Bing AI Performance: New Tool Tracks Content Citations in Search & Copilot

Microsoft this week unveiled a fresh tool offering publishers unprecedented visibility into how their content is utilized by artificial intelligence systems, a move that could reshape the dynamics of content discovery and optimization. The “AI Performance” dashboard, introduced within Bing Webmaster Tools on February 10, 2026, provides data on citations of publisher content across Microsoft Copilot, AI-powered summaries in Bing search and partner integrations.

The public preview, announced by Principal Product Managers Krishna Madhavan, Meenaz Merchant, Fabrice Canel, and Saral Nigam, marks Microsoft’s first step toward providing comprehensive “Generative Engine Optimization” tooling within its webmaster platform. For publishers, the dashboard addresses a long-standing concern: a lack of transparency regarding how AI systems are leveraging their work.

Previously, publishers operated with limited understanding of whether their content appeared in AI-generated responses, a situation industry observers have described as a “black box” for attribution. The new dashboard aims to lift the veil, offering four key metrics: citation frequency, page-level activity, grounding query phrases, and temporal trends. According to Microsoft, the dashboard “shows where and how content from your site is referenced as a source across AI experiences.”

The implementation provides publishers with total citation counts, revealing how many times their content has been used as a source in AI-generated answers within selected timeframes. The dashboard also highlights the average number of unique pages from a site appearing as sources daily, reflecting overall citation patterns. “Grounding queries” expose the specific phrases AI systems used when retrieving referenced content, offering insight into the AI’s retrieval logic. Microsoft acknowledges this data represents a sample of overall activity and plans to refine the metric as data processing capabilities expand. Page-level citation activity identifies which specific URLs receive the most references, while a timeline visualization displays changes in citation activity over time.

Microsoft emphasized its commitment to respecting content owner preferences, stating that Bing respects all directives expressed through robots.txt and other supported control mechanisms. This addresses growing publisher concerns about AI systems accessing content without clear consent or established compensation frameworks.

The dashboard’s metrics differ significantly from traditional search performance reporting. Citations in AI-generated answers are independent of click-through rates, impressions, or average positions that define conventional SEO. A page can receive numerous citations without generating substantial traffic, or conversely, drive traffic without frequent citations. This distinction is critical, as AI-generated answers often satisfy user information needs directly, reducing the require to click through to source websites.

Research from PPC Land indicates that visitors arriving via AI search demonstrate 4.4 times higher value compared to those from traditional organic search, measured by conversion rates, although overall traffic volumes may decline. The grounding queries metric provides insight into how AI systems match content to user prompts, differing from traditional search keywords as they represent AI retrieval logic rather than user-entered queries. Understanding these phrases allows publishers to optimize content structure and depth for AI comprehension.

Microsoft recommends publishers strengthen content depth and expertise, recognizing that pages cited for specific grounding query phrases often demonstrate clear subject focus and domain authority. Improving structure and clarity through headings, tables, and FAQs can also help surface key information. Supporting claims with evidence, including examples, data, and citations, builds trust, while regular content updates ensure AI systems reference the most current information.

The company highlighted the importance of the IndexNow protocol, a real-time notification system that informs search engines and AI systems when content is added, updated, or removed. With 60 million websites joining IndexNow daily and 1.4 billion URLs submitted each day as of October 2023, Microsoft views real-time content freshness as critical for maintaining citation relevance. Comprehensive XML sitemaps are also positioned as essential infrastructure for content discovery in AI-powered search environments.

For local businesses, Microsoft emphasized the importance of accurate business information, directing them to Bing Places for Business to ensure key details remain current and eligible for inclusion in AI-generated responses. The redesigned platform, launched October 3, 2025, allows businesses to import listings from Google Business Profile or create new profiles manually.

The AI Performance launch positions Microsoft ahead of Google in providing dedicated tooling for AI citation visibility. While Google has deployed AI Overviews extensively, it has not yet released comparable analytics showing publishers how frequently their content appears in AI-generated summaries. Research from PPC Land showed that ChatGPT citations predominantly reference content ranking in traditional organic search positions 21 or lower approximately 90 percent of the time, a pattern that challenges conventional SEO assumptions.

Microsoft’s Copilot advertising business surpassed $20 billion in annual revenue by April 2025, with search and news advertising revenue climbing 21 percent in recent quarters. Microsoft Advertising research showed Copilot achieving 73 percent higher click-through rates compared to traditional search advertising.

The release of the AI Performance dashboard signals Microsoft’s commitment to transparency between AI systems and the open web, addressing concerns about attribution and visibility in AI-generated content. The dashboard provides foundational visibility into a rapidly evolving content discovery landscape where AI-generated answers increasingly mediate between users and original sources.