How Do You Make a Business Visible in ChatGPT Search?

A growing number of B2B buyers now ask ChatGPT, or a similar assistant, who supplies something before they ever type a query into Google. For a business used to thinking about search purely in terms of ranking pages, this raises a genuinely different question: how does a company make itself visible to a system that does not display a results page, does not accept payment for placement, and does not publish a ranking algorithm anyone can study.

The honest answer is that nobody outside the companies building these systems knows the full mechanics, and those mechanics change. What is known, and what is consistent with how large language models and their retrieval layers work, is that they draw on a mixture of training data, indexed web content and, for tools with live browsing or search plug-ins, real-time retrieval. A business increases its chances of being mentioned by making sure all three of those inputs have something clear and consistent to work with.

This article sets out what is reasonably well understood, what remains platform-specific or uncertain, and what a B2B company can practically do — without pretending that any agency, including Evans, can guarantee a mention.

What ChatGPT is actually drawing on when it answers

ChatGPT and comparable assistants combine several sources depending on the version and mode in use: the model's training data, which is a snapshot fixed at some point in the past; and, for tools with browsing or search capability, live retrieval from the web at the moment of the question. When live retrieval is involved, the system is behaving more like a research assistant that reads a handful of pages and summarises them than like a search engine returning ten blue links. That means being mentioned depends less on beating competitors in a ranking and more on being one of the sources the retrieval step actually finds and judges usable.

This has a practical consequence: a company that is difficult to describe in one clear sentence — because its website buries what it does in vague language, or its services are described differently on different pages — gives an AI system very little confident material to draw on, even if a human reader could work it out with effort.

The five questions companies ask about visibility

Why is my business not appearing in ChatGPT?

Usually because there is not enough clear, consistent, independently corroborated information about the company available to be retrieved and trusted — not because of a technical penalty. A company with a thin or vague website, few independent mentions, and no consistent description of what it does across the web gives an AI system little to work with. This is different from a Google penalty; there is no equivalent of a manual action to appeal.

Can SEO help?

Partially. Technical SEO fundamentals — a crawlable site, clean HTML, fast loading, clear internal linking — help any automated system find and read content, AI retrieval included. But classic ranking tactics built around backlinks and keyword density do not reliably translate into AI citation, which appears to reward clarity and specificity at the passage level more than domain-wide authority scores.

Does schema markup help?

Structured data (schema.org markup for Organization, Product, FAQ, LocalBusiness and similar types) gives machines an explicit, unambiguous statement of facts that would otherwise have to be inferred from prose. It is reasonable to expect this to help AI systems resolve who a company is and what it does with more confidence, though no platform publishes exactly how much weight it carries, and it is not a substitute for clear content.

Do reviews help?

Independent reviews and third-party mentions act as corroboration — a second source confirming what a company says about itself. This is consistent with how these systems are understood to weigh trust generally. Exactly how heavily any one platform factors reviews into an AI answer is not published, so this should be treated as a plausible, sensible signal rather than a guaranteed lever.

Does location matter?

For a query with a location component — 'supplier in Yorkshire', 'installer near me' — a business needs its location stated clearly and consistently across its website and third-party listings for a location-based answer to include it. For non-local B2B queries, location signals are far less relevant than being a clear, well-corroborated answer to the underlying question.

What is realistic to expect

Practical steps that are within a company's control

• State plainly, in one consistent sentence repeated across the site, what the company does, for whom, and where
• Answer the specific questions real buyers ask, as direct, self-contained passages rather than long unfocused pages
• Add structured data confirming organisation, location, products and FAQs
• Encourage and maintain genuine third-party reviews and directory listings, kept consistent with the website's own claims
• Keep the site crawlable and fast, with clear internal linking between related pages
• Build a small number of genuinely useful, specific pages rather than a large volume of thin ones

Where this fits into a wider digital strategy

AI visibility is not a standalone technical project. It depends on the same foundations as a credible commercial website: clear positioning, honest and specific content, and a structure that makes the business easy to understand quickly — for a human buyer or a machine. Evans Sales Consultancy treats AI search visibility as part of the wider digital infrastructure that supports market entry, positioning and sales, not as an isolated technical fix.