Pompy.app: a heat pump comparison site with over 600 models

A complete platform: a database, selection tools and a back office for handling requests.

• Product database - more than 600 air-source heat pump models from 3 to 20 kW, each described by more than 150 technical parameters.
• Filters and rankings - narrowing the list by power, type, brand, refrigerant and price, plus ready-made shortlists for less experienced users.
• Comparison tool - specifications of several selected models side by side in a single table.
• Content and data panel - a headless CMS with a PostgreSQL database and a metrics dashboard: new requests, most viewed models, comparison statistics.
• Request handling - a form that collects details of the building, heating demand and location, routed automatically to the right adviser or installer.

Heat pump information was scattered across manufacturer websites and trade catalogues. There was no independent place to compare models from different brands.

The client came to us with an idea and an extensive data spreadsheet. The data needed thorough standardisation, but it was the starting point for a central platform. We had to turn raw data into a consistent, scalable system and, at the same time, design the whole product from scratch.

The service had to be friendly to someone looking for a heat pump for their home, and at the same time deliver valuable requests to installation companies.

Three tools take users from a general overview to a specific model: filters, rankings and the comparison tool.

Filters narrow the list by heating capacity, unit type, brand, refrigerant and price range. Rankings show recommended units in each category to people who are new to the subject. The comparison tool sets selected models out in a table, so differences in efficiency, noise, costs and warranty are visible on one screen.

From interest to request

At every stage, users can fill in a short quote request form. It collects contact details and information about their needs, and the request goes automatically to the right person along with the technical context the user provided.

The client got a product they develop on their own: they add new models, content and guides without developers.

The database was designed so new records don't slow it down, and its content grows with the market. The architecture is modular, so adding new unit categories doesn't require rebuilding the system. The dashboard shows the number of requests, most viewed models and comparison statistics in real time, so decisions about further development are based on data.