Pawel Grabowski. Independent consultant

I was recently told about a talk at an upcoming conference: "How to rank first in ChatGPT?"

And as interesting (and certainly click-baity) as the topic sounds, it really begs a question...

What for?

To be clear, I am not questioning appearing in ChatGPT recommendations. Not at all. That is absolutely essential today.

It's the focus on "ranking first" that's a little alarming.

And that's for two reasons.

1/ AI does not have a meaningful equivalent of a #1 Google ranking

It's true that when LLMs recommend products, they often present them as a list. But that's just because it's the easiest way to communicate options.

However, the actual position at which you land isn't that important.

What's important is WHAT LLMs say about your product, in what context, and to whom.

Consider this. You're running a highly specialized management software for consultants. What's more, your solution is best for anyone working solo, focusing more on diagnosis and intelligence than implementation.

Unless the LLM specifically presents your product that way, whether it lists you first or second really makes zero sense.

• Users interacting with AI are not necessarily behaving like searchers scanning a SERP. They are having a conversation. True, it's one with a machine, but just like a normal conversation, what happens next and how the user interprets the response is shaped by everything that came before it.
• They will pick whichever ones match their situation. And here's the kicker, the user might have already provided that context to the LLM. So, the recommendations already match their need to some degree, at least. The rest is up to how relevant they are.

But there is another matter.

2/ The first spot might not exist because there might not be a list

Customers use AI to talk over their issues, ask for advice, run ideas, start a conversation that might or might not lead to product recommendations.

What's more, often, these conversations do not start with "what is the best product for ..." query.

They start with things like:

• I am considering Asana. But is that the best software to manage my dispersed team?
• Are those 40% mechanical keyboards really that effective? I noticed many writers praising them on Reddit. Present me with pros and cons. (That one is mine, a real chat, actually.)
• What are those super well-known dynamic open-back headphones that are famous for big massive wide soundstage?

Notice something about them. None of those prompts require an answer in the form of a list!

ChatGPT could ask more questions to clarify what the user actually wants. Present Asana or another product clearly, and specify what other alternatives exist (and present their context.) Or simply provide the name of those headphones and some additional information behind them.

There is no list. There might be no rankings to achieve at all.

But there is context:

• The context in which your company has been recommended (the user's situation, problem, needs, previous conversation, or buying circumstances)
• The context of the recommendation (what AI actually says about you, why it recommends you, which capabilities it emphasises, how it frames you against alternatives)

And when you're working on AI visibility, these two matter a million times more than whether you have been mentioned first.