Skip to content
leve.devFull-Stack·Case Study 2 / 3
Full-Stack·2026·Live

KurklinikFinder

Finding the right mother/father-and-child rehab clinic in Germany meant wading through endless undifferentiated lists. KurklinikFinder turns all 116 of them into a clean, filterable comparison — built SEO-first, down to an on-demand data feed for LLM crawlers.

Next.js 16SupabaseTypeScriptTailwind v4Visit site
116
clinics
10+
regions
119
static pages
SEO
built-in first
KurklinikFinder landing: 'Finde deine Kurklinik' with filters for clinic type, indication and region
Filter by clinic type, indication and region instead of scrolling endless lists.
01 — The Problem

No site let parents actually compare.

There was no place that distinguished cure clinics from rehab clinics and let mothers and fathers filter simply and practically. So I migrated the data — a WordPress CSV export mapped into a clean database with regional tagging — and built statically generated detail pages for every one of the 116 clinics.

02 — SEO as a First Principle

Built to be found — by people and by machines.

This is a discovery product, so search visibility isn't a finishing touch — it's the architecture:

  • A dynamic sitemap covers every clinic slug and blog post; each clinic ships MedicalClinic + LocalBusiness JSON-LD and its own meta title, description and OG image.
  • An llms-full.txt route generates an on-demand markdown dump of the entire clinic database, so LLM crawlers can ingest all 116 clinics in one read.
  • A map view (MapLibre), clinic comparison, reviews with a custom score, and an MDX guide section round it out.
  • Built on Next.js 16 with the React Compiler and the new caching directives — with an AI-native dev setup (custom agents and skills) living right in the repo.

Planning something similar?

I design and ship AI agent systems, data platforms and full-stack products — from first idea to production.

Get in touch