Bing AI Performance, the AI citation report in Bing Webmaster Tools, gained a Topics Portfolio tab on October 13, 2026. It groups the queries behind your AI citations into broad themes and shows queries where other sites are cited and yours is not.
The feature launched four weeks ago, so this post focuses on how to use it. Microsoft labels it beta, and the numbers come with caveats that matter for small sites. This piece covers what the tab adds, how to open it, what each column means, where the data is weak, and how it compares with what Google shows you.
The short version: Bing is the only major search engine that shows site owners the queries behind their AI citations. Topics Portfolio adds context: how big each topic is, how crowded it is, and where you are missing. Treat it as a planning aid. It does not report clicks, rankings or traffic, and Microsoft says the numbers are approximate.
What Topics Portfolio adds to Bing AI Performance
According to Microsoft’s October 13 announcement on the Bing Webmaster Blog, Topics Portfolio "groups the grounding queries where your site appears into broader thematic clusters." Grounding queries are the short phrases an AI system used to retrieve content it then cited.
The post says the tab sits next to Citations & Pages inside AI Performance. It appears only "if there is sufficient data for your site." The topic table has four columns, and a separate list called Top Opportunity Queries sits beside it.
Click into a topic and you get a detail page. Microsoft describes three summary cards there: topic size, your citations in the topic, and the query count. Below them are two tables. Your Grounding Queries lists the top queries in that topic where you already appear. Opportunity Queries lists the top queries where other sites appear and you do not. Citation volumes for opportunities are shown as ranges, not exact counts.
You can view one, two or three months of data. Each topic row and opportunity query has thumbs up and thumbs down buttons for feedback.
What Bing AI Performance already measured
Topics Portfolio sits on top of a report that launched as a public preview on February 10, 2026. Microsoft’s February launch post says it covers three surfaces: Microsoft Copilot, AI-generated summaries in Bing, and select partner integrations.
In June, Microsoft added four preview views to the same report. Its June 16 Bing Search Blog post lists Intents, Topics, Citation Share and Compare. The June Topics view groups your own queries. The October tab adds the outside view: how many sites compete in each topic and what you are missing.
| Metric or view | What Microsoft says it shows | Added |
|---|---|---|
| Total Citations | Times your content was shown as a source in AI answers | February 2026 |
| Average Cited Pages | Average unique pages from your site cited per day | February 2026 |
| Grounding queries and pages | Phrases used to retrieve your cited content, mapped to URLs | February 2026 |
| Intents, Topics, Citation Share, Compare | Query intent labels, themes, your share of citations per query, period overlays | June 2026 (preview) |
| Topics Portfolio | Topic size, sites cited, query count, opportunity queries | October 2026 (beta) |
The AI Performance help page adds the practical details. Data refreshes daily with a short delay. Standard ranges are 7 days, 30 days and 3 months, plus custom ranges. Grounding query and page data export to CSV and Excel.
How to open the Bing AI Performance report
You need a verified site in Bing Webmaster Tools. Microsoft’s June 2025 getting-started post calls the tool free. It lists five ways to verify, including importing a site you already verified in Google Search Console.
- Sign in to Bing Webmaster Tools and pick the site.
- Open AI Performance from the left menu.
- Look for the Topics Portfolio tab next to Citations & Pages.
- Set the window to three months for the most data.
- Open a topic to see its two query tables.
Two notes on step 3. The October post calls it the Topics Portfolio tab in one place and "the Topics tab" in another, so the label may vary. And if you see no tab at all, your site may not have enough citation data yet. Microsoft says that directly.
For agencies with many client sites, verify each property under a shared agency account where your contracts allow it. Set a reminder to check monthly. The data is too thin to watch weekly on most small sites.
Reading the Bing AI Performance topic table
The four columns answer different questions. Here is how Microsoft defines each one, and how we suggest reading it.
| Column | Microsoft’s definition | How to read it |
|---|---|---|
| Topic | A broad cluster of related grounding queries where your site appears | The theme. The label is AI-generated and may be imperfect. |
| Topic Size | Approximate total citations across all sites for the topic | How much AI answer demand the theme has |
| Sites Cited | Approximate number of sites appearing for the topic’s queries | How crowded the theme is |
| Queries | Approximate number of distinct grounding queries in the topic | How wide the theme is |
A large topic with few sites cited is the most useful row. It means AI answers draw on a small set of sources for a busy theme. If you already appear there, that coverage is worth protecting.
A large topic with many sites cited is harder. Your share will likely be small no matter what you publish. The June Citation Share view can show whether your slice is growing.
A small topic with your site cited often is usually your niche. For an association, that might be its certification, its standards or its annual meeting. Those pages should stay current.
Opportunity queries need judgment
Top Opportunity Queries lists, in Microsoft’s words, "the strongest queries across your whole portfolio where other sites appear and yours does not." This is the list most teams will want to act on. It needs a filter first.
Microsoft’s own example shows why. A branded shoe query can show up as an opportunity in a running shoes topic. A nonprofit could see another group’s program name, or a vendor’s product, flagged the same way. Some opportunities are not yours to win.
We suggest sorting each opportunity into one of three piles:
- Fit: you have expertise and no good page. Write one.
- Gap: you have a page, but it is thin or dated. Update it.
- Skip: branded, off-mission or outside what you do.
The help page’s content advice is ordinary SEO advice. It recommends descriptive headings, short sections, tables and FAQ-style content, plus evidence and fresh information. Our post on writing for AI citation covers the same ground in more depth.
Limits of Bing AI Performance data
Microsoft is open about the weak spots. The October post lists these:
- Topic labels are AI-generated and may be imperfect.
- Topics with too little activity are hidden, so smaller sites may see a short list or no tab.
- Branded queries can appear as opportunities.
- Topic Size, Sites Cited and Queries are approximate. Opportunity volumes are ranges.
The post’s guidance is to "Treat the numbers as directional signals for planning, not as a precise ledger." The help page adds that the data is a sample, very low activity may not appear, and totals can differ between views.
Two more points matter for reporting to clients or boards. First, citations are not clicks. The help page says a citation "does not represent traffic, clicks, or user engagement." Second, citation counts say nothing about ranking or importance. Our post on how AI citations and rankings have decoupled explains why those two numbers now tell different stories.
Finally, the report covers Microsoft surfaces and select partners only. It does not tell you what ChatGPT, Claude, Gemini or Perplexity cite. For that, see our comparison of manual tracking and paid platforms.
How Google’s AI reporting compares
Google now has its own report, with a different design. Search Console’s Generative AI performance report covers AI Overviews and AI Mode, and Google’s help page says it became available to all websites worldwide as of August 31, 2026. It counts impressions, meaning times links to your site were shown in a generative AI feature. It groups data by page, country, date and device, with no query view and no topic grouping. We covered it, and the opt-out next to it, in our post on Google’s generative AI performance report. Bing AI Performance reports citations and the queries behind them; Google reports impressions by page. Use both, and do not add their numbers together.
A monthly routine for small teams
Most of our readers do not have an analyst for this. A 30-minute monthly check is enough.
- In Bing AI Performance, export grounding queries and pages for the last 30 days. Save them with the date in the file name.
- Open Topics Portfolio on the three-month window. Note your top five topics by your citations.
- Mark any topic where you were cited last month and are not now.
- Sort the opportunity queries into fit, gap and skip.
- Pick one gap page to update and one fit topic to write about.
- Next month, use Compare to see whether citations moved after the change.
Keep the routine boring. Do not rewrite a site around one month of beta data. If you need a primer on the wider practice, start with our explainer on answer engine optimization.
Frequently Asked Questions
What is Bing AI Performance?
It is a report in Bing Webmaster Tools that shows how often your pages are cited as sources in AI answers. It covers Microsoft Copilot, AI-generated summaries in Bing and select partner integrations. It launched as a public preview on February 10, 2026.
What is Topics Portfolio?
A beta tab in AI Performance, announced October 13, 2026. It groups the grounding queries where your site appears into broad topics, shows each topic’s size and how many sites are cited, and lists opportunity queries where other sites appear and yours does not.
Why don’t I see the Topics Portfolio tab?
Microsoft says the tab appears only when there is sufficient data for your site. Smaller sites may see a short list or no tab. The help page says missing grounding queries do not indicate a penalty.
Does Bing AI Performance show clicks or traffic?
No. A citation means your content was shown as a source in an AI answer. Microsoft’s help page says it does not represent traffic, clicks or engagement. Use your web analytics for visits.
How accurate are the Bing AI Performance topic numbers?
Microsoft calls them approximate and says to treat them as directional signals for planning. The data is sampled, topic labels are AI-generated, and opportunity volumes are shown as ranges.
Can I export Topics Portfolio data from Bing AI Performance?
The October post does not mention exporting topic data. The help page says grounding query data, page-level citations and time series data export to CSV and Excel, and that exports reflect the current filter.
Is Bing Webmaster Tools free?
Microsoft’s June 2025 getting-started post describes it as free. You can verify a site by importing it from Google Search Console, a DNS record, an XML file, a meta tag or Domain Connect.
Does Google offer a similar report?
Search Console has a Generative AI performance report for AI Overviews and AI Mode. It counts impressions by page, country, date and device. It does not show queries or topics the way Bing’s report does.
Should I block Bing to avoid AI citations?
That is a business decision for each site. Microsoft says Bing respects preferences set in robots.txt and other supported controls. Read Bing’s documentation on which control does what before you change anything.
