Google Search Console adds first-party AI performance reports for AI Overviews and AI Mode

Google has begun shipping first-party performance reporting for its generative AI search surfaces inside Search Console, giving site owners an official way to measure how often their content appears in AI Overviews and AI Mode. According to a write-up on dev.to, the new views sit within the existing Performance report and also take in related AI surfaces in Discover.

What the reports cover

Dev.to reports that Google detailed the launch in an official announcement, with supporting Search Console documentation explaining how the figures are counted. The core measurement is impressions — instances where a page contributes to an eligible AI experience — and those impressions can be broken down by dimensions such as pages, devices and countries, mirroring the slicing site owners already know from conventional search reporting.

For publishers, marketers and businesses, that separation is the point. Visibility inside a generative answer does not look like a classic blue-link result, and dev.to notes that this exposure previously sat inside broader performance data, leaving owners to infer how much of it came from AI features.

A reporting lens, not a ranking change

The rollout does not introduce a new ranking system or a separate optimization playbook. It is best understood as a new lens on data Google already collects: a defined home inside Search Console where AI-driven visibility can be inspected on its own, rather than treated as an unmeasured byproduct of overall organic performance.

Where the numbers stop short

The reports are deliberately scoped, and dev.to is careful to flag the boundary. They emphasize impressions and surface-level reporting, not a query-by-query account of every AI answer appearance. A report can show that AI visibility is trending up or down without explaining which queries triggered it, how an answer was composed, or why a particular source was selected.

There is also a limit on interpretation. An impression is a visibility metric whose meaning depends on the AI surface where it was counted; it does not by itself prove anything about content quality, user behavior after a visit, or commercial outcomes such as leads and sales.

How to use the data

The dev.to piece suggests a practical workflow:

• Establish a baseline by reviewing the AI-specific report over a consistent date range.
• Identify which pages show up in the AI-focused data and assess whether they address their topic clearly.
• Compare device and country patterns before changing content or publishing priorities.
• Look for sustained movement over time rather than reacting to a short reporting window.
• Pair the findings with analytics and business data, which remain necessary to understand what users do after an AI appearance.

Why it matters

AI-generated answers are changing how content is presented to searchers, and site owners have had limited direct visibility into that shift, left to infer AI exposure from overall Search Console numbers. Google now provides a first-party signal where there was previously little, which should sharpen conversations between content, search and business stakeholders: instead of asking only whether overall traffic rose or fell, teams can ask whether exposure in AI experiences is changing and which pages are involved.

The data has clear limits — no full query-level detail and no direct line to conversions — but it moves AI answer visibility from assumption toward measurable reporting, and it gives editorial teams a concrete starting point for deciding which pages deserve closer attention.