Microsoft Clarity Citations Dashboard — Free AI Citation Measurement (2026)

Key takeaways

• Microsoft Clarity's Citations dashboard went generally available on May 13, 2026. It is free and shows how often your pages are cited in AI answers, your citation share against competitors, and the queries AI used to find your content.
• Setup takes any one of three options — install the Clarity tracking code, connect Google Search Console, or connect Bing Webmaster Tools.
• Two caveats: Share of authority can read higher than reality (officially stated), and the list of AI platforms covered is not disclosed.
• Combined with GSC's generative AI report and Bing Webmaster Tools, this completes a free measurement stack — these tools tell you where you were cited; why you are not named and what to fix is the job of diagnostics.

GA4 barely registers AI traffic, and citations happen a step before the visit. One more free tool that fills part of this measurement gap has now reached general availability — Microsoft Clarity, best known for behavior analytics, now includes AI citation measurement.

What shipped

Microsoft Clarity's Citations dashboard went generally available on May 13, 2026 (Microsoft Clarity Blog). In line with Clarity's pricing, it is free, and domain verification takes any one of three routes — installing the tracking code, connecting Google Search Console, or connecting Bing Webmaster Tools (Microsoft Learn).

"Visibility has become less about ranking in search results and more about whether your content is used to shape the answer itself." — Microsoft Clarity Blog, 2026

The official documentation makes the same point — the dashboard does not measure search rankings, impressions, or click-through rates. It measures the influence that occurs inside the AI experience, before a user ever visits your site.

The five metrics — definitions and how to read them

Metric	Official definition	How to read it in practice
Page citations	Total times your pages were referenced in AI answers during the period	Identify which content and formats get cited consistently
Share of authority	Your share of total citations across the same query set	Relative position vs. competitors — see the calculation caveat below
AI referral traffic	Share of sessions arriving from AI assistants	How often citations turn into visits
Grounding queries	Queries AI systems used to retrieve your content before answering	Not what users typed — what AI searched — a clue for aligning content
My cited pages	Cited pages with counts and associated queries	Separate pages to expand from pages needing structural work

Among these, grounding queries are the information other tools rarely surface. What users type and what AI systems search internally are different, and free tools that expose the latter are scarce. Knowing which queries AI uses to find your pages tells you which topic phrasings your content should be aligned with.

August 2026 expansion — Topic Insights groups citations by topic

In its first monthly product newsletter in August 2026, Microsoft announced a Topic Insights expansion to Clarity's AI visibility reporting (Search Engine Journal, 2026-08). Instead of individual URLs and queries, it groups AI citations by topic — showing which topics AI systems associate with your brand, how frequently those topics appear, and where coverage gaps remain.

Notably, the announcement came through Microsoft Advertising. Microsoft recommends comparing grounding queries against existing paid search terms and using competitive citation data to adjust landing pages — a signal that free measurement data is starting to feed paid-media practice.

Two caveats when reading the numbers

First, Share of authority can read higher than reality. The official docs state this directly — the calculation is applied at a daily level, includes only the queries where your domain was cited that day, and excludes days without citations. Being an "active participation" calculation, it may produce higher values than broader approaches (Microsoft Learn). When using this number in external reporting, state the calculation method alongside it.

Second, there is no list of which AIs are counted. Official materials describe coverage only as "supported AI experiences." Since you cannot confirm whether ChatGPT, Gemini, or Perplexity are included, do not generalize these figures as "citations across all AI." The docs also state that citation counts do not reflect position or prominence within an answer.

Where it sits in the free measurement stack

Three free AI visibility measurement tools are now available.

1. Google Search Console's generative AI report — exposure in Google surfaces (AI Overviews, AI Mode). Covered in our GSC generative AI report explainer.
2. Bing Webmaster Tools' Citation Share and Web IQ — citation share in Bing surfaces. Covered in Bing made it official: rankings ≠ citations.
3. Clarity Citations — behavior analytics combined with citations and referrals in one view. This article.

All three share the same limitation — they show where and how often you were cited, but not why your brand goes unnamed in answer bodies or what to fix. Citations and brand mentions move independently, as we documented in our ghost citation analysis. That next step is the domain of AI visibility diagnostics such as RanketAI — AI brand visibility analysis measures brand mentions and source citations separately across ChatGPT, Gemini, and Perplexity, and page structure diagnostics checks whether your pages are structured to be citable in the first place.

Frequently asked questions

We have GA4 — why look at this separately?

GA4 measures what happens after the visit. AI citations happen before the visit, inside the answer, and many end without a click. Where AI traffic gets lost in analytics is covered in our ChatGPT traffic measurement gap analysis.

Doesn't this overlap with GSC and Bing Webmaster Tools?

They measure different areas. GSC covers Google surfaces, Bing's tools cover Bing surfaces, and Clarity combines citations, referrals, and behavior from its own aggregation. The three numbers will differ — that is expected, and they complement rather than duplicate each other.

What exactly is a grounding query?

The search query an AI system ran internally to retrieve supporting material before generating an answer. It can differ from what the user typed, and it shows which phrasings lead AI to your content.

Do we have to install the tracking code?

For domain verification alone, connecting Google Search Console or Bing Webmaster Tools is enough. Session-based metrics such as AI referral traffic, however, only produce meaningful values with the tracking code installed.

Related reading

• GSC generative AI report explainer — the Google layer of free measurement
• Bing made it official: rankings ≠ citations — citation share metrics in Bing
• Why GA4 barely sees ChatGPT traffic — the structure of the measurement gap
• Ghost citations — why citations and mentions move separately — the metric to check after citations