What does Bing AI Performance actually measure?
Bing AI Performance gives site owners a view of how their content is cited in AI-generated answers. This guide explains how to interpret the report’s query context and changes over time.
Read each report view for what it observes
Bing Webmaster Tools extends its reporting to citation activity across Microsoft Copilot, AI-generated summaries in Bing and selected partner integrations. The table separates the questions its report views can help you investigate from conclusions they do not establish.
What the report view describes, and what it leaves open
Scroll to read all columns.
| Report view | What it describes | What it does not establish |
|---|---|---|
| Intent and topics | Classifications of grounding queries and broader query clusters. | A definitive account of user intent or a complete map of a site’s subject coverage. |
| Citation share | The site’s percentage of all citations shown across sites for the same grounding query. | Traffic share, a ranking, a quality score or a competitor-domain comparison. |
| Compare | Citation activity viewed across selected time periods. | That a content change caused the difference. |
| Citation counts and cited pages | Reported citation activity across supported AI experiences and URLs | Placement, page importance or a recommendation. |
Sources: AI Performance in Bing Webmaster Tools; AI visibility insights: intent, topics and citation share
Treat query context as a useful but imperfect classification
Grounding queries in the report are a sample of overall citation activity. A query missing from the report is therefore not proof that no citation activity occurred. These classifications come from evolving AI and machine-learning systems; during preview, labels may be broad, especially for niche subjects. Inspect the available query and cited-page context before deciding that a topic is missing from your content.
Sources: AI Performance in Bing Webmaster Tools; AI visibility insights: intent, topics and citation share
Investigate a change before editing a page
When citation activity differs between periods, choose a grounding query or cited URL and compare the relevant periods. Record plausible influences, such as changes in user behavior or demand, evolving models, freshness signals, partner refresh cycles, competing content and broader changes across the web. Then inspect the relevant page or claim and decide whether the evidence supports an edit.
Sources: AI visibility insights: intent, topics and citation share
Sources
- AI Performance in Bing Webmaster ToolsMicrosoft Bing · Accessed Explore publications using this source
- AI visibility insights: intent, topics and citation shareMicrosoft Bing · Accessed Explore publications using this source
Read next
Put the report in context
For broader distinctions between AI visibility signals, see the related guide to AI citation visibility. For help evaluating measurement approaches, see the guide to choosing a GEO tool.
Check my brand