Test us on transactions we've never seen
Upload up to 250 recent completions — address, sale date, and sale price. We'll value each property as of the sale date. You compare against the price you already know.
These historical valuations have no commercial value — they exist purely to prove accuracy.
Our current model was trained on sales completed before 1 July 2025.
Sales completed after that date are a blind test — we mark every row you upload so you can see which is which.
Why this test is structurally blind
You submit recent completions
Transactions from the last 6 months. Address, sale date, and the price you know.
We can't have trained on them
Our model was trained on sales completed before 1 July 2025. Anything completed since then cannot be in our training set — and your results flag every sale as blind or seen, so you don't have to take our word for it.
We value as of the sale date
Each property is valued as of the date it sold — a like-for-like comparison. PE5, PE10, PE15, PE20, MdAPE, segmented by property type, region, and price band.
250 properties is enough for statistically robust accuracy metrics, segmented by property type and region. This is more transparency than any other AVM provider offers.
CSV format
Required columns
| Column | Description |
|---|---|
address |
Full address including postcode (e.g. "10 Downing Street, London, SW1A 2AA") |
sale_date |
Completion date (YYYY-MM-DD or DD/MM/YYYY). Must be within the last 6 months. |
sale_price |
Sale price in GBP (e.g. 250000) |
Optional columns
postcode |
Postcode in a separate column — improves matching if not in the address |
uprn |
UPRN — skips address matching if provided |
property_type |
D/S/T/F — used for segmentation in the report |
Upload your data
Create a free account to start backtesting.
Ready for live underwriting?
Once you're satisfied with our accuracy, purchase a valuation bundle and start using Gadsden Valuations for origination screening, portfolio monitoring, and retrospective revaluation.
View pricing