Case study
Voolu
PropTech platform turning fragmented property, zoning, and valuation data into natural-language queries, with an AI valuation model and Mapbox visualisations.
- WordPress
- PHP
- Custom theme
- Mapbox
- AI integration
- Semantic HTML
What it is
A geospatial intelligence platform for property professionals. Users ask the system a question in plain English ("flood risk for this postcode", "comparable valuations within 500m") and the platform pulls together zoning, valuation, environmental, and demographic layers into a single answer.
Two surfaces carry the Voolu name. This case study is the public marketing site at voolu.co.uk: a custom WordPress theme. The product web app behind it, app.voolu.co.uk, is a separate Next.js build I led the frontend on.
Build notes
- Custom WordPress theme. No page builder, no off-the-shelf template.
- Mapbox for the spatial layer, with custom styles matched to the brand palette rather than a default basemap.
- Waitlist capture wired through the WordPress backend. No third-party form service sits in the critical path, so a signup does not depend on someone else's uptime or someone else's cookie.
- Performance budget enforced on the marketing surface: real headings, responsive imagery, deferred non critical assets.
The part that is easy to underestimate
Marketing sites for data products have a specific failure mode: the site promises a capability the product has not shipped yet, and the gap becomes the sales team's problem. The useful work here was less about the theme and more about keeping the claims on the page tied to what the platform could actually answer on the day.