Patients no longer arrive via Google: how ChatGPT is choosing dentists in Mexico

In short: When a US visitor asks ChatGPT for a dentist in Mexico, they no longer get ten blue links: they get a short, ranked list of clinics, with stars and a suggested specialty. And that patient behaves differently: traffic arriving from ChatGPT converts at 15.9%, versus 1.76% for Google organic: nine times higher [1]. For your clinic, the question is no longer how you rank on Google; it's whether the AI puts you on that short list.

The test: a dentist in Los Cabos, asked in English

We did something simple. We asked ChatGPT, in English, for a dentist recommendation in Los Cabos — the exact question a visitor from California or Texas would type while planning a trip. The reply wasn't a results page. It was an editorial pick: five clinics, each with its stars, its location, a suggested specialty ("best overall", "best for implants"), and a map. A digital concierge that had already decided for the patient.

For the five clinics that showed up, that's a patient channel that costs them nothing. For the rest — dozens of good practices in the same city — it's a waiting room that empties without the owner knowing why. Nobody canceled. Nobody complained. They simply stopped calling.

Nine times: why an AI patient isn't just any patient

Seer Interactive — a US marketing-data consultancy — measured how traffic converts by source. Traffic from ChatGPT books or buys 15.9% of the time; traffic from Google organic, just 1.76% [1]. Nine times higher. The reason is common sense: whoever arrives from the AI has already been pre-selected. They already read why that clinic and not another. They come to book, not to compare. Google organic still sends you browsers; the AI sends you decided people.

Where the AI gets its names

Mexican dental tourism — Los Algodones, Tijuana, Cancún, Los Cabos — lives on foreign patients who research from afar, months before boarding a plane. That research changed shape. It used to run on searches and open tabs; today you describe your case to the AI and get two or three names. 45% of consumers already use AI assistants to find local services, up from 6% a year earlier [2], and the medical visitor is among the profiles that research most before traveling.

Here's the part almost no one sees: where the AI gets those names. The mechanism is the one that repeats across AI search: models lean on sources that already aggregate and structure businesses: the directories and aggregators. Platforms that gather many clinics with structured data, reviews, and comparisons. The AI doesn't visit each dentist's website one by one; it reads whoever already did the work of ranking them.

And the result concentrates in a few hands. In our July 2026 monitoring of English-language 'dentist in Los Cabos' queries — the typical US visitor's question — a handful of clinics concentrate most of the mentions: the same five or six names repeat answer after answer, while the vast majority of the area's practices appear in none.

It's worth separating the phenomenon from its solution. The phenomenon is that the AI is already a discovery channel, and it hands out patients in a concentrated way. The solution isn't an "AI ranking" trick; it's fundamentals work — boring and verifiable — that the clinics who show up already have and almost no one else has done.

Four fundamentals the AI actually reads

These aren't tricks. They're the signals a model uses to decide whom to name. In order of effort.

1. Fresh, answered reviews. A clinic in Tijuana can have twenty years of history and happy patients, but if its latest reviews are eighteen months old, to the AI it's frozen. Models read recency: reviews from the last few weeks, with a reply from the practice, weigh more than a high, old average. Answering every review — the good and the awkward — signals a business that's alive and attentive. Don't ask for five stars; ask the patient to describe what you did. That clinical detail is exactly what the AI cites.

2. Published prices or ranges. Almost every dental-tourism question includes "how much does it cost". A patient from San Diego looking for implants in Los Algodones wants the number before driving three hours. Most clinics hide it behind "contact us"; the AI can't cite what it can't find, so it recommends the one that did publish its range. You don't have to set a price to the peso: an honest range — "zirconia crown, from X to Y dollars" — makes you the useful answer the model prefers to give.

3. The same information everywhere. An orthodontist in Cancún might show one phone number on Google, another on Facebook, and a third in the directory where an aggregator listed her years ago. To a human it's an oversight; to the AI it's a low-confidence signal, and sometimes a reason not to name her. Name, address, phone, hours, and services have to say exactly the same thing on every platform. Consistency isn't cosmetic: it's what lets the model trust that the data is correct.

4. Presence in the directories machines read. The AI doesn't improvise: it leans on sources that aggregate and structure businesses. A clinic in Puerto Vallarta can have the best website in town and still be invisible, because the model rarely visits a standalone site — it goes to the registries and directories that already ranked the sector. Being in the right sources, with machine-readable data, is what puts you in the deck the AI deals from. Outside those sources, you don't exist to it.

Measure first, optimize later

If you run a clinic in any of these cities, the opening scene is already happening to you, whether you see it or not. Someone in the US asked ChatGPT this week for a dentist in your area, and the AI named five. The only question that matters is whether your name was among them.

And that question has an exact answer, not an opinion. Before hiring anyone who promises to "get you into the AI", measure your baseline: in what fraction of answers about your specialty and city do you appear today, and who is keeping those patients. That number is the starting point for everything else. Plexa's visibility report gives it to you: it measures what the AI says about your clinic versus your competitors, across real dental-tourism queries. Claim your listing and start by measuring. Optimizing without that number is shooting blind — exactly the mistake the clinics that already show up did not make.