For the past couple of years, one of the most frustrating gaps in search measurement has been AI Overviews. Your content might be appearing in them regularly - cited, summarised, used as a source - and you would have no reliable way to confirm it. Impressions were invisible. Click attribution was murky. You were essentially making educated guesses about whether your content strategy was working in AI search.
That gap is now closing. Google has introduced dedicated AI Overview reporting directly inside Google Search Console, covering performance in AI Overviews, AI Mode, and the AI features within Discover. This is a significant shift - not because the data is perfect, but because it is the first time Google has given marketers a structured, queryable view of how their content performs inside AI-generated results.
What the New Reporting Actually Shows You
The reporting allows you to see how your content is being used across AI Overviews, AI Mode, and AI-powered Discover - three distinct surfaces that Google's AI systems now populate. These are separate from traditional organic listings, and Google is treating them as such by reporting on them separately within GSC.
The core value here is segmentation. Previously, any traffic arriving from a search where an AI Overview appeared was bundled into your standard organic data. There was no clean way to isolate AI-influenced sessions or to tell whether a click came from the AI-generated panel or the traditional blue links below it. The new GSC data gives you dedicated impression and click data tied specifically to these AI surfaces.
This matters enormously for content teams trying to make the case for GEO investment internally. If you can show a stakeholder that specific pages are generating impressions in AI Overviews - even when those pages are not ranking position one in traditional results - you have concrete evidence that your content is being trusted and used by Google's AI systems. That is a fundamentally different conversation to have.
Using the Data to Identify What Is Already Working
The first practical step is diagnostic: pull the AI Overview data and cross-reference it against your existing content. Which pages are appearing in AI Overviews? What topics, formats, and content structures do those pages share? This is not about reverse-engineering an algorithm - it is about understanding your own content through the lens of what Google's AI has decided is trustworthy and useful enough to cite.
Look particularly at pages that rank modestly in traditional search - say, positions four through fifteen - but are generating AI Overview impressions. These are your hidden performers. They may not be winning the traditional click, but they are being surfaced inside AI-generated answers. That tells you the content is structurally strong and thematically relevant. The opportunity is to build on those topics, extend coverage, and reinforce your authority in that space.
Equally, look at pages that rank well traditionally but are absent from AI Overview data. Those gaps may indicate that the content is too shallow, too commercially biased, or structured in a way that does not translate well into AI-generated summaries. Both types of insight are actionable - one tells you where to double down, the other tells you where to rework.
The Click-Through Question in AI Search
One thing the new data will likely confirm for many sites is that AI Overview impressions do not automatically translate into high click-through rates. This is expected, and it does not make AI visibility worthless. When a user sees your brand cited in an AI-generated answer, that is a brand impression. Whether or not they click, they have been exposed to your content as a credible source.
The more useful question to ask of the data is this: which AI Overview appearances are generating clicks, and what do those queries have in common? If you can identify the types of intent - navigational, transactional, or specific comparison queries - that drive actual clicks from AI-generated results, you can weight your content strategy accordingly. Not all AI visibility is equal, and the GSC data gives you the means to start ranking it by commercial value.
For UK businesses in particular, this matters as Google's AI features become more embedded in the standard search experience. AI Overviews are now a mainstream feature for UK users, not a beta experiment. The traffic patterns are real, and so is the need to measure them properly.
AI Mode and Discover: Two Surfaces Worth Watching Separately
The inclusion of AI Mode and AI-powered Discover data alongside AI Overviews is easy to overlook, but each surface has a distinct user intent profile. AI Mode is a more conversational, research-oriented experience where users are exploring topics in depth rather than running a quick query. Appearing in AI Mode results suggests your content handles complex, multi-part questions well - which is a different structural requirement to a standard featured snippet.
Discover, meanwhile, is a push surface - content served to users based on interest signals rather than active searches. AI-powered Discover is becoming an increasingly important channel for content that earns ongoing authority on a topic. If your content is appearing there, it means Google's systems have associated your brand with a subject area strongly enough to proactively surface it. That kind of topical authority is exactly what good GEO practice is designed to build.
Separating the three surfaces in your analysis is important. A content piece that performs well in AI Overviews may do nothing in Discover. Understanding those differences will help you plan content that serves multiple AI surfaces rather than optimising narrowly for one.
Turning GSC AI Data Into a Working Content Brief
Once you have a baseline picture of where you are appearing and where you are not, the data starts to shape your editorial decisions. Pages generating AI Overview impressions on queries where you currently have thin supporting content represent a clear brief: go deeper on that topic. Add context, address follow-up questions, include structured information that an AI system can extract and cite reliably.
Pages generating zero AI Overview impressions despite ranking traditionally on relevant queries are a different brief: audit the content for structure, depth, and credibility signals. Is there a clear, direct answer to the likely query? Is the information presented in a way that can be parsed by an AI system, or is it buried in lengthy prose? Are there trust signals - author attribution, date stamps, factual specificity - that help Google's AI assess the content as reliable?
This is where GSC's AI reporting becomes a genuine strategy tool rather than a vanity metric. It gives you a feedback loop between your content and Google's AI systems - one that did not exist six months ago. The brands that start using it methodically now will build a material advantage over those treating AI visibility as unmeasurable.