In Citevio's engine panel measurement on 11 August 2026, Perplexity named a practice or agency 4 times and ChatGPT did so 2 times across 16 buyer-intent agency-selection questions, not patient-facing dental questions. Both gave no name for most of the set. Google AI Overviews results are not published here, and Gemini was not measured because of quota errors.
AI answers should be checked engine by engine and question by question. This page reports a narrow agency-selection measurement, not a measurement of patient-facing dental recommendations.
What did we ask, and what did each engine do?
In Citevio's engine panel measurement on 11 August 2026, we ran 16 buyer-intent agency-selection questions, not patient-facing dental questions, against four engines in one measurement window. For each answer we recorded whether a business or agency was named directly, whether it appeared only as a cited source, or whether the engine gave no relevant name.
| Engine | Named by name | Cited as source only | No relevant name |
|---|---|---|---|
| ChatGPT | 2 | 0 | 14 |
| Perplexity | 4 | 1 | 11 |
| Google AI Overviews | Not published; see below | ||
| Gemini | Not measured: quota error on all 16 questions | ||
In Citevio's engine panel measurement on 11 August 2026, Perplexity produced more named mentions than ChatGPT on the same 16 buyer-intent agency-selection questions (4 versus 2). Neither engine answered with a name on most questions. That does not establish a trend.
We also ran the same questions against Google AI Overviews, but Citevio does not publish AI Overviews measurement results yet; that column is held back until the measurement is repeated.
This is one measurement window, not a trend. The completed run-consistency measurement explains why repeated checks matter. Until a pattern repeats across two runs, treat this table as a snapshot, not a rule.
Is Perplexity easier to win than ChatGPT?
No conclusion follows from this measurement. Perplexity named a business more often than ChatGPT did in one measurement window. That is not the same as being easier to win, and Citevio has not tested which engine is easier.
The engines are not interchangeable in this measurement. The useful unit is engine by engine and question by question, with the scope stated clearly: buyer-intent agency-selection questions, not patient-facing dental questions.
Why do two engines disagree about the same practice?
Citevio's panel measures the outcome, whether a name appears, not the mechanism behind it. We did not test why ChatGPT and Perplexity diverge on identical questions, so we are not claiming a cause here.
No major AI crawler runs JavaScript. GPTBot, OAI-SearchBot, ClaudeBot and PerplexityBot read the raw HTML only, so a page whose content is drawn by JavaScript can rank on Google and still be blank to ChatGPT. Google Gemini is the exception, because it uses Google's rendering service. What that raw-HTML limit means in practice for a dental site is set out in why JavaScript-built dental websites can be invisible to AI.
Google documents that AI Overviews use Google Search eligibility requirements, while OpenAI and Perplexity document search-backed answers with cited sources. These documented mechanisms explain why source retrieval matters, but they do not establish the cause of the differences in this measurement.
What can this measurement not tell you?
Citevio's engine panel measurement on 11 August 2026 was run on buyer-intent agency-selection questions, not patient-facing dental questions. The two question types are not interchangeable, and Citevio has not measured whether the same pattern holds for patient queries. The 4-versus-2 gap could change in another measurement window; this page makes no forecast or timeline.
A separate site-level evidence base should not be folded into this panel: Citevio read 527 US dental practice websites field by field across seven metros plus nine nearby towns between 6 and 26 June 2026.
Why this measurement says nothing about Gemini — and what Google does publish
The Gemini column of this measurement is empty, and an empty column is a gap rather than a result. It is worth stating plainly because the obvious wrong reading — "Gemini did not name anyone" — is not what happened. Nothing was learned about Gemini here either way. What can be added instead is what Google itself publishes about appearing in its AI features, which turns out to be the opposite of what most of the market sells.
Google's Search Central documentation on AI features is unusually blunt for a platform document. On whether there is a technical entry ticket:
"You don't need to create new machine readable files, AI text files, or markup to appear in these features. […] There are no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary."
Source: Google Search Central, AI features and your website, read 18 August 2026.
The same page describes how these features assemble an answer, noting that both AI Overviews and AI Mode "may use a 'query fan-out' technique — issuing multiple related searches across subtopics and data sources — to develop a response". That matters more for a dental practice than a checklist would: an answer built from several searches at once has several chances to reach a page, and several chances to miss one.
What that document does not contain is a route for local businesses. Business Profile appears once, inside a general list of good practice, not as a described path into an AI answer. So the honest position on "how do I get into Gemini" has two halves and both are negatives: Google does not publish a separate local mechanism, and Citevio has not measured one.
First, AI Overviews and AI Mode sit inside Google Search; the Gemini app is a different surface. The documentation quoted above is written about Search, and reading it as a statement about the Gemini app would be putting words in Google's mouth. Second, the fact that this measurement excluded Gemini says nothing about whether Gemini names dental practices. It says the column returned a quota error on all 16 questions and was therefore dropped, which is a limitation of our run rather than a finding about the engine.
Whether the Gemini app and AI Overviews are the same thing is a question with its own answer elsewhere, in the FAQ of how to appear in Google AI Overviews for dental. How a column gets dropped rather than counted as a loss, and why that rule matters to every number on this site, is on how we measure AI visibility.
Which questions are commonly asked about engine agreement?
Do ChatGPT and Perplexity give the same agency-selection answer?
Not reliably. In Citevio's engine panel measurement on 11 August 2026, Perplexity named a practice or agency 4 times and ChatGPT did so 2 times across one window of 16 buyer-intent agency-selection questions, not patient-facing dental questions. Both engines gave no name for most questions.
Is Perplexity easier to get recommended by than ChatGPT?
In Citevio's engine panel measurement on 11 August 2026, Perplexity named a business more often than ChatGPT did in one window (4 versus 2 of 16 buyer-intent agency-selection questions, not patient-facing dental questions), but that is not the same as being easier to win. Citevio has not tested which engine is easier.
Why do AI engines give different answers to the same question?
Citevio's panel measures the outcome, not the cause. Different engines can use different retrieval and ranking processes, but this measurement was not designed to isolate why their answers differed.
Did this measurement include Google AI Overviews and Gemini?
Both were part of Citevio's engine panel measurement on 11 August 2026, but neither produced a publishable result. Citevio does not publish Google AI Overviews measurement results yet, and the Gemini column returned a quota error on all 16 buyer-intent agency-selection questions, so that column is a measurement gap, not a result.
How many of the 16 questions did ChatGPT name a business or agency for?
Two of 16. In Citevio's engine panel measurement on 11 August 2026, ChatGPT named a practice or agency by name for 2 of the 16 buyer-intent agency-selection questions, cited a domain as a source only for 0, and returned no relevant name for the other 14.
How many of the 16 questions did Perplexity name a business or agency for?
Four of 16, plus one source-only citation. In the same 11 August 2026 measurement, Perplexity named a practice or agency for 4 of the 16 questions, cited a domain as a source only for 1 more, and returned no relevant name for 11.
What's the difference between an engine 'naming' a business and 'citing' it as a source?
Named means the practice or agency appears by name inside the answer text itself. Cited means the domain only shows up in the source list, without the business being named in the answer text. Citevio's 11 August 2026 panel tracked these as two separate outcomes, not one visibility number.
Do the three outcome columns in the engine panel table add up to all 16 questions for each engine?
Yes. For ChatGPT: 2 named plus 0 cited-only plus 14 with no relevant name equals 16. For Perplexity: 4 named plus 1 cited-only plus 11 with no relevant name equals 16. Every one of the 16 buyer-intent agency-selection questions from the 11 August 2026 measurement falls into exactly one of the three columns, per engine.
Why does this page treat buyer-intent agency-selection questions separately from patient-facing dental questions?
Because the two question types are not interchangeable, and this measurement tested only the first kind: which agency or vendor to hire, not which dentist to see. Citevio has not measured whether the 4-versus-2 pattern between Perplexity and ChatGPT holds for patient-facing dental questions.
Could the 4-versus-2 gap between Perplexity and ChatGPT look different in another measurement window?
Yes. This is one measurement window from 11 August 2026, not a trend, and this page makes no forecast about future windows. Citevio's separate run-consistency work explains why a single window shouldn't be read as a settled pattern; until a result repeats across more than one run, treat a single-window gap like this one as a snapshot.
What do OpenAI, Perplexity and Google themselves document about how their AI answers get assembled?
Google documents that AI Overviews use Google Search's own eligibility requirements. OpenAI and Perplexity document that their answers are search-backed with cited sources. Those are the mechanisms each company has published; Citevio's 11 August 2026 measurement did not test why ChatGPT's and Perplexity's specific answers diverged on identical questions, so it isn't the source of that explanation.
Does an empty Gemini column mean Gemini never named an agency in this test?
No — that reading would be wrong. Gemini returned a quota error on all 16 questions in the 11 August 2026 measurement, so the column was dropped rather than counted as a result. Nothing was learned about Gemini either way; that's a limitation of this particular run, not a finding about the engine.
Why doesn't this page publish a Google AI Overviews result alongside ChatGPT and Perplexity's?
Google AI Overviews was part of the same 11 August 2026 measurement, but Citevio does not publish AI Overviews measurement results yet. That column is held back until the measurement is repeated, rather than published as a one-off reading.
Does this 11 August measurement use the same 'at least three runs per engine' methodology Citevio uses for client tracking?
No, and that's a stated limit on what this page can claim. Citevio's client-facing methodology runs every question through all four engines across at least three separate runs, and a finding without its ratio doesn't go into a client report. This page's engine panel is a single measurement window with results published for two engines only, so it should be read as one window, not a tracked pattern.
Is Citevio's engine panel measurement openly licensed, and does it include client data?
Yes to the first, no to the second. Citevio's dental market research, including this engine panel measurement, is published openly under CC-BY-4.0 with a DOI, so the underlying numbers can be checked and reused. Client data is never published in any of it.
Citevio's dental market research is published openly under CC-BY-4.0 with a DOI, so the numbers can be checked and reused; client data is never published.
Sources: OpenAI, ChatGPT search; Perplexity, What is Perplexity?; Google Search Central, AI features and your website; Citevio open dataset, Zenodo DOI 10.5281/zenodo.23143214; Citevio engine panel measurement, 11 August 2026 (own data, methodology: how we measure AI visibility).
See how your own practice comes up
A free scan shows whether ChatGPT and Perplexity mention your practice by name today.
Run my free AI scanDelivered within 48 hours · no sales call — ever