Does ChatGPT Know Your Brand? A 20 Minute Test for Marketing Leaders

Open ChatGPT and ask it to recommend companies in your category. If your brand does not appear, buyers using AI to shortlist suppliers are not seeing you. The test below takes about 20 minutes. The fix is not technical. AI systems repeat what other credible sources say about you, which means the problem is usually that not enough other sources say anything.

Something Changed and Most Companies Missed It

Research suggests around 82 percent of consumers now find AI powered search more helpful than traditional search. Only about 22 percent of marketers track whether their brand appears in AI responses at all. That gap is the whole story. Buyers moved. Measurement did not.

In 2012, a brand was invisible if it was not on page one of Google. In 2026, a brand is invisible if ChatGPT does not mention it when someone asks for recommendations in that category. The difficult part is that nobody tells you. There is no ranking report for this. A company can lose a shortlist position for two years without ever knowing it happened.

Where AI Actually Gets Its Answers

This is the part most people get wrong, and it changes what you should do about it. AI assistants do not simply read your website and summarise it. They pull from many sources and weigh them by how credible and independent those sources appear.

One analysis of more than 5 million citations across Google AI Overviews, ChatGPT, Claude and Perplexity found Reddit sitting in second place among the most cited domains. Google signed a reported 60 million dollar deal with Reddit in 2024 to use it as a training source. OpenAI signed a similar arrangement shortly after.

So when someone asks an AI which companies to consider, part of that answer is shaped by what strangers said about you in a forum thread two years ago. You did not write it. You may not know it exists. It is part of your brand anyway.

The 20 Minute Test

Do this yourself. Do not delegate it, because the value is in seeing the answers directly.

Open ChatGPT, Perplexity, Google AI Mode and Gemini. Run the same prompts in each. Write down what comes back.

Prompt 1: The category question

"What are the best [your category] companies in [your market]?"

This is the shortlist question. If you are not here, you are not being considered.

Prompt 2: The problem question

"I need to [the problem you solve]. Who should I talk to?"

Buyers often describe a problem rather than a category. Different phrasing, different answer.

Prompt 3: The direct question

"What do you know about [your company name]?"

Now check three things. Does it know you exist. Is what it says accurate. Is the tone positive, neutral or negative.

Prompt 4: The competitor question

"What are the alternatives to [your strongest competitor]?"

If they are the reference point and you are not on the alternatives list, that is a specific and fixable problem.

Prompt 5: The founder question

"Who is [your founder or CEO name]?"

For organisations where the leader carries reputation, this matters as much as the company question.

Reading Your Results

Write your answers into a simple grid.

Four outcomes are possible, and each means something different.

Not mentioned at all. The most common result, and the least alarming. It usually means not enough independent sources describe what you do. This is a coverage problem.

Mentioned but described wrongly. More urgent than being absent. AI is confidently telling buyers something inaccurate about you. This is usually caused by inconsistent information across your own profiles.

Mentioned but buried. You appear seventh in a list of ten. You are in the consideration set and not in the shortlist. This is a credibility weighting problem.

Mentioned accurately and early. Good. Now check the other three platforms, because results vary considerably between them.

The Most Common Cause, and It Is Not Technical

When we run this exercise with organisations, the same root cause appears again and again. Their own sources disagree with each other.

The website says the company works in 40 countries. A directory listing says 37. LinkedIn gives the founder one job title, a speaker profile gives another, and a third site gives a third. One page describes the company one way, another page describes it differently. None of these are lies. They are just old, or written by different people at different times, and nobody ever reconciled them. An AI system reading all of this cannot form a confident description. And when it cannot describe you confidently, it does the safe thing. It describes somebody else instead.

This is the fix that costs nothing and helps most. Write one description of your organisation, with one set of numbers. Then put exactly that text everywhere. Website, LinkedIn, every directory, every profile, every partner page.

Not a rewritten version for each platform. The same text. Consistency is what allows a machine to state something as fact.

Why This Is a Brand Problem, Not an SEO Problem

Most advice on this subject comes from technical marketers, and it focuses on technical fixes. Structured data, machine readable files, page formatting. Some of it helps at the margins. But it misses what is actually happening. AI systems cite you based on what other credible sources say about you. That is not a new idea. That is what brand reputation has always been.

A brand was never what a company said about itself. It was what other people said about it when the company was not in the room. Every brand strategist has been saying this for thirty years. What changed is that those conversations are now machine readable, and a machine repeats them to your buyer in seconds.

So the work is the same work it always was. Be known for one clear thing. Have credible third parties confirm it. Be consistent everywhere. We set this out in more detail in our guide to brand positioning maps, which is the same discipline applied to a different surface.

What to Do With What You Found

Four actions, in order of value.

1. Fix your own inconsistencies first. This takes a week and costs nothing. One description, one set of numbers, everywhere. Until this is done, everything else is building on sand.

2. Get independent sources to describe you. AI weighs third party mentions far more heavily than your own website. Industry association listings. Trade publications. Podcast appearances. Client case studies published on the client's own site. Speaker directories.

One genuine industry body listing is worth more than fifty directory submissions, because it implies someone checked. This is the same logic that governs whether a corporate brand is believed internally, applied to an external audience that happens to be a machine.

3. Publish something only you have. AI cites what it cannot find elsewhere. Original data from your own work. A framework with a name. Patterns you have seen across clients that nobody else has visibility into.

Generic content gets summarised. Original content gets cited.

4. Show up where people actually ask. If your buyers discuss their problems in a public forum, and nobody from your organisation has ever contributed usefully there, you are absent from a source AI reads heavily.

This has to be genuine participation. Dropping links gets you removed, and removed content helps nobody.

What Not to Bother With

Two things are getting recommended widely and are worth skipping.

Machine readable files aimed at AI crawlers. Google has stated publicly that it does not use these for Search, including AI Overviews and AI Mode. Some other systems do read them. It is a ten minute job with a small payoff, so treat it as optional rather than a priority.

Buying mentions. Paid placements designed to manufacture citations are the modern version of buying backlinks. It worked briefly last time too.

Run This Every Quarter

Results shift as models update and as the sources they read change. Keep your grid. Run the same five prompts every three months and compare. You are looking for movement, not perfection. Three things worth tracking each time: whether you appear at all, whether the description is accurate, and which competitors appear alongside you. That third one is often the most useful, because the companies AI groups you with are the companies your buyers are comparing you to, and that list is frequently not the one you expected.

For organisations working through this properly, it usually connects back to positioning rather than promotion, which is where our brand strategy and management work begins.

The Uncomfortable Conclusion

If AI does not mention you, the instinct is to look for a technical explanation. Usually there is not one. Usually it means that not enough credible independent sources have anything specific to say about what you do, and the sources that do exist disagree with each other. That was a brand problem before AI existed. It is simply visible now, and it is visible to your buyer in about four seconds.

About Global Brand Academy

Global Brand Academy is a global strategic training and transformation partner, working with organisations, leaders, executives, sales professionals and entrepreneurs to build capability where it matters most across brand, leadership, culture, sales, customer experience and AI. Built on over 30 years of real world experience across 40+ countries, GBA has served more than 1,000 organisations and impacted over 1.2 million leaders, professionals and entrepreneurs. Our work is anchored around seven transformation domains and delivered through an integrated ecosystem of keynotes, masterclasses, coaching and certification, so learning moves from insight to capability and from intent to impact.

Global Brand Academy is founded and led by Dr Jerome Joseph, an award winning global speaker and author of twelve books, whose doctoral research examined the impact of brand and culture on organisational performance. For organisations discovering that AI describes them poorly or not at all, that background is directly relevant, because the cause is almost always an unclear or inconsistent brand position rather than a technical one.

Frequently Asked Questions

How do I check if AI mentions my brand?

Open ChatGPT, Perplexity, Google AI Mode and Gemini, and run the same five prompts in each. Ask for the best companies in your category and market, describe the problem you solve and ask who to talk to, ask directly what it knows about your company, ask for alternatives to your strongest competitor, and ask about your founder. Record whether you appear, in what position, and whether the description is accurate.

Why does AI not mention my company?

Most often because not enough independent third party sources describe what you do, or because the sources that exist contradict each other. AI systems weigh credible outside mentions more heavily than your own website, and when the available information is inconsistent they avoid stating anything confidently about you.

How do I get my brand mentioned in ChatGPT?

Start by making every description of your organisation identical across your website, LinkedIn, directories and profiles, including all numbers. Then build genuine third party coverage through industry bodies, trade publications, podcasts and client published case studies. Publish original material such as proprietary data or a named framework, since AI cites what it cannot find elsewhere.

Is AI brand visibility the same as SEO?

They overlap and they are not the same. SEO optimises a page to rank. AI visibility depends on how many credible independent sources describe your brand consistently, which is closer to reputation than to page optimisation. Technical fixes help at the margins, but the substantial work is brand and credibility work.

How often should we check AI brand visibility?

Quarterly is sufficient for most organisations. Models update and the sources they draw on shift, so results change over time. Use the same prompts each time so the comparison is meaningful, and track whether you appear, whether the description is accurate, and which competitors appear alongside you.

Does adding an AI text file to my website help?

Google has stated publicly that it does not use such files for Google Search, including AI Overviews and AI Mode. Some other AI systems do read them. It is a small job with a small payoff, so it is reasonable to do but should not be treated as a priority ahead of consistency and third party coverage.