UK House Price Change Analysis

This project visualises the long-term change in house prices across the United Kingdom using government-published price paid data from 1995 to 2024. The goal was to explore regional disparities in property value growth and present a geospatial heatmap to highlight patterns of affordability, demand, and investment hotspots.

The final output is an interactive choropleth-style dot density map using Plotly and Mapbox, showing price change by postcode centroid, allowing a deeper understanding of how the housing market evolved across regions.

Key challenges included:

• Massive dataset of over 10 million property transactions to clean and compress
• Significant postcode and NUTS region mismatches required reconciliation
• Outlier prices and rapid appreciation rates distorted trend clarity
• Needed a colour gradient that accurately reflected percentage change without exaggerating urban spikes

• Used Python and Pandas to clean, merge, and filter raw data from the UK Land Registry
• Removed anomalies such as sales under £10,000 or over £10M to improve map readability
• Mapped each property to its NUTS2 region using GeoPandas and postcode lookup tables
• Calculated average price by year and region, then computed % change over the 29-year period
• Used Plotly’s scatter_mapbox to generate a responsive UK-wide heatmap with scaled colourbars

The implementation of this sales performance dashboard delivered significant business value:

• Highlighted South East hotspots and rural underperformance with clarity
• Created the basis for a regional affordability index or price elasticity estimator
• Useful for public policy, investment planning, and spatial inequality studies

The technical architecture was designed for scalability and reliability:

• Language: Python
• Libraries: Pandas, Plotly, GeoPandas, NumPy
• Mapping: Plotly’s Mapbox visualisation with region-based filtering
• Data Source: HM Land Registry Price Paid Dataset
• Optimisation: Sampling + rolling averages to preserve performance