Why Dutch Property Data Is So Fragmented

The Netherlands maintains several world-class public registries for property data. BAG tracks 9.9 million registered address objects. Kadaster records every ownership transfer. WOZ provides annual tax valuations for every property. CBS publishes detailed housing statistics. Yet if you are building a property application, none of these alone gives you what you need.

The public registries

BAG (Basisregistratie Adressen en Gebouwen) is the national address and building registry. It contains 9.9 million+ registered objects, is updated daily, and has been freely accessible via PDOK since its inception. Every address gets a stable 16-digit verblijfsobject ID. BAG tells you what exists and where, but nothing about market activity.

Kadaster (specifically the BRK, Basisregistratie Kadaster) records ownership and transaction history. Unlike BAG, it is not free. Kadaster charges per document or query, and bulk access requires a subscription agreement. If you need to know who owns a property or what it last sold for, this is the authoritative source.

WOZ (Waardering Onroerende Zaken) is the municipal property tax valuation. All 342 municipalities assess every property annually, using January 1 of the prior year as the reference date (peildatum). WOZ values have been publicly accessible since 2016 via the WOZ-Waardeloket. The structural problem: the 2025 WOZ value reflects the market as of January 1, 2024. In a market that moved 4.9% in 2024 alone, that lag matters.

CBS (Centraal Bureau voor de Statistiek) publishes demographic and housing stock data at the neighbourhood level, but typically on an annual cycle with a publication delay of 6 to 12 months.

EP-Online, operated by RVO (Rijksdienst voor Ondernemend Nederland), tracks energy performance labels. As of 2024, roughly 60% of Dutch dwellings have a registered label.

The listing portals

None of the public registries track what is currently for sale or rent. That data lives in commercial portals, and the market is split along professional lines.

• Funda is the dominant sales platform but only lists properties from NVM member brokers (roughly 75% of the market). Non-NVM brokers and private sellers are excluded.
• Pararius is the largest rental platform, particularly strong in the expat and international segment.
• Jaap.nl and Huislijn aggregate from multiple broker networks but with varying coverage.
• Hundreds of independent makelaar websites carry listings that may not appear on any major portal.

There is no equivalent of the UK's Rightmove or Zoopla that covers close to 100% of the market. The result: a complete view of what is available requires monitoring over a thousand individual sources.

Why this matters for PropTech

Every PropTech startup in the Netherlands discovers this fragmentation the hard way. Whether you are building an automated valuation model, a rental alert app, or an investor analytics dashboard, you end up spending your first six months on data engineering: connecting to PDOK for BAG data, negotiating with Kadaster for transaction history, scraping listing portals, building normalization and deduplication logic. This is table-stakes infrastructure, not product differentiation.

The cost of fragmentation

Fragmentation creates real costs beyond engineering time. Data quality suffers because each source uses different address formats, property type taxonomies, and update frequencies. Without a shared identifier like the BAG ID, joining data across sources requires fuzzy matching that inevitably introduces errors. And because public registries update on different schedules (BAG daily, WOZ annually, CBS with a delay), any composite view of a property is a snapshot stitched together from data of different ages.

The Netherlands has excellent property data. It is just spread across a dozen systems with different access models, update frequencies, and identifiers. Bringing it together is the infrastructure challenge that defines Dutch PropTech.