Writing · Series: How AI Search Works · Follow-on 1
Why the Bing AI Performance Report Is the Most Underused Tool in GEO.
Microsoft's Bing AI Performance report is the only first-party citation dashboard from any AI search platform. Most practitioners aren't using it. Here's what it shows, what it doesn't, and how to turn it into a content strategy signal.
Published1 June 2026
ByThomas Cox
Read time9 minutes
Filed underSeries · GEO Measurement · Follow-on 1
Every practitioner working in GEO in 2026 is dealing with the same measurement problem: there is no reliable first-party data on how AI systems are using their content. Google Search Console shows organic performance. It does not show AI Overview citations. You can observe that an AI Overview exists for a query, but you cannot see, from Google's own tooling, which pages were retrieved to generate it.
Microsoft has changed this for its own platform. The Bing AI Performance report, launched in public preview in February 2026, is the only first-party citation measurement tool available from any AI search platform. If you're not using it, you're leaving the only empirical window into AI citation behaviour on the table — regardless of where your primary traffic originates.
The AI Performance report is accessible via Bing Webmaster Tools (bing.com/webmasters) for any verified site. The metrics it provides:
Total Citations — the number of times pages from the verified site were cited as sources in AI-generated answers during the selected time period. This is the closest available metric to an AI visibility score: not ranking position, not impressions, but direct citation events.
Average Cited Pages — the daily average of unique pages from the site cited in AI-generated answers. A higher number indicates broader content coverage across AI-cited topics; a lower number may indicate concentration of citations on a small number of pages.
Grounding Queries — sample query phrases that Copilot's system generated internally when retrieving and citing content from the site. These are not the user's original prompts. They are the reformulated queries that Copilot's system determined would retrieve relevant information — the actual retrieval signal, not the user-facing query.
Page-Level Citation Activity — citation counts broken down by URL, showing which specific pages are cited most frequently.
Visibility Trends — a timeline view showing how citation activity changes over the selected date range. Available from November 1, 2025 onwards.
Microsoft Clarity users who have linked their Clarity workspace to Bing Webmaster Tools also get the AI citation data surfaced inside Clarity's reporting interface — useful if Clarity is already your primary analytics view.
The grounding queries: the most valuable data point.
The grounding queries data is the most strategically valuable element of the report, and the most commonly misunderstood.
When a user asks Microsoft Copilot a question, Copilot does not send the user's verbatim question to Bing for retrieval. It generates a set of distilled search terms — the grounding queries — that the system determines will retrieve relevant information. These grounding queries are shorter, more keyword-like than the user's original conversational prompt.
The grounding queries visible in the AI Performance report are the actual retrieval signals that caused your content to be cited. They show you what Copilot was looking for when it found your pages — which is often different from what you expected.
Practically: if a grounding query that consistently cites your content is "enterprise data security compliance framework 2025" but your page is titled "Our Approach to Data," the grounding query is telling you that Copilot is finding your content for a specific informational need that your title and metadata do not clearly signal. That is an optimisation signal — not just for Copilot, but for the semantic clarity of your content more broadly.
What the report does not show.
Microsoft's documentation is honest about the report's limitations. Practitioners should be equally honest with clients and stakeholders.
No click data. The report shows citation frequency — how often your content was cited as a source in an AI-generated answer. It does not show how many users clicked through from that citation to your site. The relationship between citation and traffic is unknown from this data alone.
Sampled, not exhaustive. Microsoft's documentation notes that grounding queries are sampled and may be refined retroactively as more data is processed. The data is directionally reliable, not a complete audit trail.
Bing/Copilot only. The report covers Microsoft Copilot (including Bing's AI-generated summaries and Edge integration) and select partner integrations. It does not cover Google AI Overviews, ChatGPT, or Perplexity. It is one platform's data, not a cross-platform picture.
No competitive view. The report shows your site's citations. It does not show which other sites Copilot cited alongside yours for the same grounding queries. Knowing the competitive citation landscape requires separate research.
How to use the report strategically.
Step 1: Verify in Bing Webmaster Tools. If the site is not verified in Bing Webmaster Tools, do this first. Verification is free and straightforward (XML file, meta tag, or DNS record). It is also the prerequisite for Bing indexation monitoring, which is the baseline for Copilot web citation eligibility.
Step 2: Identify which pages are cited and for which grounding queries. Sort the grounding queries by volume. Map each query to the page it most frequently cites. This gives you a picture of what Copilot perceives each page as being authoritative about — which may or may not match what the page is optimised for.
Step 3: Identify citation gaps. Look for grounding queries where Copilot is retrieving results from other sources rather than from the site. These represent either content that does not exist on the site (content gaps) or content that exists but is not structured clearly enough to be retrieved (structural gaps).
Step 4: Use citation trends to detect content freshness issues. A declining citation trend for a page that was previously well-cited often indicates that more recently updated content from other sources has displaced it. AI retrieval systems favour fresh content. A freshness pass — updating data, adding recent examples, refreshing publication dates where appropriate — can recover lost citation frequency.
Step 5: Feed grounding queries into broader content strategy. The grounding queries are, effectively, what AI systems believe users are asking about in your domain. They are a research signal independent of traditional keyword data. Use them to identify content opportunities that keyword research may not surface — questions AI systems are actively retrieving for, in your topic area, that your site is not yet the best source for.
The Bing AI Performance report is a first-party measurement tool from one platform, covering one ecosystem. It is limited. But it is the most direct empirical data available anywhere on how AI systems actually use web content.
More importantly: it demonstrates what platform-level citation measurement can look like. If Microsoft has built this, the expectation should be that Google and others will follow. The GEO measurement landscape of 2027 will likely look very different from 2026.
In the meantime, the practitioners using the data that exists will be better positioned than those waiting for the data they wish existed.
/ Next in the series
Article 8 asks why Anthropic, OpenAI, and Perplexity haven't published any equivalent guidance — and what that silence means for how to approach them. Read Article 8 →