For years, AI visibility measurement meant sampling ChatGPT yourself — or paying a vendor to run prompts and count mentions. On 10 November 2026, OpenAI flipped that model: Businesses in ChatGPT went into beta, and Business Hub became the first place an AI assistant reports how often it named your brand and how often someone clicked through.
That is a structural shift. Mentions are the product outcome SEO teams have been proxying with third-party trackers. Now the engine that generates the answer also ships the scoreboard — with limits.
What OpenAI launched
OpenAI's Businesses in ChatGPT program lets verified businesses:
- Claim a Business Profile (phone or DNS verification)
- Edit structured business data and product feeds
- View Business Hub metrics: mentions, profile views, clicks to the website, and discovery topics (aggregated themes of conversations where the business appeared)
- Optionally publish a Business Agent (US ecommerce and online services) and use it as a destination for ChatGPT Ads
What Business Hub does not show: raw prompts, conversation text, competitor names, or per-query breakdowns. OpenAI aggregates to 28-day windows (live dashboards seen in the wild used "data through" dates in UTC). Small numbers are suppressed behind undisclosed thresholds.
The UI is explicit about attribution limits. Tooltips state that clicks do not confirm website visits or conversions — a click is a click on a link in ChatGPT, not GA4-validated traffic.
The first numbers we can compare
Until launch week, the only public Business Hub figures were marketing screenshots. PPC Land and Search Engine Roundtable published arithmetic on live and example dashboards by 11 November 2026:
| Dashboard source | Period | Mentions | Clicks | Clicks per mention |
|---|---|---|---|---|
| OpenAI example business | 7 days | 15,356 | 5,432 | ~35% |
| BuildUp Bookings (live) | 28 days | 690 | 24 | ~3.5% |
| Zianvo (live) | 28 days | 28 | 1 | ~3.6% |
OpenAI's illustrative row is roughly ten times more click-efficient than the two live accounts journalists could verify on launch week. That does not prove the UI is wrong — it proves mention-rich and click-rich are not the same problem.
Most brands in AI answers are named without a prominent link. Semrush's cross-engine work (cited in How to Measure AI Visibility KPIs Beyond CTR) found ChatGPT cites brands in 87% of appearances but names them in only ~21% of answers — the inverse pattern on Gemini. Business Hub foregrounds mentions and profile engagement; your site analytics still carry referral rarity.
Why engine-owned metrics change the workflow
1. Mentions become a first-class KPI — for one platform
Third-party AEO tools estimate mention rate by re-running prompts. Business Hub reports what ChatGPT's own logging sees — closer to ground truth for in-product recommendations, with aggregation and privacy filters applied.
That helps teams stop arguing about sample size on ChatGPT specifically. It does not replace multi-engine observation. Only 2.7% of domains are cited by all five major engines. ChatGPT's dashboard says nothing about Gemini, Perplexity, or Google AI Mode.
2. Verification becomes distribution infrastructure
Claiming a profile is not vanity. OpenAI positions verified feeds and business metadata as inputs to how ChatGPT represents products, locations, and services. That parallels Google's entity graph story — but here the verification gate is inside the assistant.
Teams that delay verification leave default or third-party descriptions in the retrieval path. Early movers get editing rights and Hub visibility; laggards still appear in answers shaped by web crawl and licensed data — see How ChatGPT and Claude Answer "Near Me" Queries.
3. Discovery topics hint at prompt clusters — not keywords
"Discovery topics" are thematic buckets, not search-query strings. They are useful for positioning and content gaps ("we show up in conversations about X but not Y") but useless for classic keyword reporting. Pair them with repeated buyer-question runs in your category, not Search Console queries.
4. Business Agents and ads stack on the same identity
US ecommerce and online-service businesses can publish a Business Agent and sponsor it in ChatGPT Ads. Visibility strategy splits:
- Organic mention growth — earned evidence, reviews, third-party coverage (earned media still dominates citation mix)
- Paid and agent surfaces — controlled profile, scripted paths, measurable clicks
The Hub will increasingly blend both. Mention rate without click rate may still be a win when the answer recommends you by name — the citation gap is common.
How to read Business Hub without fooling yourself
- Track mentions and clicks together. A high mention count with near-zero clicks matches ChatGPT's cite-heavy behavior; optimize for named recommendations and profile completeness, not only landing-page CTR.
- Do not annualize OpenAI's demo row. Use your own 28-day windows; compare week over week after material content or PR moves.
- Keep independent sampling running. Hub metrics can diverge from what buyers see on edge prompts, logged-out users, or new model releases. Cross-check with a fixed category prompt set — the same discipline as measuring without fooling yourself.
- Reconcile with analytics cautiously. ChatGPT referral traffic remains small relative to organic search for most sites (recent panel data). Branded search lift often catches influence Hub clicks miss.
What this does not solve
No competitive benchmark. You cannot see share of voice versus rivals inside Hub.
No prompt transparency. Aggregation protects privacy but blinds SEO forensics.
No multi-engine view. Google's generative performance reporting and ChatGPT's Hub are siloed — expect separate AEO strategies per engine.
Selection bias on who can claim. Beta rollout and waitlists (including non-US businesses reported on launch week) mean early public dashboards skew toward digital-native ecommerce — not every B2B brand.
Practical playbook
This week
- Check eligibility and claim the profile if your category is supported.
- Audit business metadata and product feeds for the same facts your best sales page states — inconsistency breeds silent wrong answers.
- Screenshot baseline Hub metrics before your next PR or product launch.
This quarter
- Build a mention KPI for ChatGPT alongside citation KPIs from sampled prompts on other engines.
- Map discovery topics to content and earned-media targets — one topic → one evidence asset (review platform, comparison article, docs page).
- Train stakeholders: mentions are awareness; clicks are optional on ChatGPT relative to Gemini-style naming-heavy answers.
For agencies
Standardize client reporting: Hub (ChatGPT mentions/clicks) + independent corpus (competitor mention rate, source-to-mention). See agency playbook.
How Obsurfable fits
Business Hub covers one engine's logged view of your brand. Obsurfable's Explorer corpus records prompts, answers, brands mentioned, and citations across buyer questions — browseable and comparable over time. The App matches that evidence to your company for mention rate, share of voice, and source-to-mention, so you see whether Hub gains show up in the answers your category actually asks — and whether competitors move on prompts ChatGPT never reports.
FAQ
Should we stop using third-party AEO tools now that ChatGPT has a dashboard?
No. Use Hub as authoritative ChatGPT mention and click data where you qualify. Keep independent observation for other engines, competitive share, and prompt-level drift.
Why are live click rates so much lower than OpenAI's example?
Likely mix of business type, answer format (name-only vs linked), profile completeness, and threshold suppression on small accounts. Treat the example as illustrative UX, not a benchmark.
Does claiming a business change whether ChatGPT recommends us?
OpenAI describes verified data as an input to representation; it has not published causal lift studies. Claiming is table stakes for accurate profiles and metrics — not a substitute for earned evidence.
Is Business Hub available everywhere?
Beta expanded on 10 November 2026 with US-centric Business Agent and ads features; some international businesses reported waitlists. Check OpenAI's current eligibility page before planning workflows.
Bottom line
ChatGPT Business Hub is the first engine-owned AI visibility dashboard — mentions and clicks from the assistant that generates the answer. Claim profiles early, read mentions and clicks together, and do not confuse OpenAI's demo efficiency with your live funnel. Multi-engine brand visibility still requires independent observation; Hub is one disciplined input, not the whole scoreboard.