AI visibility is still a discoverability problem.
Many firms talk about AI visibility as if it were something entirely new. In practice, the same structural issues often sit underneath it: weak semantic clarity, unclear page relationships, fragmented trust signals and offer structures that are difficult to interpret. If a site is hard to understand, it is less likely to be surfaced confidently by systems that need to retrieve, interpret and summarize information at speed.
Why structure matters more than AI messaging.
Adding AI language to a website does not make the business easier for AI systems to understand. What matters more is whether the site clearly expresses what the firm does, who it serves, which pages matter most, how services relate to one another and where authority is reinforced.
Why many B2B firms are difficult to interpret online.
Many B2B websites reflect the history of the business rather than the logic of the offer. Pages accumulate over time. Navigation expands without a clear hierarchy. Service language becomes too internal or too generic. Expertise exists, but it is distributed across disconnected signals. The result is not always low quality. More often, it is low legibility.
What firms should fix first.
The first step is not usually more content. It is a clearer structure: sharper service language, stronger page roles, more coherent internal linking, clearer hierarchy and better alignment between real-world authority and visible digital signals.
AI tools do not solve structural problems.
AI can accelerate output, but it does not remove the need for structural clarity. In complex B2B environments, visibility still depends on whether the business is expressed clearly enough to be interpreted, trusted and connected to the right signals. That judgment cannot be automated.
How AI systems actually search and why one query is rarely enough.
When a user asks ChatGPT or Perplexity a question, the system typically does not rely on a single search. It breaks the request into several background lookups, each testing a different part of the intent. A question such as "which B2B software firms in Switzerland offer enterprise data integration" may trigger separate retrieval paths around company type, geography, use case and credibility signals. The answer is then synthesised across those results rather than pulled from one simple query.
This changes what visibility actually requires. A firm that ranks for one obvious keyword may still be absent from the final answer if it does not appear across the related queries the system is using to interpret the request. Structural clarity, entity recognition and coherent service language are what make a firm legible across that wider set of signals.
For most B2B firms, that is a structural problem before it is a content problem.
Why Bing matters more than most B2B firms realise.
Research into which search engine most influences ChatGPT recommendations points consistently to Bing, not Google. In one study, 87% of brand citations in ChatGPT answers aligned with Bing's top results rather than Google's. A brand could rank well on Google and still be largely absent from AI-generated answers simply because its Bing presence was weak.
For B2B firms focused exclusively on Google, that is a meaningful gap. In the context of AI visibility, Bing is not a secondary concern.