Why AI visibility is NOT another optimisation discipline and why it should concern CMOs and GTM leaders

Date : September 14, 2026 By Andrei Iunisov

For years B2B digital marketing followed a reasonably predictable model: a potential customer had a problem, searched Google, clicked several results, visited vendor websites, downloaded content, compared options and eventually spoke to sales.

Marketing teams built their acquisition strategies around this journey. SEO generated rankings and traffic. Paid search captured high-intent demand. Content moved visitors through the funnel. Analytics showed which channels generated leads and opportunities.

That model has not disappeared, but it is becoming incomplete. More buyers can now perform a significant part of their research without visiting your website at all. They can ask ChatGPT, Gemini or Claude questions such as:

• What are the best solutions for my problem?

• Which vendors should I consider?

• How does Vendor A compare with Vendor B?

• What are the alternatives to the market leader?

• Which solution is appropriate for a company of my size?

• Which vendors support my technology stack or regulatory requirements?

The AI system can research the market, compare companies, summarise reviews and recommend a shortlist. As a result, the website visit may happen much later in the buyer’s journey. In some cases, it may not happen at all.

This changes the marketing question from:

“How do we get more people to our website?”

to:

“How do we make sure our company is part of the buyer’s consideration set before they even reach our website?”

A useful look inside how ChatGPT finds information

A recent analysis by Metehan Yeşilyurt and the Peec AI team provides an interesting view into how this process may work technically.

While analysing ChatGPT’s web retrieval behaviour, the team found data showing the search queries generated by ChatGPT, the pages it retrieved and the individual passages selected from those pages.

Their conclusion was particularly interesting: ChatGPT does not necessarily evaluate a web page as one single unit.

It can break content into passages and evaluate the relevance of each passage to the user’s question. In the example analysed by the team, the passages that survived this retrieval process were then provided to the model and used to help generate the final answer.

AI systems need to understand what your company does, why it matters and whether it is relevant to a specific customer question.

Your website needs to explain the business, not just market it

Many B2B websites still contain surprisingly little useful information.

They say things such as:

We deliver innovative solutions that empower organisations to achieve their full potential.

Or:

Our industry-leading platform helps companies accelerate digital transformation.

These sentences sound acceptable, but they communicate almost nothing.

• Who is the product for?

• What specific problem does it solve?

• When should somebody use it?

• What alternatives exist?

• How is it different?

• What systems does it integrate with?

• What does implementation involve?

• What evidence supports the claims?

Historically, vague marketing language could sometimes survive because the main objective of a page was to rank for a keyword and persuade a human visitor to click a CTA. AI research creates another requirement. Your content also needs to provide clear, extractable answers to the questions customers ask while evaluating the market.

The aforementioned analysis found that individual passages could be selected based on their relevance to a particular query. The researchers summarised the implications using 3 concepts: semantics, originality and authority.

From a C-level perspective, it could even simplier:

Every important part of your market story should make sense on its own.

If your competitive advantage requires someone to read 4 pages and interpret several vague claims before they understand it, you have a positioning problem as much as a content problem.

AI search makes positioning more important, not less

There has been a tendency to discuss AI visibility as another optimisation discipline. First we had SEO, then AEO, GEO, LLM optimisation and various other acronyms. There is certainly technical work involved. However, companies can easily miss the bigger issue:

You cannot optimise your way out of weak positioning.

Imagine asking an AI assistant:

What are the best procurement platforms for mid-sized manufacturing companies?

To recommend your company confidently, an AI system needs evidence connecting several concepts:

• your brand,

• procurement software,

• mid-sized businesses,

• manufacturing,

• relevant capabilities,

• and ideally evidence that other credible sources recognise those connections too.

If your website describes the company as an “AI-powered digital transformation platform for modern enterprises”, the problem is not primarily technical SEO. The problem is that your positioning is unclear. This is why AI visibility should concern CEOs, CMOs and GTM leaders rather than sitting exclusively with the SEO/GEO team.

It exposes whether your company’s market narrative is actually understandable.

The new funnel starts before the click

The traditional digital funnel might look like this:

Google search → website visit → content → conversion → sales

A growing number of buying journeys may instead look more like:

Business problem → AI research → vendor discovery → comparison → shortlist → branded Google search → website validation → webinar download → demo → remarketing campaign → sales

The website still matters enormously, but its role can change. Instead of being where the customer discovers you, it may become where they validate a decision already partially formed elsewhere.

That distinction matters because a buyer arriving after an AI recommendation may already know:

• what category you operate in,

• your main competitors,

• your supposed strengths,

• your approximate positioning,

• and potentially some concerns about your product.

Your website now needs to confirm, deepen or correct that understanding. Marketing teams therefore need to start paying attention not only to traffic acquisition, but also to what happens before the traffic exists.

Marketing teams need an AI visibility feedback loop

I do not think companies need to panic and rebuild their entire marketing strategy around ChatGPT or other AI engines. However, they should start measuring what AI systems understand about their business. A useful process is relatively straightforward.

• First, identify the commercial questions that matter during the buyer’s journey.

• Then monitor the answers across major AI platforms.

• Look at:

1. whether your company appears,

2. which competitors appear,

3. how the company is described,

4. what strengths are associated with your brand,

5. what weaknesses are mentioned,

6. which sources are influencing the answers,

7. and which topics consistently exclude you.

Then compare those findings with your existing GTM strategy.

Perhaps the problem is missing content. Perhaps your positioning is not clear. Perhaps competitors have much stronger third-party authority. Perhaps review coverage is weak. Perhaps your product is strong, but almost nobody outside your own website discusses it. Perhaps AI systems misunderstand what you actually offer.

These are not merely “GEO problems”. They can reveal weaknesses across the entire go-to-market system.

What should marketing teams do next?

My recommendation would be to treat AI visibility as another measurable part of the buyer’s journey, not as a separate experimental marketing channel.

I would start with a relatively small set of commercially important prompts; the questions prospects actually ask before they shortlist a vendor, request a demo or speak with sales. Run those questions across the main AI platforms and document where your company appears, how it is positioned and which competitors consistently appear instead.

Then identify the reasons behind the gaps.

If the positioning is unclear, improve the messaging. If important questions are not answered on the website, create or update the relevant content. If competitors have stronger third-party authority, review where they are being mentioned and whether similar opportunities exist for your company. If AI systems repeatedly describe your product incorrectly, analyse which sources are contributing to that understanding.

The objective should not be to increase an abstract AI visibility score. The objective is to improve your probability of being considered by the right buyers. If your target customer is using AI to build a shortlist, your job is to make sure your company has a credible reason to be on it.