AI Visibility for Local Medical Practices · Finding 04 of 9 · August 2026 edition

AI recommends a shortlist of providers, typically three to four.

This study analyzed 3,234 AI answers and 10,133 citations to 72 patient questions on ChatGPT, Gemini, Perplexity and Google AI Overview, checked against 18,492 practice listings in five US metro markets: suburban Chicago, Philadelphia-South Jersey, Tampa-St. Petersburg, Phoenix West Valley and suburban Houston.

Get your free AI Visibility Snapshot

Summary

On local medical provider discovery questions in Javvo's study, ChatGPT named about 3 providers per answer and Perplexity about 4, while Gemini averaged about 7 and read more like a directory page than a recommendation. Only 14% of provider discovery answers elevated a single practice above the rest.

AI answers a provider discovery question with a list, not a single name

On provider discovery questions, ChatGPT named about 3 providers per answer and Perplexity about 4. Google AI Overview averages about 3 names whenever it names anyone, measured across all question types, since it appears too rarely on provider discovery searches to measure there. Gemini is the outlier. It averages about 7 providers per answer and reads more like a directory page than a recommendation.

Providers named per provider-discovery answer

2.9 3.0 4.0 6.6 ChatGPT Google AI Overview* Perplexity Gemini
Mean providers named, by engine
Engine Mean providers named
ChatGPT 2.92
Google AI Overview* 3.02
Perplexity 4.03
Gemini 6.56
Average providers named per provider discovery answer.

*Google AI Overview measured across all question types.

The length of the provider list varies depending on the question

Condition questions generally drew the longest lists, at about 5 providers named per answer. Urgency and insurance questions drew tighter lists of about 3, and direct questions about a specific practice drew the shortest lists, at about 2.

Crowning a single winner in AI answers is rare

Only 14% of provider discovery answers elevated one practice above the rest, and another 11% narrowed the field to two. The rest presented options and left the choice with the patient. When an engine does commit to a single recommendation, it tends to commit to a person. Every single-winner pick that named an individual doctor rather than a practice occurred in podiatry, where doctors are named most.

See how AI answers questions about your practice today.

Get your free AI Visibility Snapshot

Run AI visibility for the practices you serve.

AEO for partners

August 2026 edition.