On October 13, 2026, Microsoft announced a beta feature called Topics Portfolio inside the AI Performance report in Bing Webmaster Tools. It groups the grounding queries where your site appears in Copilot-style AI answers into thematic clusters — football, taxes, AI chatbots — so you can finally see the shape of your AI search presence, how big each theme is, how crowded it is, and where you are absent. If you do content or SEO for a mid-sized company, this is the cheapest AI-search research tool shipped this year. Here is how to read it without fooling yourself.
What topics Portfolio actually shows
Bing's AI Performance report, live in public preview since February 11, 2026 and extended in June 2026 with intents, citation share, and site comparison, already showed the grounding queries where your site appears and the pages cited for each. Since OpenAI-style answer engines synthesize rather than rank, that report was the first mainstream window into AI visibility. The AI Shelf Index findings we covered earlier this month showed why this window matters: classic rankings do not guarantee AI citations.
Topics Portfolio adds the zoomed-out layer. The landing page lists the topics where your site has meaningful presence, personalized to your own grounding activity rather than a generic topic list, with three ecosystem measures per topic.
| Column | What it tells you |
|---|---|
| Topic | A thematic cluster of related grounding queries where your site appears |
| Topic Size | Approximate total citations across all sites for the topic — how much AI search demand it represents |
| Sites Cited | Approximate number of sites appearing for the topic's queries — how crowded the space is |
| Queries | Approximate number of distinct grounding queries in the topic |
Below the topic table sits Top Opportunity Queries: the strongest queries across your whole portfolio where other sites appear and yours does not. Click into a topic and you get two tables — Your Grounding Queries (where you already appear) and Opportunity Queries (same topic, competitors cited, you not). Together they give you both halves of a topic: what you own and what is still open.
The beta caveats you must not skip
Microsoft shipped this with the beta label for honest reasons, and the October 13 announcement is unusually explicit about the limitations. Treat the numbers as directional signals for planning, not a precise ledger:
- Topic classification is AI-generated and occasionally imperfect; a query may land in a topic you would have labeled differently.
- Topics with insufficient site presence or ecosystem activity are not shown. Smaller sites may see a short list, or not see the feature at all.
- Branded queries can surface as opportunity queries — Bing's own example is a running-shoes site seeing a competitor brand name flagged as an opportunity.
- Topic Size, Sites Cited, and Queries are approximate, and opportunity citation volumes are deliberately shown as ranges.
Every topic row and opportunity query carries thumbs up and down controls that feed the classification team directly. Use them; that is what the beta is for.
A six-step workflow to turn the portfolio into an editorial plan
- Open a one- to three-month window first. Monthly granularity smooths the classifier noise; the announcement recommends this cadence.
- Start where your citations are already concentrated. Defend established topics before chasing new ones — presence compounds.
- Work the Opportunity Queries top-down. For each, ask whether an existing page should earn the citation or whether the intent genuinely needs a new angle. Cannibalization discipline applies to AI search too.
- Filter branded and misclassified queries out manually. Do not build content plans on a topic assignment you disagree with — flag it with the thumbs down instead.
- Recheck monthly and expect drift. The classifier improves, and AI demand moves fast; a range that held in October can widen by December.
- Export decisions into your content calendar, not just your dashboard. An opportunity query with no resulting page decision was an hour wasted.
How it fits your measurement stack
Topics Portfolio covers one engine: Bing-side AI search, which feeds Copilot. It does not cover ChatGPT or Gemini answers, so it is not a substitute for cross-engine share of model sampling — McKinsey research cited by Writer found just 16% of brands systematically track AI search performance at all, so most teams are starting from zero and should take the free win. Bing's view is also evidence-rich in a way Google's is not yet: Google added AI and multimodal reporting to Search Console in September 2026, but Topics Portfolio gives you named-topic ecosystem context — demand size and competitive crowding, not just your own appearances.
Used well, it slots into the third-party evidence layer of our five-dimension AI visibility framework: define measurable visibility, sample across engines, and now let Bing hand you a prioritized gap list grounded in real grounding activity. For most mid-sized companies the practical takeaway is simpler than it looks in the dashboard: one hour in this report will produce a more defensible content roadmap than a week of generic AI visibility tools, because it is built from your own site's actual citation footprint, in a market where the familiar source is what earns the citation to begin with.
Conclusion
Bing's Topics Portfolio beta, announced October 13, 2026, is the first free tool that shows your AI search presence as a portfolio: themes you own, themes that are crowded, and named queries where competitors are cited instead of you. Its numbers are ranges and its classifier is imperfect, but as a planning input it is immediately actionable, and it will be most useful exactly now, while most brands still track none of this. Spend an hour, export a decision list, and let the gap queries shape next month's content plan.
If you want a measurement stack that actually makes sense — Bing's new portfolio layers, Search Console's AI reporting, and cross-engine citation tracking under one framework — that is what we build at Vivid. Start the conversation on our contact page.
Read next
- AI Overviews tripled on brand searches, today's map of the branded SERP takeover.
- CERT-In: AI agents reached live systems, undetected, today's security wake-up call for agent deployments.
- Citi, BofA and JPMorgan are locking AI agents out, today's agentic-commerce reality check.
- QuickBooks' 223% citation lift: the GEO playbook, an earlier data point on earning citations.
