Official statement
Other statements from this video 4 ▾
- 0:39 Can Google Search Console really measure the performance of your content on TikTok and Instagram?
- 2:34 Why does Google reject SVG favicons in its search results?
- 2:43 How will the new site reputation policy in the EEA transform your SEO strategies?
- 5:48 Could the favorite sources button be a game changer for niche site SEO?
Google is rolling out dedicated performance reports for generative AI features in Search Console for all sites. These new reports allow quantification of the real impact of AI Overviews and other AI features on your visibility. For SEOs, this finally provides the opportunity to move beyond speculation and measure whether these new interfaces are cannibalizing classic organic traffic or generating new opportunities.
What you need to understand
What exactly do these new reports bring?
Google has now integrated specific data on generative AI features into Search Console. Until now, it was impossible to differentiate a traditional impression from one appearing in an AI Overview. These reports change the game by isolating the performance related to AI features.
Specifically, you can now see how many impressions and clicks come from AI-generated snippets, compare these metrics with your traditional organic performance, and identify which content is cited by AI. This provides unprecedented transparency on a channel that has previously remained a black box for practitioners.
Why is Google making this data available now?
The timing question is not insignificant. AI Overviews have sparked backlash among publishers noticing a drop in traffic without being able to document it. Google is responding to this pressure by opening up its metrics, likely to defuse accusations of opaque cannibalization.
This transparency also has a strategic dimension: by showing that some sites are gaining traffic through these new interfaces, Google legitimizes the widescale deployment of AI in its results. This is a clear signal that these features are no longer experimental but are becoming a full-fledged channel.
What metrics are available in these reports?
The reports include presumably the impressions, clicks, and potentially the click-through rate specific to AI features. You should be able to segment by URL and query, and observe temporal evolution. Google has not detailed whether the data is historical or starts only at the deployment of the report.
A grey area remains regarding the exact definition of what constitutes an AI impression. Is it only about AI Overviews, or does it also encompass AI-enhanced featured snippets, automatically generated insights, or even Discover results boosted by generative algorithms? This ambiguity will require clarification. [To be verified]
- The reports are deployed for all sites, not just large publishers
- They allow isolating AI performance from traditional organic performance
- This transparency addresses criticisms about AI traffic cannibalization
- The exact metrics and available history remain to be confirmed in practice
- This is a signal that AI features are becoming a sustainable channel, not a temporary experiment
SEO Expert opinion
Is this transparency really complete?
Let's be honest: Google never does anything out of pure altruism. If these reports are appearing now, it's because regulatory pressure and publisher grievances have reached a critical threshold. It remains to be seen whether the data provided is truly comprehensive or if Google is only showing part of the picture.
Experience with other Search Console reports often reveals discrepancies between official data and field observations. Impressions may be undercounted, clicks attributed differently depending on the type of interaction. It will be necessary to cross-check these new metrics with Analytics to validate their coherence. [To be verified]
Is the impact on traffic necessarily negative?
Contrary to popular belief, not all sites lose traffic with AI Overviews. Some content benefits from increased visibility by being cited as sources. Sites that provide structured factual data, verifiable statistics, or detailed methodologies are more likely to be referenced.
The real issue concerns low-value content: basic definitions, generic tutorials, aggregations without analysis. These pages lose their purpose when AI synthesizes the answer directly. If your strategy relies on superficial informative content, then yes, you will suffer. However, if you produce expert, sourced, nuanced content, you could gain.
Can you optimize to appear in AI results?
Officially, Google repeats that there is no specific optimization for AI Overviews, and the same SEO principles apply. In practice, some patterns are emerging: well-structured content with semantic markup, verifiable numerical data, citations from identified experts, transparent methodologies.
Structured data (FAQ, HowTo, Dataset) seem to play a role in source selection. Content that clearly answers specific questions, with a proper informational hierarchy, is overrepresented. However, be careful; correlating does not mean causation: these best practices also boost traditional SEO, making it hard to isolate the pure AI effect. [To be verified]
Practical impact and recommendations
How can you access these new reports in Search Console?
As the rollout is gradual, not all accounts have simultaneous access yet. Log into Search Console, check the Performance section to see if new filters or tabs regarding AI features appear. Google might also create a dedicated section in the sidebar menu.
If you don’t see anything yet, that’s normal. The rollout takes several weeks. Check regularly, especially if you manage a site in a sector where AI Overviews are already massively deployed (health, finance, technology). These verticals are prioritized in deployments.
What should be prioritized in analyzing this data?
Start by identifying the queries that generate AI impressions but few clicks. This is a signal of cannibalization: users get their answer without clicking. For these queries, two strategies: either you accept the loss and redirect your efforts, or you enrich the content to create a need for clicks (deeper insights, interactive tools, exclusive data).
Next, identify content that generates traffic via AI. Analyze what differentiates them: structure, depth, factual data? Duplicate these patterns on other strategic content. Finally, monitor temporal evolution: a sudden drop in AI impressions may indicate that Google has changed its reference sources.
What corrective actions should be implemented?
If you notice a drop in classic organic traffic correlated with an increase in AI impressions without clicks, several levers exist. Enrich your content with elements that AI cannot synthesize: original case studies, proprietary data, well-argued positions, calculator tools.
Optimize your structured data so that Google can easily identify and attribute your content as sources. Add mentions of expert authors with verifiable credentials (indirect AuthorRank). Test formats that naturally create a need for clicks: detailed comparisons, interactive guides, simulators.
- Check weekly for report availability in your Search Console
- Identify queries with high AI impressions but low CTR
- Cross-check GSC AI data with Analytics to validate coherence
- Audit the content cited by AI to identify winning patterns
- Enrich cannibalized content with exclusive data and in-depth analysis
- Implement or optimize relevant structured data (FAQ, HowTo, Dataset)
❓ Frequently Asked Questions
Les rapports IA dans Search Console remontent-ils des données historiques ou seulement à partir du déploiement ?
Ces rapports couvrent-ils uniquement les AI Overviews ou aussi Discover et d'autres features IA ?
Peut-on segmenter les données IA par type de device, pays ou période comme dans les rapports classiques ?
Une baisse d'impressions IA est-elle forcément négative pour le site ?
Ces données peuvent-elles être exportées vers des outils tiers via l'API Search Console ?
🎥 From the same video 4
Other SEO insights extracted from this same Google Search Central video · duration 6 min · published on 26/10/2026
🎥 Watch the full video on YouTube →
💬 Comments (0)
Be the first to comment.