discovery signal, not a diagnosis. Segment the report by page, country, device, query theme, and comparison window before changing content. Then use the marketplace seller GEO guide to identify fields you can actually improve, and the generative engine optimization framework to audit the evidence behind the pages gaining or losing visibility.

How FirstShelf can help

The new Search Console report tells you where your pages showed up inside AI Overviews and AI Mode — but impressions without clicks leave the harder question open: why some listings get surfaced and recommended while others never enter the answer. That is the layer the free FirstShelf GEO audit scores. It reads your listing the way an AI system does and grades the signals that drive selection: semantic density, structure quality, entity authority, and platform fit.

Used together, the two close the loop. Search Console shows which pages earn AI impressions; FirstShelf shows what to fix on the pages that don’t, and its dashboard tracks whether your visibility moves after each change. To see why your listings do — or don’t — make it into AI answers, run a free 60-second audit at firstshelf.ai.