Google Is Finally Showing AI Search Visibility. The Report Still Leaves Important Questions Unanswered

Google’s new Search Console reports reveal which webpages appear in AI-generated search experiences. For publishers and communication professionals, the harder question is what that visibility actually achieves.

By Hezron Ochiel

Google is finally giving website owners a clearer view of how their content appears in AI-generated search results.

But the new reporting leaves a question that matters to publishers, researchers and communication professionals: what does an impression tell us about the influence of a source?

Imagine a university publishes research on healthcare training. Its findings appear among the links presented in an AI-generated answer. The university records an impression, but cannot tell whether its research informed the main explanation, received meaningful recognition or helped a prospective student make a decision.

That is the measurement problem Google has begun to address.

On 3 June 2026, the company announced dedicated generative AI performance reports in Search Console, giving website owners a separate view of visibility within AI Overviews, AI Mode and generative features in Discover.

Google already included AI-related appearances in its wider performance reporting. The development introduces a more focused view, initially available to selected websites.

The announcement follows growing pressure from publishers seeking greater transparency about traffic from AI search. In an earlier analysis of Google’s claims about AI-generated website clicks, I examined concerns raised by website owners who wanted evidence they could independently assess.

The new reports address part of that concern. They also expose how much remains unknown.

What Google is now showing

The reports provide five categories of information: impressions, pages, countries, devices and performance over time.

Website owners can see how often their URLs appear in supported generative AI features, identify the pages involved and examine geographical patterns. Device information is available for Search, while performance trends can be reviewed across different periods.

Separate reporting is provided for Search and Discover. This article focuses on Search, including AI Overviews and AI Mode.

A user might ask Google which Kenyan institutions offer particular health programmes and receive an AI-generated response with supporting links.

If a university’s webpage appears in that response under Google’s counting rules, the appearance may contribute to its generative AI impressions.

For the communication team, this indicates that its content is reaching an AI-enabled search environment.

It does not reveal how much attention the page received or what the person took away from the answer.

The three questions the report cannot answer

1. Was the source meaningfully credited?

A webpage may appear among the supporting links of an AI-generated answer without being central to the explanation.

The reporting identifies pages receiving impressions but does not provide a detailed account of each page’s contribution to an individual generated response.

Consider a research institute whose work appears alongside material from several international organisations.

The institute may know that its page was displayed. It may still be unable to establish whether the answer used its original findings, briefly referenced its work or relied primarily on another source.

For organisations investing in research and expert knowledge, this difference matters.

2. Which questions produced the visibility?

Knowing that a page appeared is helpful. Understanding the question that led to its appearance can be more useful for communication planning.

A university’s research page might appear when someone searches for training programmes, health workforce shortages or developments in medical education.

Those questions reflect different information needs.

Google’s first dedicated report does not provide the same query-level detail available through conventional Search performance reporting.

Communication teams can identify pages receiving AI impressions but may struggle to determine which audience needs generated those appearances.

3. Did the appearance lead to anything?

A person may encounter a university in an AI-generated answer and visit its website immediately.

Another may remember its name and search for it later.

A journalist might use the information to identify an interviewee, while a prospective student may compare institutions before applying.

These outcomes matter, but an impression cannot establish them.

The challenge is familiar to public relations professionals.

A media mention confirms that coverage occurred. It does not establish whether audiences understood the message, changed their opinion or took action.

AI search reporting needs the same measurement discipline.

Microsoft has moved further on citation reporting

Google’s announcement is worth examining alongside developments at Microsoft.

On 10 February 2026, Microsoft introduced AI Performance in Bing Webmaster Tools.

The reporting allows website owners to examine citations across supported AI experiences, including Microsoft Copilot and AI-generated Bing summaries.

It includes citation counts, cited webpages, trends and samples of the grounding queries used when retrieving information.

Microsoft subsequently expanded the reporting in June with additional views covering query intent, topics and citation share.

The two platforms use different reporting methods, so their figures should not be treated as directly comparable.

Nevertheless, the distinction is revealing.

Google’s first dedicated report shows which webpages appear in its generative search experiences. Microsoft’s reporting offers additional evidence about citation activity and some of the queries associated with retrieval.

Even these tools have limitations.

Microsoft cautions that citation counts do not establish a page’s ranking, authority or precise role within an individual AI-generated answer.

Publishers therefore need to examine visibility, attribution and influence separately.

Why this matters for public relations

The implications extend beyond SEO.

Communication teams have traditionally monitored media coverage, social media engagement, website traffic and stakeholder enquiries.

AI-generated answers introduce another place where people encounter institutional information.

Someone looking for a specialist may ask an AI assistant for recommendations. A prospective partner may request a summary of an organisation’s research. A journalist may seek background on a public agency.

The answer can influence the user’s initial understanding before they visit an official source.

I explored this wider relationship between discovery and reputation in How AI Search Is Changing Trust, Visibility, and Public Relations in Africa. One concern was how institutions could ensure their expertise remained accurately represented as AI systems became part of everyday information searches.

Google’s reporting now gives communication teams another way to investigate that concern.

They can examine which institutional webpages appear in AI search while continuing to monitor whether the organisation’s programmes, experts and achievements are described correctly.

A strong showing in conventional search does not guarantee accurate representation in AI-generated answers.

Both deserve attention.

The African knowledge attribution problem

For African researchers and institutions, attribution raises an additional concern.

A university may publish original research that contributes to international discussions. A local organisation may produce detailed evidence about a humanitarian problem. An African professional body may develop guidance used beyond the continent.

When AI systems explain these subjects, the original knowledge producers may be difficult to identify from the displayed citations.

Sometimes the original material is inaccessible or poorly documented. In other cases, an intermediary publication may be more prominent than the source from which the information originated.

Consider a hypothetical Kenyan health research institution that publishes a detailed report.

An international publication subsequently summarises the findings. A later AI-generated response cites the international article without visibly acknowledging the original study.

The institution’s research has contributed to the information available, but the recognition attached to that knowledge has shifted.

This example illustrates a possible attribution problem. It does not establish systematic preference for international publishers.

For African organisations, the issue is worth examining because recognition affects how expertise becomes discoverable to journalists, researchers, policymakers and potential partners.

Google’s reporting provides information about visibility. Examining the sources displayed within AI answers is necessary to understand attribution more fully.

What communication teams should measure

The arrival of AI search reporting provides an opportunity to improve digital communication evaluation.

I would begin with three connected measures.

Visibility: Which webpages appear in AI search, how frequently do they appear and where is that activity concentrated?

Google’s reports can help answer parts of this question.

Attribution: When an AI system provides an explanation, which sources does it cite? Is the organisation named? Are its original findings represented correctly?

Bing’s citation reporting and documented manual testing can provide additional evidence.

Impact: Does the information contribute to enquiries, applications, media requests, research engagement or other outcomes relevant to the organisation?

Answering this question requires website analytics, stakeholder feedback and appropriate records of audience action.

These measures form part of the broader approach to digital authority discussed in my Ultimate Guide to AI Search, Brand Reputation, and Public Relations in Africa. The purpose is to connect what audiences encounter online with the information they need and the decisions they make.

A university, for example, might discover that its research pages receive substantial AI impressions.

It could then test questions about its areas of expertise across different platforms and record which sources appear.

Finally, it could examine relevant website visits, research enquiries and professional contacts.

These observations would not automatically establish that AI search caused any changes. They would provide a stronger basis for evaluation than impressions alone.

What Google should show next

Google’s new reports address a genuine information gap.

Publishers now have a clearer way to identify which pages appear within its generative search features and how that visibility changes.

The next stage should provide greater insight into the circumstances surrounding those appearances.

Query-level context would help organisations understand audience information needs.

Citation and source-use information would help publishers examine attribution.

More detailed click and referral reporting would make it easier to investigate how AI search contributes to website activity.

Even with those improvements, measuring reputation and audience decisions would remain a wider task.

An AI-generated answer can shape awareness without producing an immediate website visit. A visible citation may receive little attention. A user may make a decision days after first encountering an organisation.

Communication professionals have long dealt with similar limitations in media measurement.

Google’s new data should be understood as another useful source of evidence.

The next measurement challenge

Google has opened a useful window into AI search visibility.

For publishers, the reports provide information that was previously difficult to separate from conventional search performance.

For communication teams, they offer another way to examine how institutional knowledge is encountered.

The remaining questions concern recognition, accuracy and outcomes.

African institutions have particular reason to pay attention, given the importance of making locally produced knowledge accessible and properly attributed.

An impression is worth recording.

Understanding what happened because of that impression is the more demanding task.

Hezron Ochiel is a Strategic Communications Expert at Kenya Medical Training College (KMTC), a leading government health training institution in Kenya. He is a PRSK Moran Award winner, visibility strategist and founder of Hezron Insights. His work focuses on strategic communication, AI visibility, digital authority and leadership storytelling across Africa.