Specialist surgery
An orthopaedic surgeon, Sunshine Coast
A shoulder surgeon in a crowded regional market. We rebuilt his website as 103 pages a machine can read, and ChatGPT, Claude and Gemini now all recommend him, two of them first.
The numbers
- 3 of 3
- Major AI assistants that recommend him for shoulder surgery on the Sunshine Coast, ChatGPT, Claude and Gemini Clean API sessions with web search on and no account history, three phrasings per engine, 30 September 2026
- 1st
- Named first by Claude and by Gemini, in every answer where they named a surgeon Claude named him first in 3 of 3 phrasings. Gemini named him first in 2 of 2, and declined to name any surgeon on the third
- 2 of 3
- ChatGPT answers that recommend him, each linking straight to the website we rebuilt Clean API session with web search on, 30 September 2026
- 103 of 103
- Pages on the rebuilt website carrying structured data a machine can read Every URL in the live sitemap fetched and checked, 24 September 2026
What we did
- Rebuilt his website end to end, 103 pages, with structured data on every one
- Published the surgeon as his own entity, separate from the group practice
- Defined the subspecialty precisely, shoulder and upper limb rather than orthopaedics generally
- Published procedures, fellowships, hospital appointments and locations as structured facts
- Verified every publication against Crossref after finding three citations that did not resolve
Specialists have a structural problem that clinics do not. They appear inside a group practice page listing eleven surgeons and every joint in the body, so an engine asked about shoulders in a specific region has nothing to work with except a name on a team page.
The fix started with the website. We rebuilt it as 103 pages, every one carrying structured data, so there was finally something specific for an engine to read. Then entity work: publishing the surgeon as an identifiable person with a named subspecialty, specific procedures, fellowships, hospital appointments and locations, so a model can answer a narrow question with a specific name.
On 30 September 2026 we asked ChatGPT, Claude and Gemini three different versions of the same patient question, through API sessions carrying no account history, and never mentioning him in the prompt. Claude named him first in all three answers. Gemini named him first in both answers where it named a surgeon, and on the third phrasing declined to recommend anyone. ChatGPT recommended him in two of the three, linking directly to the website we rebuilt, with one or two other surgeons listed above him.
We do not report this as a ranking of surgeons, because it is not one. It is a record of what three engines returned, on one day, for three phrasings.
More results
Other case studies
General practice
A single-site GP and skin cancer clinic
10 → 109 AI-attributed bookings per month, May to September 2026
- $39.59 → $10.54
- Google Ads cost per new patient phone call, June to September 2026, on an unchanged daily budget cap
- 120
- Bookings attributed to ChatGPT specifically, May to 23 September 2026
General practice
A 22-location GP group, Sydney
1 in 3 Tracked AI answers that cite the group by name
- 50%
- Of Claude answers name the group, its strongest engine
- 22
- Locations published as distinct machine-readable entities
General practice
A Newcastle general practice, established 1938
51% Of tracked AI answers name the practice
- 35%
- Of tracked answers cite the practice's own site
- 53%
- Citation rate on Gemini and Perplexity, its strongest engines
Find out what AI says about your practice
We run your practice against every major AI engine and show you exactly where you stand, and who is being named instead of you. No charge, no lock-in.