Google Just Launched AI Performance Reports in Search Console

boliviainteligente-Cr50M5NNV1U-unsplash

On June, 3 Google announced the launch of Search Generative AI performance reports inside Search Console, with dedicated views for Search and Discover. The reports show how often your URLs appear inside AI Overviews and AI Mode, with breakdowns by impressions, pages, countries, devices, and dates.

The rollout is going to a subset of websites first, so most accounts haven’t seen it live yet. It’s on the way.

What the New Reports Show

According to Google’s announcement, the dedicated views include:

  • Impressions inside AI features, which shows how often URLs from your site appeared in generative AI surfaces in Search and Discover. 
  • The pages view shows which specific URLs are surfacing in AI features. 
  • The countries view breaks visibility out by geography, which matters because AI Overviews behave differently market by market. 
  • The devices view is available for Search results and shows the split between mobile and desktop. 
  • And the dates view offers hourly, daily, weekly, and monthly granularity for tracking performance over time.

The data also continues to flow into the overall performance report, so the total visibility number teams already track will now include generative AI impressions. The dedicated view is the new layer on top of that.

Why This Matters

For the first time, AI visibility on Google becomes a measurable number inside the dashboard SEO teams already use every day.

Until now, conversations about AI Overviews visibility have relied on third-party tools, sampled data, and educated estimates. SEO leads have been telling executives “we think we’re showing up in AI Overviews for these queries” without a primary source to point to.

Now there’s a primary source. The same Search Console most teams open every morning.

That changes a few things in practice. 

  • It makes AI visibility a line item that can be reported on, resourced, and compared over time. 
  • It surfaces the relationship between traditional ranking performance and AI inclusion, which is information teams haven’t had access to before. 
  • And it gives content strategy a feedback loop that’s been missing since AI Overviews started rolling out.

The Search Behavior is Also Changing

Here’s a pattern the new reports are going to make impossible to ignore.

Search queries inside Google Search Console are getting longer and more conversational, with structures that look closer to how people prompt an LLM than how people used to type into a search box. 

Previsible’s data shows that more than 65% of queries on some client sites are now eight words or longer and follow a clear prompt-like pattern, with full questions, specified constraints, and the kind of context the user is feeding the search engine the same way they’d feed it to ChatGPT.

This is behavior bleeding back into Google from how people have learned to interact with AI tools. 

The new Search Console reports will make clear which of those longer, prompt-like queries are pulling content into AI Overviews and AI Mode and which aren’t. 

That’s useful information for any team thinking about how to structure content for the queries buyers are actually running today.

What to Do Before the Reports Land

You don’t have to wait for the dashboard to start preparing for what it’s going to show.

Pull a query report from existing Search Console performance data and filter for queries with eight or more words. Look at the shape of those queries and the pages they land on. The pages already capturing prompt-like queries are the ones to study for what’s working structurally.

From there, audit those pages. Whether the answers are available in the response HTML or only after rendering. Whether key facts live in tables, lists, and direct sentences. Whether the specifications a model would need to cite, such as pricing, integrations, use cases, and supported industries, are present and clear.

Setting that baseline now means when the reports go live on your account, you can compare the data against your own understanding of how content should be performing.

The Rollout

The reports aren’t live for most accounts yet. Google is rolling them out to a subset of websites first to gather feedback before going wider. We haven’t seen them in our own accounts yet either.

We’ll keep this post updated as the rollout expands and as we start working with the data inside client accounts. 

The framework above is what we’d recommend doing in the meantime, because the work of structuring content for AI extraction stays the same regardless of when the measurement layer arrives.

Your buyers are already asking AI who to trust. Let's make sure they find you.

blog-cta
Ana Author

Uncovering organic growth opportunities that drive revenue for Fortune 500s and high-growth brands across search and AI discovery. Making SEO make sense for anyone in the room.

Continue reading

Google IO David Jordan

Google I/O and the leap to full agentic search 

10 mins read

Why AI Content at Scale Means Nothing Without Quality Controls in Place

6 mins read
The Google logo in colorful letters is displayed on top of a glass office building against a blue sky background, representing Google's commitment to AI Search Optimization and innovation.

Google Just Published Its First Official Guide to AI Search Optimization. Here’s What Actually Matters

5 mins read