New Zealand patients choose between the public system run by Te Whatu Ora and private care, and the choice usually comes down to insurance and waiting time. This directory was built for the private side of that decision: clinics, GPs and specialists in Auckland, Wellington and Christchurch.
The Challenge
- Health is a regulated, high-trust topic, and Google filters out thin or anonymous directories.
- Patients do not only want a clinic. They want to know whether their insurer covers it and how a referral works.
- The public-versus-private question appears in AI answers as often as in standard search.
The Strategy
1. Query trees built on NZ health logistics
Head terms were private hospitals Auckland, GP clinics Wellington and specialist doctors Christchurch. Pages were mapped to answer the whole chain:
- Cost and insurance: "Private health insurance consultation cost orthopedic surgeon Auckland"
- Comparison: "Southern Cross vs Nib approved cardiology clinics Wellington"
- Process: "Te Whatu Ora referral pathways vs private medical appointments New Zealand"
Insurer panel names and Medical Council of New Zealand standards run through the copy, so the pages match the way New Zealanders actually phrase these searches.
2. A trust filter on the data
- Listings, accreditations and service data were collected with Python scripts.
- Facilities rated below 4 stars, or without verifiable Medical Council registration, were removed and 301-redirected to stronger category pages.
- Titles rotated across five sentence patterns naming real areas, including Auckland CBD, Takapuna and Wellington Central.
- Missing data shows a dash.
3. One accountable author
A single author card links every page to a named expert, so the directory has an owner a reader or an algorithm can check.
The Result
Search Console, 8 to 28 October 2026:
| Metric | Value |
|---|---|
| Total clicks | 23 |
| Total impressions | 7,350 (7,024 from New Zealand) |
| Average CTR | 0.3% |
| Average position | 15.4 overall, 14.6 in New Zealand |
An average position of 14.6 puts the typical query on page two, a few places from the clicks that sit on page one. Nearly all impressions come from New Zealand, which confirms the geo targeting is working.
Why it worked
Rotating five title formats stopped pages for neighbouring suburbs from competing with each other while still capturing long-tail searches across regions. Filtering out weak providers kept the pages credible, and the clean schema gives AI systems something safe to parse and cite.
How to apply this
- Map the site to recognised entities and link providers to the official medical bodies.
- Let one template answer the treatment question and the logistics question (insurance, waiting time, location).
- Automate data audits that remove low-rated entities and keep redirects clean.
- Write definitions and comparison tables that generative engines can display directly.