OpenAI launches business profiles in ChatGPT with mention and click tracking

A shopper asks ChatGPT which service to choose, then follows up about the terms of your product. If the answer draws on an old page or an incomplete description, your company risks losing that person before their first website visit. OpenAI has offered businesses a way to participate in this conversation: on November 10, the company introduced Business Hub, a dashboard for managing company information within ChatGPT. According to TechCrunch, the feature is currently in beta.

For founders and product managers, this marks a concrete shift. Your website remains your source of information, but a buyer’s introduction to your product may begin in someone else’s interface. OpenAI is now giving companies tools to update information in that interface and observe at least some user interest. There is no evidence yet that the new dashboard increases sales.

What companies will be able to change

Business Hub lets companies claim their business profiles and, after the required checks, manage information about their products and services. It works much like business listings in search engines: companies get a place to keep their information up to date. Eligible businesses will also be able to create a dedicated agent that ChatGPT users can ask about their offering.

Imagine an online store selling coffee machines. A shopper wants to know which model suits a small kitchen and how two machines differ. The store supplies product information, and the agent uses that information to answer questions. The user can clarify details in a conversation. How well the agent handles a particular catalog will need to be tested separately: the feature’s launch alone does not establish that.

The OpenAI documentation describes business ownership verification and identity verification through Persona. It also sets out the conditions for using the dashboard. For a team, this creates an additional operational task: deciding who owns the profile and who will update the information when the offering changes.

Interest analytics do not yet mean sales

Business Hub provides a limited set of business engagement metrics. TechCrunch lists profile views, company mentions and clicks through to the website. These figures may help product managers answer an initial question: are users noticing the product inside ChatGPT, and are they continuing to explore it outside the chat?

But several steps separate a mention from revenue. A user may have seen the company’s name, compared it with a competitor and chosen another option. A click is not the same as an inquiry or a purchase, either. Dashboard metrics should therefore be compared with the company’s own web analytics and sales data. Promising comprehensive attribution based on the launch description would be premature.

For a small SaaS business, a practical experiment might look like this: review the product information, correct the available fields, record baseline metrics, then monitor website visits and inquiries. It is important to choose a measure of value in advance—for example, qualified leads. The team can then assess results against its own funnel, rather than just the number of mentions.

Where business control ends

Verifying a profile does not give a company the right to control every ChatGPT answer about it. The OpenAI terms explicitly state that the platform may supplement and modify information using other sources, and may reject submitted information. The dashboard therefore cannot be treated as a guarantee of preferred wording or a consistent presence in recommendations.

Before publishing an agent, developers would benefit from assembling a set of real customer questions about compatibility, product limitations, returns and disputed terms. Answers should be checked against the company’s current documents. This is a routine quality check for a customer-facing interface, especially when an error could influence a purchasing decision.

The next sensible step for a founder is to check whether Business Hub is available and whether the information about their business is up to date. For a product manager, it is to assign an owner for that data and decide how to measure results. The new dashboard merits a small, testable experiment: it adds a customer touchpoint, but its contribution to each company’s sales remains to be established.