Bing now shows the AI answers where competitors get cited and you don't

On October 13, Microsoft introduced Topics Portfolio, a beta feature in the AI Performance report inside Bing Webmaster Tools. It takes the questions where your site gets cited in AI answers, which Bing calls grounding queries, and groups them into broader themes. Bing's own examples are football, taxes and AI chatbots. Search Engine Roundtable notes there is a drill down view for each topic, with the questions inside it.

Instead of reading hundreds of separate questions, you see your presence by subject: where you are strong, where demand is big and where the field is crowded. Bing says the view is personalized to your site.

Each topic comes with three columns. Bing defines them this way, and every one is an estimate:

• Topic Size: the approximate total citations across all sites for that topic. Think of it as how much AI search demand the topic has.
• Sites Cited: the approximate number of sites showing up for questions in that topic. This tells you how crowded it is.
• Queries: the approximate number of different questions in the topic.

A big topic with few sites cited is the sweet spot: plenty of people asking, not many answers getting the credit. A big topic with hundreds of sites cited is a street fight, and a small business usually wins more by getting specific.

Bing also lists what it calls the strongest questions across your whole portfolio where other sites appear and yours does not. Citation volumes show as ranges, not exact counts.

This is the most useful list in the report. It is, in effect, a list of questions your customers ask an AI where a competitor got the mention. Each one is either a page you should write, a page you have that doesn't answer the question clearly, or a question you can safely ignore.

One warning from Bing itself: branded questions can show up as opportunities. Its example is a shoe site seeing a competitor's brand name on the list. You are not going to rank for someone else's brand, so skip those and move on.

Here is a made up example to show how the pieces fit. Say a dental office opens the report and sees two topics. "Dental implants" is large, with a lot of sites cited. "Emergency dentist" is smaller, with far fewer sites cited, and the office already shows up there a little.

The smart move is not to chase implants first. It is to strengthen the emergency page: same day hours, what counts as an emergency, what it costs to be seen, and how to get in tonight. Then the office checks Opportunity Queries, finds a question like "Can I see a dentist on a Saturday for a broken tooth?", and makes sure a page answers exactly that. Small, specific wins first, the crowded topic later.

Keep expectations honest. Bing says topic labels are generated by AI and may occasionally be wrong. Topics with too little data aren't shown, so a smaller site may see a short list or no tab at all. All the numbers are approximate, and Search Engine Roundtable notes there is no click data yet. Bing's post also doesn't say which AI experiences the citations come from.

So use it as a compass, not a scoreboard. It tells you which direction to walk. It won't tell you how many customers are waiting when you get there.

Bing's own advice is to pick a one, two or three month window, start with the topics where your citations are already concentrated, then work through the opportunity questions. Here is that, turned into steps:

• Set up Bing Webmaster Tools if you haven't. It's free.
• Open AI Performance and look for the Topics Portfolio tab. If it isn't there yet, Bing doesn't have enough data on your site, which is useful to know too.
• Start with your strongest topic and make sure your best page on it answers the top questions clearly, near the top.
• Go through Opportunity Queries, skip competitor brand names, and pick three questions you can answer better than anyone.
• Write or fix one page per question, with real details: your city, your services, your prices or ranges if you share them, and who you help.

Check back in a month and see whether those topics grew. That is the whole loop: find the gap, answer it clearly, check again.

If you want help turning that list into pages that AI tools can find and quote, our founders will walk you through it. Talk to them below. See you in the next edition.