I Asked ChatGPT the Same Five Questions About a Category — Who Gets Mentioned? The Answer Surprised Me

A few days ago I read a report that stopped me in my tracks.

1,094 product categories. For each one, they asked ChatGPT five common buyer questions — what is this thing, how does it compare to the competition, are there alternatives, what's it good for, which one should I buy.

Then they tallied which brand came up most often — and most consistently — in each category.

Guess what they found?

Only 15% of categories produced a "clear winner."

The remaining 85%? Either several brands were locked in a tug-of-war, or no brand showed up consistently at all.

Think about that.

What does that tell us?

An Overlooked Fact: There Is No Overlord in AI Search

Let's first define what we mean by a "category winner."

This study set a brutal bar. To win, a brand had to be mentioned by ChatGPT in at least four of the five questions for the same category — and to beat the second-place brand by at least five percentage points.

Only then did it count.

Out of 1,094 categories, only 166 cleared that bar.

166. Sounds like a lot — until you remember that's 166 out of 1,094.

In other words, when you ask ChatGPT five related questions, eight or nine times out of ten, a different brand pops up each time.

Why does this matter?

Because all those years you spent on SEO — wasn't the whole point to have a user search a keyword and see you first? Now the search box has become a chat box. Does the work you did still count?

The Busier the Category, the Less Anyone Can Hold It Down

The study also turned up something counterintuitive.

They sorted the 1,094 categories by ChatGPT search popularity, then split them into a top half and a bottom half.

The high-popularity half — the things people ask about every day — only 11.3% produced a winner.

The low-popularity half, on the other hand, came in at 19%.

And that high-popularity half accounted for 98% of all the search volume in the entire study.

What does that mean?

Where everyone is asking the most questions, it's a free-for-all scramble — no single brand can dominate the space.

Think about it: this completely flips the search market you thought you understood. In the old world, the big keywords were occupied by giants, and small brands only survived on long-tail keywords (rare, low-volume search terms). Now? On the hot keywords, everyone has a shot — nobody gets to run away with it.

That's a good thing. A really good thing.

Old SEO Metrics Can't Explain the New Battlefield

So do brands that do well at traditional SEO also do well on ChatGPT?

The researchers went ahead and compared three legacy metrics: brand-name search volume, organic traffic, and authority score.

Only one of them correlated consistently with being a "category winner" — brand-name search volume.

Organic traffic and authority score had almost nothing to do with whether ChatGPT mentioned you.

And here's a number that stings even more. Among the most-cited domains, only 21% were also the brands that showed up most frequently inside ChatGPT's answers.

In other words, being cited does not equal being remembered.

Being cited means ChatGPT thinks your article is useful. Being remembered means ChatGPT thinks this category belongs to you.

Those are two different things.

A lot of teams are still competing over who gets cited more — and that's already the wrong direction.

Once a Winner Digs In, Though, It's Hard to Knock Them Off

There was one set of numbers in the study I turned over in my head for quite a while.

Brands that had already established themselves as a "category winner" — when tracked month over month, more than 90% of them were still holding the top spot.

But for the "emerging leader" and "unsettled landscape" categories, the leading brand swapped nearly 2,000 times during the study period.

2,000 times.

The closer the race, the more frequent the swaps.

See the pattern? This market is not monolithic.

Where a mind-share monopoly has already formed, it's hard for anyone else to dislodge the leader. But in the 85% where no monopoly has formed yet, the deck is being reshuffled every single day.

That's the opening.

Not the "if I just grind on SEO I'll rank first" kind of opening. The "if I push in right now, I might become one of that 15%" kind of opening.

So What Should Brands Actually Do?

After I finished the report, I walked away with three fairly direct takeaways.

First, stop obsessing over your ranking in any single answer. When a user buys something, they don't just ask ChatGPT one question. They'll work their way from "what is this" all the way to "which one should I buy." Ranking first in one answer is useless. What you need is to show up consistently across five related answers.

That's the new "dominating the results."

Second, brand-name search volume matters more than you think. It's essentially the only traditional metric that lines up with ChatGPT's perception. Put bluntly: whether people actively search for your name determines whether ChatGPT, when it talks about a category, brings you to mind.

Third, those 85% of categories with no winner — right now is exactly when you should be entering them. Where the winner is already locked in, pushing hard is a waste. But where it's a free-for-all? Anyone has a chance.

I've pasted the link to the study here — no signup required.

Go feel it out for yourself.

The biggest thing I took away from it: search is shifting from "who ranks first" to "who keeps getting remembered." It used to be a game of position. Now it's a game of mind share.

And mind share is not something you can ever manufacture with technical tricks.

It comes from genuinely owning a category — so that the moment anyone thinks about this thing, they think of you.

What do you think?