Google Counts AI-Result Clicks, but It Does Not Show You an AI-Only Click Total

No. Google’s dedicated Generative AI performance report shows impressions, not clicks, click-through rate, or the queries that triggered an appearance, according to Google’s announcement of the reporting rollout.

Qualifying clicks from AI Overviews and AI Mode still count in Search Console’s overall Web Performance data, but Google’s current documentation does not describe an AI-only click total or filter that separates them from conventional Web Search activity. In other words, the dedicated report measures AI visibility rather than attributable AI traffic. Google confirms that traffic from these features is included under the Web search type in its guidance for AI features and websites.

The short answer: the dedicated AI report does not show clicks

You cannot use the dedicated Generative AI report to determine how many people clicked your website from Google’s AI search features. It reports impressions and provides breakdowns such as page, country, date, and device where applicable. It does not list clicks, CTR, or triggering queries as available metrics. Secondary coverage of the global rollout describes the same reporting gap: site owners receive AI visibility data without corresponding clicks or queries.

The apparent contradiction disappears once you separate two questions:

• Does Google count qualifying AI-result clicks? Yes. External-link clicks from AI Overviews and AI Mode count in Search Console.
• Can you isolate those clicks in the dedicated AI report? No. That report exposes impressions, while Google’s cited documentation does not describe an AI-only click total or filter elsewhere.

This distinction matters for reporting. Saying “Google does not count AI clicks” would be inaccurate. The more precise conclusion is that Google counts qualifying clicks within overall Web Performance data but does not expose them as a dedicated AI-only click metric in the documented reporting interface.

What each reporting layer can actually tell you

Search Console and analytics systems answer different questions. The dedicated report tells you where AI visibility is changing. The overall Performance report shows blended Google Search activity. Analytics and CRM systems describe what visitors did after reaching your website.

Reporting layer	What it shows	What it cannot establish
Generative AI performance report	AI-feature impressions, with page, country, date, and applicable device breakdowns	AI-only clicks, CTR, triggering queries, visits, leads, or revenue
Overall Web Performance report	Blended clicks, impressions, CTR, position, queries, pages, countries, and devices	Which clicks came specifically from AI Overviews or AI Mode
Web analytics and CRM	Landing-page sessions, engagement, conversions, opportunities, and revenue	Whether a particular visit or outcome came from a specific AI appearance

An AI impression means that a link to your site was shown in a generative AI feature. It does not mean the user clicked, visited the page, engaged with your content, submitted a form, or became a customer.

Google provides dedicated Search reporting for visibility in AI Overviews and AI Mode. It also provides separate reporting for generative AI features in Discover. Device breakdowns apply to Search reporting, so teams should not assume that every dimension works identically across Search and Discover.

The overall Web Performance report remains useful because it includes the clicks and other standard metrics associated with Web Search. Its limitation is attribution: the documented reporting does not identify which portion of those clicks came from AI features.

Analytics and CRM data extend the measurement chain. However, those downstream events still cannot be matched reliably to an individual AI Overview or AI Mode impression using the available Search Console data.

What counts as a click from AI Overviews or AI Mode

A click from AI Mode or an AI Overview to an external webpage counts as a Search Console click. An interaction that keeps the user within Google generally does not count as a website click. Google explains these element-specific rules in its definitions for Search Console impressions, positions, and clicks.

AI Mode follow-up questions are treated as new queries. Each follow-up can therefore create its own impression, position, and click events under the standard counting rules. The dedicated Generative AI report does not expose those triggering queries, even though the underlying interactions are handled as queries for counting purposes.

Click, impression, and position data are generally attributed to Google’s selected canonical URL. If Google displays one page version but considers another URL canonical, the canonical URL will usually receive the performance data. This can affect page-level analysis when a site has duplicate, alternate, or redirected versions of the same content.

Do not calculate an “AI CTR” by dividing dedicated-report AI impressions by overall Web clicks. Those figures do not describe the same traffic set:

• The impressions are specific to generative AI features.
• The clicks are blended across AI features and conventional Web Search.
• The resulting percentage would therefore not represent an attributable AI-feature click-through rate.

You can place the metrics next to each other for context, but you should not combine them into a rate that the underlying data cannot support.

A practical way to measure AI visibility, engagement, and pipeline

A defensible measurement framework uses three separate layers.

1. AI visibility: Track Generative AI report impressions by page, country, date, and device where available.
2. Site engagement: Review blended Web clicks in Search Console alongside landing-page sessions, engagement, and relevant on-site actions in analytics.
3. Business outcomes: Track form submissions, bookings, qualified opportunities, pipeline, and revenue in your conversion or CRM systems.

Start by exporting AI impressions by page for a defined reporting period. Identify pages with sustained or meaningful increases rather than reacting to a single short-term movement. Then compare those page-level trends with total Web clicks and landing-page sessions over the same period. Finally, inspect conversions or qualified opportunities associated with those landing pages.

This comparison is directional, not causal. If AI impressions, sessions, and opportunities all rise, the pattern may justify further investigation, but it does not prove that the opportunities came from AI Overviews or AI Mode. Conventional results, paid campaigns, referrals, direct visits, seasonality, and delayed buyer behavior may also have affected the outcome.

A hypothetical dashboard could display:

• AI-feature impressions: 12,000
• Total Web clicks: 800
• Landing-page sessions: 740
• Qualified opportunities: 6

Keep all four figures separate. Label total Web clicks as blended search activity, not AI clicks, and do not divide them by AI impressions.

If AI impressions rise while sessions or conversions remain flat, treat that pattern as a diagnostic prompt. Review whether the visible pages:

• Match the likely intent behind the topic
• Present accurate, useful, and sufficiently complete information
• Make relevant services or next steps easy to find
• Provide clear paths to related pages
• Use an appropriate call to action

Flat downstream performance does not prove that AI visibility generated no clicks. The attribution gap prevents that conclusion. It may instead indicate that visibility occurred for low-intent topics, that the page did not create a useful next step, or that other traffic sources changed during the same period.

Google recommends pairing Search Console with analytics when evaluating engagement and conversions. The practical aim is not to force every metric into one attribution model. It is to preserve the distinctions between visibility, website behavior, and business outcomes.

Limits to note before making decisions from the report

Several constraints affect how the Generative AI report should be interpreted:

• Search Labs experiments are excluded.
• A property may not display the report if it has insufficient generative AI impressions.
• A site excluded from eligible generative AI features may not generate the impressions needed for the report.
• Recent data may be preliminary and can change.
• Chart and page-table totals can differ because Google uses different aggregation methods.
• Performance data is generally attributed to the selected canonical URL.
• Standard reporting constraints, including row limits, apply.

These conditions are documented in Google’s Generative AI performance report help page. A missing report therefore does not by itself prove that a site has never appeared in a generative AI feature. Likewise, a difference between chart and table totals is not necessarily a reporting error.

Google has said it intends to add more metrics over time, but it has not specified which metrics will be added or when. Clicks, queries, CTR, and AI-specific filters remain unconfirmed rather than promised. Reporting plans should be based on currently documented fields, not anticipated features.

Reporting controls are also different from publisher controls. Google introduced a separate way for publishers to opt out of AI Overviews and AI Mode while retaining their conventional Search and Discover presence. Opting out removes the corresponding AI-feature impressions and traffic, according to reporting on the standalone publisher control. Google also says the choice is not used as a ranking signal in regular Search.

Revisit dashboard definitions if Google documents new metrics or filters. Until then, use dedicated impressions to assess AI visibility, analytics to evaluate on-site engagement, and CRM data to measure business outcomes without presenting any layer as exact AI-click attribution.

Why do I not see the Generative AI performance report in Search Console?

The most direct explanation is that the property has not received enough impressions in Google’s generative AI features. The report may also be absent if the site has been excluded from those features and therefore is not generating eligible impressions.

A missing report should not automatically be interpreted as proof that the site has never appeared. Check the property, selected date range, impression volume, and applicable publisher controls before drawing that conclusion.

Will Google add clicks or queries to the AI report later?

Possibly, but Google has not confirmed that either metric will be added. It has said that additional metrics may arrive over time without naming specific fields or providing a timeline.

Do not build dashboards or forecasts around future click, query, CTR, or AI-filter support until Google documents those capabilities.