There are three slots. Nationally that's a lottery. In your city it isn't.

Start with the number that decides everything: an AI answer names three to five companies.

Not ten blue links. Not a page two. Three to five, in a sentence or a short list, and then the answer ends. There is no scrolling. Whoever is not in that set was not in the conversation.

A June 2026 analysis measured how those slots distribute across 91 companies in three national categories, using ten models and three hundred prompts. The result is not a bell curve, where almost everyone sits in the middle. It is a ladder where the top rung takes nearly everything.

Out of a hundred answers, how many the category leader takes, and how many the company in the middle of the pack takes:

Category            Leader        The leader   The middle of the pack
Observability       Datadog        78 in 100                 9 in 100
Outbound sales      Apollo         61 in 100                10 in 100
AI coding tools     Copilot        55 in 100                 8 in 100

The top name is in nearly every answer. The fifteenth is in almost none. Other work, by people with no stake in it, points the same way: across a large set of questions, the handful of sites at the top of AI Overviews took the overwhelming majority of citations and everybody else together took almost none.

That study’s conclusion, for anyone who is not first or second, is that writing a few more pages will not close a gap of forty answers in a hundred, and that the only road left is becoming the definitive reference for one narrow niche.

It is right about the arithmetic, and it stops one step before the useful part.

That lopsidedness is made by the question, not by AI

Here is what the national numbers actually measure: what happens when you ask a question tens of thousands of companies could answer, and narrow it down by nothing.

“What’s the best observability tool?” puts the whole world in the running. Every company on earth that does the thing is competing for three slots. Of course the same names keep winning: the model has nothing to narrow the field with except prominence, and prominence is exactly what the biggest company has most of. That pile-up is not something AI does. It is what a question with no limits does.

Now add a constraint the customer was always going to add anyway.

“Who’s the best plumber in Tulsa?” Now the field is not tens of thousands of companies. It is a few dozen. And the model still names three to five.

The same slots. A hundred times fewer competitors.

That is the entire argument, and it is why a national visibility report and a local one are different products even though they run the same way.

What changes when the candidate set is small

Three things, and they all move in your favor.

The gap between first and fifth is small. With forty companies in the running and five slots, the difference between being in the answer and being out of it cannot be that wall of seventy answers in a hundred. A field that size has no room for one. How small it is in your category is not something to guess at, it is something to measure, and that is what the report does.

Prominence stops dominating. The national leader’s advantage is that it is written about constantly, everywhere. Nobody is written about constantly in Tulsa. When nobody in the field is overwhelmingly present, small differences in how readable you are, how current you look, and how other people describe you decide the order. Those are the things you can change.

Nobody is defending. The national top three employ people whose job is this. In a city category, the companies in the answer are usually there by accident. They have not run the test. They do not know they are in it. They are not maintaining anything.

                                      National      City and category
Companies in the running               10,000+                20 - 60
Slots in an answer                       3 - 5                  3 - 5
Answers the leader takes         55-78 in 100       not measured yet
Gap from first to fifth            ~50 in 100       not measured yet
Anyone actively working on it              yes           almost never

The right-hand column is a piece of reasoning, not a measurement. The two empty cells can only be filled by real measurement, and they stay empty until we have one to publish with the data behind it. The left-hand column is that study’s published figures.

The trap inside the good news

Two of them, and they’re the reason this isn’t free money.

A small field is small for everyone. If a competitor in your city reads this and moves first, that same shallow gap closes on you instead. The window is open because nobody is looking, and that is a temporary condition by definition.

Local questions aren’t always answered locally. Ask about a service that people buy remotely and the model may ignore your city entirely and return national names. Whether your category is genuinely local is not something to decide at a desk. You go and watch what the models do, and it is the first thing to check, because if the answer is no, everything above stops applying and what if you opened after 2023 says takes over instead.

What to do with this

Find out how many companies you are actually up against. Ask the buying question in your city, across the three search-enabled models, several times. Collect every name that appears. Those names, not the list of every business in your town, are your real competition. There are almost always far fewer than people expect, and seeing them written down changes how the problem feels.

Find out what the ones holding the slots are standing on. Look at what the models cite when they name them. In local categories it is very often a small number of third-party pages: a directory, a regional roundup, a review site, one local news piece. Those you can work on, in a way you cannot work on a competitor with a decade of press coverage behind them.

Pick the constrained question, not the broad one. Best [category] in [city] is winnable. Best [category] is not, and the question your customer asked explains why the constrained version is also the one your customer actually asks.

Then measure the ordering, not just the presence. Being fifth in a five-name answer is not the same as being first, and the two are recorded as the same thing by most tools. Mentioned is not recommended has the split.