As Google continues to integrate AI powered features like AI Overviews and AI Mode into its search results, website owners and GEO professionals are increasingly seeking ways to measure their impact. A common question that has emerged is how to extract and analyze performance data, such as impressions and clicks, specifically for these AI-driven results within Google Search Console (GSC).
However, it is crucial to understand that Google Search Console does not currently offer a direct method to isolate or filter data for AI Overviews. All performance metrics from AI Overviews are aggregated with the standard web search data. This limitation has been confirmed by Google representatives and is reflected in their official documentation [2].
“Just like the rest of the search results page, sites appearing in AI features (such as AI Overviews and AI Mode) are included in the overall search traffic in Search Console. In particular, they’re reported on in the Performance report, within the “Web” search type.” [2]
There was a brief period of excitement in the SEO community in September 2025 when a rumor of a new “AI Overviews” filter in GSC began to circulate. However, this was quickly debunked by Google’s John Mueller as a fake screenshot, and he reiterated that no such feature was planned for the immediate future [4].
This guide presents two primary workarounds to this limitation: a heuristic-based approach using only GSC data, and a more robust method that combines GSC data with third-party tools.
Understanding how impressions are counted is key to interpreting the data you see in GSC. The methodology for AI Overviews is more complex than for traditional search results and depends on how a citation appears within the overview. The visibility of a citation can be categorized into three tiers, as detailed by an experiment from Brodie Clark [1]:
| Tier | Citation Visibility | Impression Trigger | Impact on CTR |
|---|---|---|---|
| 1 | Visible Citations | Automatically on page load | Standard CTR calculation |
| 2 | Partially Visible Citations | When a user clicks “show more” | Appears higher due to pre-qualified interest |
| 3 | Hidden Citations | When a user expands the section | Appears highest, as an impression is only logged after significant user interaction |
This tiered system means that the Click-Through Rate (CTR) for citations within AI Overviews can often appear higher than for traditional organic results. This is because, for partially visible or hidden citations, an impression is only recorded after a user has already shown a level of interest by choosing to expand the content, thus making a subsequent click more likely.
You can identify likely AI-driven queries directly within your GSC data by focusing on query characteristics. This approach, as detailed by Otterly.AI, is based on the premise that longer, more conversational queries containing specific keywords are more likely to be AI prompts [7].
This method provides a quick and direct way to surface potential AI prompts from your existing GSC data without the need for additional tools.
For a more accurate and comprehensive analysis, you can combine GSC data with insights from third-party SEO platforms. Tools like Semrush, Ahrefs, and SISTRIX have developed features to track when and where AI Overviews appear for specific keywords.
By cross-referencing this data with your GSC performance reports, you can estimate the impact of AI Overviews on your site’s traffic. The general methodology, as outlined by Rich Sanger, involves a comparative analysis of keywords that have gained or lost AI Overview visibility over a specific period [5].
Directly extracting AI Overview prompt impressions from Google Search Console remains an impossibility for now. However, by using a combination of heuristic analysis within GSC and the more advanced method of integrating third-party tool data, you can gain significant insights into the impact of AI on your search performance.
The heuristic approach offers a quick, tool-free way to identify potential AI prompts, while the third-party tool method provides a more robust and quantifiable measure of impact. Employing both strategies will give you the most comprehensive understanding possible of your visibility in Google’s evolving AI-driven search landscape.
[1] Clark, B. (2024, September 19). AI Overview Tracking in Google Search Console [SEO Experiment]. Brodie Clark Consulting. https://brodieclark.com/ai-overviews-google-search-console/
[2] Google. (n.d.). AI Features and Your Website. Google Search Central. https://developers.google.com/search/docs/appearance/ai-features
[3] Schwartz, B. (2025, May 26). Google Search Console to show AI Mode performance but you won’t be able to break it out. Search Engine Land. https://searchengineland.com/google-search-console-to-show-ai-mode-performance-but-you-wont-be-able-to-break-it-out-455992
[4] Schwartz, B. (2025, September 15). Google Search Console Did Not Add An AI Overviews Filter. Search Engine Roundtable. https://www.seroundtable.com/google-search-console-no-ai-overviews-filter-40107.html
[5] Sanger, R. (2025, January 20). Measuring Google AI Overviews: Performance Insights and Strategies. Rich Sanger SEO. https://richsanger.com/measuring-the-impact-of-google-ai-overviews-on-seo-performance/
[6] SISTRIX. (2025, June 2). Track AI Overviews with SISTRIX, across all countries. SISTRIX. https://www.sistrix.com/changelog/track-ai-overviews-with-sistrix-across-all-countries/
[7] Peham, T. (2025, July 29). How to Find Real Prompts in Google Search Console (and Analyze Them in OtterlyAI). Otterly.AI Blog. https://otterly.ai/blog/analyze-real-prompts-google-search-console/