[PRD] Price Paid Data - UK House Price Analytics App

[antonsamper]

Problem Statement

UK house price data is publicly available through the Land Registry, but it's difficult to access and analyse. Users who want quick insights into housing market trends—whether they're buyers, sellers, investors, or simply curious—must navigate cumbersome government interfaces or pay for expensive property data services. There's no simple, fast way to search by any location (address, postcode, town) and instantly see meaningful statistics, trends, and interesting facts about property sales.

Solution

A web application with instant search across all UK house price paid data. Users type any location and immediately see aggregated statistics, with the ability to drill down into individual transactions. The app surfaces insights like price trends over time, property type breakdowns, and "quirky facts" (e.g., most expensive house in a postcode). Free accounts get 2-3 years of data; premium users unlock exports, comparisons, advanced filters, and full historical data.

User Stories

1. As a first-time buyer, I want to search by postcode and see average house prices, so that I can understand if an area is within my budget
2. As a homeowner, I want to see what similar properties in my street sold for, so that I can estimate my home's value
3. As a property investor, I want to compare sales volumes between two postcodes, so that I can identify active markets
4. As a curious user, I want to see the most expensive house ever sold in my town, so that I can satisfy my curiosity
5. As a user, I want to filter results by property type (flat, terraced, semi, detached), so that I can see relevant comparisons
6. As a user, I want to filter by price range, so that I can focus on properties within my budget
7. As a user, I want to filter by date range, so that I can see recent sales or historical trends
8. As a user, I want to filter by number of bedrooms, so that I can compare like-for-like properties
9. As a user, I want to see price trends over time as a chart, so that I can visualise market direction
10. As a user, I want to see a breakdown of property types sold as a pie/bar chart, so that I can understand the housing mix
11. As a user, I want to see a map with prices overlaid, so that I can visualise geographical price variations
12. As a user, I want to click on a statistic and drill down into individual transactions, so that I can see the raw data
13. As a user, I want to see aggregations (median, mean, count, min, max) for any search, so that I can quickly grasp the market
14. As a user, I want to see price records and milestones (e.g., "highest sale this year"), so that I can understand notable events
15. As a free user, I want access to 2-3 years of data, so that I can get useful insights without paying
16. As a premium user, I want to export search results to CSV/PDF, so that I can use the data elsewhere
17. As a premium user, I want to compare statistics between multiple postcodes side-by-side, so that I can make informed decisions
18. As a premium user, I want to use advanced filters like regex patterns, so that I can do sophisticated searches
19. As a premium user, I want access to full historical data (beyond 2-3 years), so that I can analyse long-term trends
20. As a user, I want to create an account, so that I can save my searches and access premium features
21. As a user, I want to see results on a map (OpenStreetMap/Mapbox), so that I can understand location context
22. As a user, I want the search to be instant/very fast, so that I don't have to wait for results
23. As a user, I want to search by partial address, street name, or area, so that I have flexibility in how I search
24. As a user, I want to see when the data was last updated, so that I know how current it is

Implementation Decisions

Data Source

• Primary: UK Land Registry Price Paid Data (updated monthly)
• Future consideration: Research legal alternatives to Rightmove/Zoopla data (direct scraping has legal risks)

Tech Stack

• Frontend: React, Tailwind CSS, Shadcn UI components
• Backend: TypeScript (framework TBD - likely Next.js API routes or separate Node service)
• Database: PostgreSQL with TimescaleDB extension for time-series optimisation
• Maps: OpenStreetMap or Mapbox (not Google Maps due to cost)

Architecture

• Time-series database design to handle monthly data updates efficiently
• Search must be near-instant - likely requires proper indexing strategy and possibly Elasticsearch/Meilisearch for search
• Data ingested monthly from Land Registry bulk downloads

Account Model

• Accounts required (no anonymous access)
• Free tier: 2-3 years historical data, basic filters, view aggregations
• Premium tier: Full history, exports, comparisons, advanced filters (regex)

Search UX

• Single search box accepting address, postcode, town, city, or any combination
• Results show aggregated statistics first
• Users can drill down into individual transactions
• Filters applied post-search to refine results

Monetisation

• Cannot charge for the raw data (it's free by law)
• Charge for premium features: exports, reports, advanced filters, full history

Out of Scope (MVP)

• Mobile app - planned for Phase 2 with location-aware features (see house prices while walking down a street)
• Price alerts/notifications - planned for Phase 2
• Rightmove/Zoopla data - requires legal research into alternatives
• Mortgage calculators - not planned
• Estate agent integrations - not planned
• User-generated content (reviews, comments) - not planned

Further Notes

• Land Registry data is released monthly; the app should clearly show data freshness
• Consider rate limiting and caching to manage infrastructure costs
• The "quirky facts" feature (most expensive, price records) could be a key differentiator for virality
• Mobile Phase 2 vision: GPS-based lookup showing prices for properties you're standing near