Clinician-reviewed formula
Every calculator runs on an NIH/WHO/ACSM-sourced formula; the reviewer name + sameAs is visible on the page.
Health & Wellness · Case Study
HealthCalcPro — YMYL Health Calculator Ecosystem
A wellness platform serving BMI, TDEE, body-fat, heart-rate, pregnancy, and sleep calculators with clinician-reviewed formulas — meeting Google's YMYL standards.
Results
Measurable Outcomes
10/10
Clinician-approved formulas
100/100
Lighthouse SEO score
Tüm hesaplayıcılar / All calculators
AI Overview citation readiness
< 1.8s
Mobile LCP (p75)
The Brief
What was the goal?
Health content sits in the YMYL category; Google penalizes generic calculators without clinician review, doesn't cite them in AI Overviews, and starves them of Discover traffic. Clinical-trust E-E-A-T signals were non-negotiable.
The Approach
Which methodology pillars I applied
I applied the Schema (MedicalEntity + Person reviewer chain), Entity (clinician sameAs stacks), and AEO (result block speakable, FAQ schema) pillars. Every formula links to an NIH/WHO/ACSM source — and the source is visible.
See the full process → 47-Point AI-Ready SEO Audit methodology.
Tech Stack
What it's built with
- Next.js 15
- TypeScript
- Tailwind CSS
- Edge Functions
- Schema.org MedicalEntity
The Build
Key decisions and tradeoffs
Clinician-approved formula document: per calculator, the source formula (NIH, WHO, ACSM), unit system, and edge values are captured in a single reference doc. That file is both developer documentation and E-E-A-T evidence for Google.
I wired MedicalCalculator → MedicalEntity → Person (author + reviewer) → Organization @id chain. The reviewer is a real clinician with a Wikidata Q-ID — "reviewed by" isn't a fake signal.
Health Age Quiz and personalized reports run on Edge-function-based state management. Every result is tagged with the speakable selector; ChatGPT can pull a citation directly from our calculator when answering a user's BMI query.
Steps Applied
Methodology in practice
MedicalEntity schema chain
MedicalCalculator → MedicalEntity → Person → Organization @id chain; complete YMYL E-E-A-T signal.
Speakable result block
Every calculation result lives at #result-block id + SpeakableSpecification; AI Overview citation-ready.
Honest Reflection
What I'd Do Differently
In v1 the reviewer name was on the page but the Person schema didn't carry a Wikidata Q-ID. Google's YMYL filters read it as an "unverifiable reviewer" and started suppressing us in featured snippets. In v2 we created a Wikidata entity for the reviewer and linked the sameAs chain. On the next YMYL project, reviewer entity linking is non-negotiable before launch.
Screenshots
Want this for your project?
Same 47-point methodology, $499 fixed price, 5-7 business days. Book the audit.