I ran an experiment this week that changed the way I think about AI visibility.
I asked AI to act like a buyer.
Not to tell me what AI visibility is. Not to analyze a website. Not to give me a list of optimization tactics.
I asked it to research companies a real buyer might consider hiring.
The first time, the buyer was looking nationally for companies helping small and midsize businesses improve their visibility and recommendations across platforms like ChatGPT, Gemini, Claude, Perplexity and Google AI.
Then I ran the research again.
This time, I changed the buyer context.
The buyer was now a small or midsize business in the Washington, DC metropolitan area looking for the same kind of help.
The consideration set changed substantially.
That is the interesting part.
AI visibility is contextual
We still talk about AI visibility as though a business either “shows up” or it doesn’t.
But that isn't how buyers make decisions.
They refine.
They start with a category. Then they add their location. Their industry. Their company size. Their particular problem. Their budget. Their requirements.
Every refinement gives AI more context about what “best” or “right” means for that buyer.
And the businesses AI considers can change along with it.
Think about the difference between these questions:
What is the best CRM software?
What is the best CRM software for a professional services firm?
What is the best CRM for a 20-person professional services firm with a long sales cycle and no dedicated RevOps team?
Those aren't three versions of the same search.
They are three different buying contexts.
And potentially three different consideration sets.
That is why I think businesses need to stop thinking about AI visibility as a position and start thinking about it as a consideration set problem.
The question isn't “Do I show up?”
The better question is:
Under what buyer contexts does my business enter the consideration set?
And then:
What happens as the buyer keeps refining the decision?
That second question may be even more important.
A company might appear when AI is identifying possible solutions at the beginning of a buying journey and disappear when the buyer adds a specific requirement.
Or the opposite can happen.
A highly specialized company may not appear in a broad category question at all.
Add the industry, geography or problem it was built to solve, and suddenly it becomes highly relevant.
That isn't necessarily a visibility failure.
It tells us something about how AI understands the business.
This is why the buyer journey matters
I've become increasingly interested in studying AI visibility across the entire buyer journey rather than testing a handful of prompts.
Imagine the buyer moving through this sequence:
Category → Location → Industry → Business size → Problem → Requirements → Comparison → Recommendation
At every turn, the context gets richer.
And at every turn, I want to know:
Who entered the consideration set?
Who stayed?
Who disappeared?
Why?
And what evidence did AI rely on to make that decision?
That gives us far more useful intelligence than asking whether a brand was mentioned in one prompt.
It also changes how I think about the 4 Ps
I've been using IAB's four dimensions to evaluate AI visibility: Presence, Prominence, Framing and Persuasion.
This experiment made their relationship even clearer to me.
Presence: Are you part of the answer for this buyer and this question?
Prominence: How strongly are you represented compared with the other options AI considers?
Framing: Does AI understand what makes you relevant to this particular buyer?
Persuasion: Is there enough credible evidence for AI to move from understanding you to recommending you?
The important phrase in all four is for this buyer.
Because context changes everything.
And yes, SEO still matters
There was another pattern in the research that caught my attention.
Many of the companies AI considered had strong traditional SEO capabilities or foundations.
I don't think that means AI visibility is simply SEO with a new acronym.
It isn't.
But I also don't think businesses should treat SEO and AI visibility as opposing strategies.
AI systems still need to find information, understand entities, interpret pages and connect evidence.
Crawlability matters.
Structured data matters.
Clear entity information matters.
Strong buyer-intent content matters.
Authority matters.
But those things are the foundation, not the entire AI visibility strategy.
I think about the distinction this way:
SEO helps machines find and understand you.
AI visibility helps determine whether they cite, frame and recommend you.
And that second part increasingly depends on evidence beyond your website.
Reviews. Independent coverage. Communities. Expert commentary. Customer experiences. Research. Podcasts. Social conversations. Consistent information across the web.
Your website is one witness.
The rest of the web is testifying too.
So try this with your own business
Don't just ask ChatGPT, Gemini or another AI platform if it knows your company.
Instead, become your buyer.
Start broad.
Then refine the question.
Add the location.
Add the industry.
Add the business size.
Add the actual problem.
Add the buying criteria.
Ask for comparisons.
Finally, ask for a recommendation.
Then study what changes.
When does your business enter the consideration set?
When does it leave?
Who replaces it?
And perhaps most importantly:
What evidence does the AI have about those businesses that it doesn't have about yours?
That gap is where the work begins.
Because the goal isn't to “rank” for one perfect prompt.
The goal is to stay in the conversation as the buyer gets closer to a decision.
Change the question. Change the context. Change who AI recommends.