Back to Blog

AI Visibility · Industry Report · September 2026

We analyzed 1,140 ChatGPT and Gemini answers. Here’s how AI decides which healthcare practices to recommend.

Patients are increasingly asking AI where to go for care. So we tested 1,140 real patient-style queries across 19 healthcare markets. What we found surprised us: there are no fixed rankings. Recommendations can change within minutes. ChatGPT and Gemini disagree nearly half the time. And the sources influencing AI recommendations aren’t necessarily the ones practices spend the most time optimizing. Across those responses, we tracked 6,492 practice recommendations and 2,716 citations to understand what actually makes one practice appear while another gets overlooked.

Glace industry report cover: 1,140 ChatGPT and Gemini answers analyzed to see who gets recommended for care

The direct answer

Who actually gets recommended when a patient asks AI for care?

Usually, a different set of practices each time. When a patient asks ChatGPT or Gemini for, say, the best dentist nearby, the AI names five or six practices. If that same patient asked again a few minutes later, most of the list would change: typically only two names show up both times. Consistency is rare. In 16 of the 19 cities we studied, not even one practice appeared in half of the AI’s answers for its own area.

So the useful question is not “what’s my rank on ChatGPT.” It is: out of 100 patients asking AI right now, how many would hear my name? We call that a practice’s presence rate, and it is what this report measures. The practices heard most often are the ones whose websites and review pages AI can actually open and cite as proof. Sites that block AI from reading them, like Yelp, were cited zero times in this entire study.

FindingThe number
There is no stable ranking. Identical questions, minutes apart, return mostly different practicesonly 2 of ~6 names repeat
Nobody owns their city. Even the most-recommended practice in a city usually appears in fewer than half of the AI’s answers for that citytrue in 16 of the 19 cities we studied
ChatGPT and Gemini recommend different practices. The names both agree on are the handful every one already knows; the rest differ engine to engineonly ~1 in 5 names is shared by both
ChatGPT shows where its answers come from. Gemini rarely does. Answers that include at least one source link a patient can checkChatGPT 99% · Gemini 7%
AI’s proof comes from practice websites, not review platforms. Most source links point to the recommended practice’s own site; Yelp, which blocks AI from reading it, is never used78% practice websites · Yelp 0
The five findings at a glance, from 1,140 answers, 6,492 recommendations, and 2,716 citations. Each is unpacked below.

How we measured it

Every practice owner has run the experiment. Open ChatGPT, type “best dentist near me,” scan the answer for your own name. If it’s there, relief. If it’s not, panic. Either way, it feels like you learned where you stand.

Our data says you learned almost nothing, because the answer changes on the next ask.

Since early August, every AI visibility scan we run at Glace asks the questions patients actually ask about their healthcare needs. Things like “best dentist in Fremont,” “emergency dentist near me,” “which dentist takes PPO dental insurance.” Nothing branded; a practice’s name never appears in a question. Each question goes to both ChatGPT and Gemini with live web search on, and each is asked repeatedly at intervals, because a single ask proves nothing. Across 19 markets, that added up to 1,140 answers, each naming five to six practices on average.

Common patient queries Asked to both ChatGPT and Gemini Multiple rounds of every question 1,140 answers in total
This analysis covers 19 practice markets scanned between August 4 and September 3, 2026: 15 dental, 3 dermatology, 1 chiropractic, mostly suburban US metros. Every practice is anonymized in the figures below.

Across those 1,140 answers, four patterns kept repeating. None of them look like a ranking.

Ask twice, get a different list

Take the same question, the same city, the same assistant, minutes apart. We compared every pair of repeat asks in the dataset, 1,138 pairs in all, after normalizing practice names so formatting differences don’t count as churn.

Here is what a typical pair looks like. The first answer names five or six practices. The second answer also names five or six. Only two of those names appear in both. Roughly half of the first list vanishes on the very next ask, replaced by different practices. Measured formally: pool every name across both answers and, in the median pair, just 29% of them appear in both.

29%

is all that two identical asks share of their recommendations

Same question, same assistant, minutes apart

2.8%

of repeat asks returned the identical list

13%

of repeat asks shared no names at all

Same question, same assistant, same market, minutes apart. ChatGPT and Gemini churn at an identical rate: a 0.29 median overlap each.

This is the single number that should change how the industry talks about AI search. Google trained everyone to think in rankings: stable positions you climb one competitor at a time. AI recommendations don’t work that way. Each answer is a fresh draw from a weighted pool, and most of the pool rotates on every draw.

Nobody owns the shelf

If AI answers were a ranking, every market would have fixtures: practices that show up in nearly every answer, the way the same three practices sit in Google’s map pack for months. They almost don’t exist. In 16 of our 19 markets, not a single practice appeared in more than half of the AI’s answers for its own city.

The typical market surfaced about 120 different practice names across its answers. More than half of those names appeared exactly once and never again. The most-recommended name in a typical market captured just 7% of all recommendation slots, and the top three names together held under 18%.

16 of 19

cities have no practice that the AI recommends even half the time

53%

of recommended names appear exactly once, then never again

<18%

of a city’s recommendations go to its top three practices, combined

Compare Google’s map pack, where the top three positions are stable for months and absorb most of the clicks.

The AI shelf in most markets has no fixtures yet. It’s a rotating carousel, and nobody has claimed a permanent slot.

That cuts two ways. It means nobody can honestly sell you “#1 on ChatGPT.” It also means the window is open: the handful of practices in our sample that broke 80% presence prove stable visibility is achievable, and whatever earns it is being earned right now, before the shelf hardens the way Google’s did.

Even the visible flicker

For each practice we tracked how consistently it appeared when the same question was asked again. The pattern is not “visible practices stay visible.” Where a practice appeared at all, it appeared in every repeat of that question only about 60% of the time. The other 39% of the time it flickered: named in some asks, missing from others.

reliably visible: named in every ask of that question 27% flickered in and out 18% never named on that question 55%
380 question-and-assistant pairs, each asked repeatedly. Flicker rates are nearly identical on both assistants: 37% of ChatGPT’s any-mention pairs and 42% of Gemini’s.

ChatGPT and Gemini are different worlds

Put ChatGPT’s recommendations for a city next to Gemini’s and you are mostly looking at two different lists. Of all the practices named in a typical city, only about one in five was named by both assistants. The rest belong to one engine’s world or the other’s: of the practices reliably visible on at least one assistant, half were invisible on the other. Being recommended by ChatGPT tells you nothing about Gemini, and the reverse.

And when they do land on the same name, it is rarely a discovery: the practices both assistants agree on are the handful the whole city already knows.

This split is not a quirk of one kind of question. Take any question type, compare the names each assistant recommends for it in the same city, and there is no question where the two agree on even one name in five:

When the patient asks about…Names both assistants agree on
A specific procedure10%
Affordable11%
Best overall13%
Top rated for a treatment14%
Accepting new patients16%
Takes my insurance17%
Emergency, near me17%
Open on Saturday19%
A symptom needing treatment19%
For my kids or family20%

For a practice, the lesson is blunt. There is no being visible “on AI.” There is being visible on ChatGPT, and being visible on Gemini, and they are separate battles: winning one buys you nothing on the other. Why would two AIs reading the same city disagree this much? Because they read it through different evidence, and show different amounts of it. That is exactly where the next finding goes.

The sources AI cites, and the ones it can’t read

Both assistants ran with live web search enabled. What the patient sees could not be more different. ChatGPT showed at least one source link in 99.3% of its 570 answers, about five citations per answer, 2,621 links in total. Gemini showed a source in just 7%: 95 links across 570 answers. Gemini grounds its answers in search too; it just presents the result as settled fact, with addresses, phone numbers, and “highlights” and no visible evidence trail.

So where do ChatGPT’s citations point? Not where the industry spends its energy. 78% of the citation links go to practice-level websites: the recommended practice’s own site, its offers page, its hosted review page. Only 22% go to platforms and directories of any kind. And the platform list is upside down relative to a decade of local-marketing advice:

SourceWhat it isAnswers citing itMarkets
reviews.birdeye.comHosted review pages81 of 57016 of 19
zocdoc.comBooking marketplace66 of 57013 of 19
deltadental.comInsurer directory45 of 57011 of 19
aae.orgEndodontist society directory24 of 57014 of 19
findadentist.ada.orgADA directory22 of 57012 of 19
invisalign.comVendor provider locator17 of 57010 of 19
healthgrades.comReviews directory17 of 57011 of 19
yelp.com · facebook.com · instagram.comWhere the reviews live00 of 19

Read that last row again. In 570 recommendation answers with 2,621 source links, ChatGPT cited Yelp exactly zero times. Facebook: zero. Instagram: zero. Google reviews: effectively zero. This is not a mystery. Yelp’s robots.txt blocks AI crawlers from its business pages,1 so those reviews cannot serve as evidence in an AI answer. Meanwhile Birdeye’s hosted review pages, which are open to crawlers, have quietly become ChatGPT’s favorite review source, cited in 16 of our 19 markets, ahead of Zocdoc and Healthgrades.

Whoever hosts your reviews on an AI-readable page becomes your review site of record. The platforms blocking AI crawlers have written themselves out of the answer.

A concierge and a brochure

The two assistants have also settled into different characters, which matters for what a patient does next:

In a typical answer…ChatGPTGemini
Length220 words391 words
Includes a phone number96%63%
Shows its sources99%7%
Formatted with headers and bullets48%100%
Ends by offering to narrow the choice8%1%

ChatGPT behaves like a concierge: shorter answers, nearly always a phone number, and sometimes an offer like “tell me your insurance and I’ll narrow it down.” Gemini behaves like a brochure: longer, heavily structured, more names, less action. Either way, the patient is handed a short list and, most of the time, a number to call. What happens when they call it is a different problem, and the reason our AI Front Desk exists.

One more pattern worth knowing: across all 1,140 answers, the assistants named 2,373 distinct providers in 6,492 recommendation mentions, and 68% of those names are practice brands, not individual doctors. AI recommends institutions more readily than people, which quietly rewards practices with a strong brand name over practices trading on a practitioner’s personal reputation.

What this means for your practice

1. Screenshots are not evidence. Any agency claiming they got you “to #1 on ChatGPT” off a single screenshot is selling a lottery ticket stub. With a 29% median overlap between identical asks, one answer proves nothing in either direction.

2. Ask about your draw rate, not your rank. The honest metric is a presence rate: out of many repeated, varied asks, how many name you? That is a number you can track month over month and actually move.

3. Your website is the evidence AI presents. With 78% of citations pointing at practice-level pages, your site, your offers page, and your AI-readable review page are not just conversion assets. They are the receipts ChatGPT hands the patient. A practice invisible to AI crawlers has nothing to be cited with, and the platforms that block those crawlers can no longer speak for you there.

4. The window is open. No fixtures in 16 of 19 markets means AI assistants haven’t crowned local winners yet. The practices that build citable evidence now are buying tickets in every draw while their competitors are still asking what their rank is.

Where Glace fits

This research comes directly out of how our visibility scan works. We never trusted a single ask, so every Glace scan puts the common patient questions to both assistants and repeats them at intervals, the same way this research was run. The score a practice gets is a presence rate, because that is the only number that means anything in a lottery. The scan also reads the answers themselves: which competitors keep appearing, which sources get cited in your market, and which question types you are absent from.

Start here, free

AI Visibility Check

See your practice’s presence rate across ChatGPT and Gemini, measured the way this research was: repeated real answers, not one screenshot

Repeated asksBoth assistantsFree

Then build the evidence

Marketing & AI Visibility

An AI-readable website, service-specific content, and reviews where AI can actually cite them, so you appear in more of the draws

ContentSEOAI answers

When the patient calls

AI Front Desk

96% of ChatGPT answers hand the patient a phone number. The AI Front Desk makes sure that call becomes an appointment

CallsBooking
Visibility gets you into the answer. The front desk turns the answer into an appointment.

The honest caveats: 19 markets is a small sample, weighted toward dentistry and the metros we work in. Run-to-run variation partly reflects randomness inherent to these models, which is precisely the point, because the patient’s answer is drawn from that same randomness. And these numbers describe August and September 2026. The models will drift, which is one more reason a one-time check is worth less than a measurement you repeat.

There is no AI ranking to win. There is a lottery, and the only strategy is to hold more tickets: more citable evidence, in more of the places AI actually reads, measured often enough to know it’s working.

Frequently asked questions

Is there an AI ranking for local businesses?

Not in the way Google has rankings. In our data, a typical answer names five or six practices, and asking the identical question again minutes later typically brings back only two of those names. AI recommendations behave like repeated draws from a weighted pool, so the meaningful measurement is a presence rate across many asks, not a position in a single answer.

Why does ChatGPT recommend different practices every time you ask?

Each answer is generated fresh. The assistant runs a new web search, samples from what it finds, and composes a new list. Practices with strong, consistent evidence across the web appear in a larger share of the draws, but very few appear in every draw. In our sample, only 2.8% of repeat asks returned the identical list.

How do I check if ChatGPT recommends my practice?

Not with a single ask. One answer is a coin flip: in our data, practices that appear at all flicker in and out of answers 39% of the time. A meaningful check asks many patient-style questions across multiple assistants and multiple runs, then reports the share of answers that mention you. That is how the free Glace visibility check works.

Does ChatGPT use Yelp reviews when recommending healthcare practices?

In our 570 ChatGPT answers with 2,621 source links, Yelp was cited zero times. Yelp’s robots.txt blocks AI crawlers from its business pages, so its reviews cannot serve as evidence in AI answers. Review pages that AI can read, such as hosted review pages connected to practice websites, were cited instead.

What sources does ChatGPT cite when it recommends a dentist?

Mostly the practices’ own websites: 78% of citation links in our data pointed at practice-level pages. Among platforms, the most cited were Birdeye-hosted review pages, Zocdoc, insurer directories like Delta Dental’s, professional directories like the ADA’s Find-a-Dentist and the AAE’s endodontist finder, and vendor locators like Invisalign’s.

Do ChatGPT and Gemini recommend the same practices?

Mostly not. In a typical city we studied, only about one in five recommended practices was named by both assistants, and the shared names were the market’s best-known practices. Of practices reliably visible on at least one assistant, half were invisible on the other, so checking one assistant tells you little about the other.

What is AI visibility for a healthcare practice?

AI visibility is how often AI assistants like ChatGPT and Gemini name your practice when patients ask for care recommendations in your area. Because answers vary between asks, it is best expressed as a presence rate, the share of relevant AI answers that mention you, rather than a rank.

How can a practice improve its AI visibility?

Make your evidence easy for AI to find and cite: an AI-readable website with clear service pages, reviews on pages open to AI crawlers, consistent presence in the directories AI actually cites, and third-party mentions connecting you to specific services and locations. Then measure repeatedly, because a single check tells you almost nothing. Our earlier article on what feeds AI answers covers the evidence side in detail.

Written by Glace

Glace helps independent healthcare practices become the credible answer patients find online, with a connected path from patient education to booked appointment.

Meet the team

Methodology and references

Findings are from 1,140 AI answers collected by Glace visibility scans between August 4 and September 3, 2026, across 19 practice markets (15 dental, 3 dermatology, 1 chiropractic). Ten frozen, unbranded question templates per specialty were asked to ChatGPT and Gemini via their APIs with web search enabled, three independent runs each. Practice names were normalized before overlap and concentration calculations; test scans and duplicate scans were excluded, and practices are anonymized in all figures. Citation counts measure source links visible in the answer text a patient reads. Gemini’s search grounding runs server-side, so its 7% figure measures the sources it chooses to display. These numbers describe model behavior in August and September 2026 and will drift as models update.

  1. Yelp. robots.txt, which disallows GPTBot, ClaudeBot, Google-Extended and other AI crawlers from business pages. yelp.com/robots.txt
  2. OpenAI. Introducing ChatGPT Search, on how ChatGPT answers are grounded in live web results with linked sources. openai.com/index/introducing-chatgpt-search
  3. Google. Grounding with Google Search, on how Gemini grounds answers in search results. ai.google.dev/gemini-api/docs/grounding
Share Post on X Share on LinkedIn