How it works
From postcode to professional valuation report in under a minute. No phone calls, no waiting, no estate agent opinions.
Enter the address
Type a postcode and select the property from our database of 19.6 million UK residential addresses. We match to Land Registry title data, EPC records, and our enriched feature set automatically.
Get your valuation
Our LightGBM gradient boosting model analyses 60+ features — comparable sales, property characteristics, location metrics, and market trends — to produce a point estimate with confidence scoring.
Download the report
A professional PDF report containing the valuation, confidence tier, comparable evidence, feature contributions, methodology disclosure, and data provenance. Ready for lenders, investors, or your own records.
Data sources
Every valuation draws on multiple authoritative government and public-sector datasets, updated monthly. No scraped listings. No asking-price estimates.
HM Land Registry
31M+ actual sale transactions with price, date, property type, and tenure. The ground truth for UK property prices.
Energy Performance Certificates
29M+ EPC records providing floor area, construction age, energy rating, wall type, heating, and more.
ONS House Price Index
Regional and local authority price indices used to time-adjust comparable sales to the valuation date.
Schools & Ofsted
Distance to nearest schools, Ofsted ratings, and school density metrics within catchment areas.
Environment Agency
Flood risk zones (rivers, surface water, coastal) mapped to individual property locations.
Crime & Demographics
Police.uk crime rates and ONS Census demographics mapped to individual property locations.
Methodology overview
We use a LightGBM gradient boosting model trained on millions of actual sale transactions. The model learns the relationship between property features and sale prices, then applies that learning to produce valuations for unseen properties.
Each valuation is validated against a walk-forward backtest: the model is trained on historical data and tested on future sales it has never seen. This mirrors real-world conditions and prevents overfitting.
Read the full methodology →Confidence scoring
Not all valuations are equally reliable. Each property is assigned an individual Forecast Standard Deviation (FSD) based on comparable evidence, feature completeness, and model prediction uncertainty. The three confidence tiers are a human-readable summary of the underlying granular FSD.
Tier 1 — High Confidence
Strong comparable evidence. Standard property type in an active market. Each property is assigned an individual Forecast Standard Deviation — the prediction interval is narrow.
- • Abundant recent comparable sales
- • Good EPC / feature coverage
- • Low prediction uncertainty
Tier 2 — Medium Confidence
Reasonable comparable evidence with some gaps. Each property is assigned an individual Forecast Standard Deviation — the prediction interval is moderate.
- • Some comparable sales available
- • Partial feature coverage
- • Moderate prediction uncertainty
Tier 3 — Low Confidence
Limited comparable evidence. Unusual property, thin market, or missing data. Treat as indicative only.
- • Few or distant comparable sales
- • Significant feature gaps
- • Wide prediction uncertainty
Ready to test us?
Create a free account and backtest against your own data — free, no credit card required. Production valuations are pay-per-use.