Case study · Real estate

Click Real Estate

A digital estate agency in Seville where buyers search by neighbourhood and sellers follow their sale from their phone.

Industry: Real estate · Seville
Our role: Design, development, hosting and maintenance
Year: 2026
Website: clickrealestate.es ↗

The problem

Big property portals treat every listing the same and leave sellers in the dark: they can’t see how many viewings their home gets or which offers come in. Click wanted the opposite — a small, curated set of homes, useful data for every neighbourhood, and sellers who see everything in real time.

What we did

Design

Product identity and component system
Search, property pages and area pages
Designed mobile-first

Logic

Listings with filters and a map by neighbourhood and town
Buyer alerts
Home valuations backed by statistical models
Private seller dashboard: viewings and offers in real time
PDF reports

Infrastructure

Next.js 15 and React 19 with TypeScript and Tailwind 4
FastAPI with async SQLAlchemy, PostgreSQL 16 and Alembic
Redis, transactional email via Resend and S3-compatible storage
Docker on Hetzner, error tracking with Sentry

Technical decisions

Leaflet maps on GeoJSON: Neighbourhoods and towns are drawn with their real boundaries, so search and area pages share the same geography.
NextAuth with Google sign-in and two-factor: The seller area holds sensitive data, so access goes through Google plus a second factor.
An async Python backend: FastAPI with async SQLAlchemy on PostgreSQL 16 and versioned Alembic migrations — the same language as the statistical valuation models.
Email, files and reports: Transactional email via Resend, files in S3-compatible storage and PDF reports generated on the server.
Docker on Hetzner, watched by Sentry: The whole system runs in containers on a Hetzner server, with error alerts in Sentry and 30+ scheduled jobs.

Screenshots

Homes for sale with filters — **Search.** Filter by area, type, bedrooms and price, with an alert for new listings.

Property page with photo gallery — **Property page.** Gallery, price per m² and a virtual furnishing proposal.

Two-step home valuation form — **Valuation.** A two-step form for owners to request a valuation.

Triana neighbourhood page with price per m² — **Prices by area.** One page per neighbourhood, with asking prices compared to the Seville average.

Built for phones

Every screen was designed at 390 px wide first, then scaled up to desktop.

Outcome

The site is live at clickrealestate.es and we continue to maintain it.

automated tests across backend and frontend: 11,000+
areas on the map: 33 Seville neighbourhoods and 44 towns: 77
API endpoints: 700+
scheduled jobs: alerts, market data, backups: 30+

Figures checked against the code and production in November 2026.

Visit clickrealestate.es ↗

Stack: Next.js 15 · React 19 · TypeScript · Tailwind 4 · NextAuth · Leaflet + GeoJSON · FastAPI · SQLAlchemy async · PostgreSQL 16 · Alembic · Redis · Resend · S3 · Sentry · Docker · Hetzner