UK Land Registry Sold Prices

Sold house prices for England and Wales, straight from HM Land Registry Price Paid Data. Every residential sale registered since January 1995, roughly 30 million transactions, with the address, the price paid, the date, the property type and the tenure. No API key, no account, no scraping of a listings site.

🔍 What does UK Land Registry Sold Prices do?

It queries the Land Registry's own open data API and returns every matching sale as a clean row, ready for a spreadsheet, a valuation model or your own product. Search a postcode, a street, a town, a council district or a whole county, then narrow it by date window and price window.

The quickest way to try it is to leave the prefilled input alone and press Start: it returns the most recent sales in Bath since the start of 2024. Swap town for your own area, or drop a full postcode into postcode, and you have the sold price history for that patch.

This is the register itself, not a portal's copy of it. Every row carries a landRegistryUrl pointing back at the official record for that transaction, so anything you publish can be traced to source.

📊 What data can I extract from Land Registry Price Paid Data?

One row per registered sale:

Field	Description
transactionId	Land Registry unique identifier for the sale
transactionDate	Date of the sale, as a plain calendar date in YYYY-MM-DD
pricePaid	Sale price in whole pounds sterling
address	Full address on one line, assembled from the address parts below
saon	Secondary addressable object name, the flat or unit within a building
paon	Primary addressable object name, the house number or building name
street	Street name
locality	Locality or village within the posting town
town	Posting town or city
district	Local authority district the property sits in
county	County on the address
postcode	Full UK postcode recorded at the time of sale
propertyType	detached, semi-detached, terraced, flat-maisonette or otherpropertytype
estateType	Tenure of the sale, freehold or leasehold
newBuild	True when the property was newly built at the point of sale
transactionCategory	standardpricepaidtransaction for an ordinary sale, additionalpricepaidtransaction otherwise
recordStatus	State of the record in the register, add, change or delete
landRegistryUrl	The official Land Registry page for this transaction
retrievedAt	When the row was collected, ISO 8601

The address is split the way the Land Registry splits it, so saon and paon are separate from street. That matters when you are matching against another address dataset, because "Flat 5, Grosvenor Court, 8" is one building name and one unit rather than free text. The joined address field is there for the cases where you just want a label.

💡 Why pull sold prices from the Land Registry?

Valuation and comparables. Every sale on a street, with dates and prices, so a comparable is evidence rather than an asking price. Asking prices are what a seller hoped for. Price Paid Data is what somebody actually paid.

Area price history. Pull a whole town or district across a date window and you have the raw material for a price index, a time series, or a chart of how a postcode has moved since 1995.

New build tracking. Set newBuildOnly and you get the first sale of every plot in an area, which is how you find which developments completed, when they completed, and what the units went for.

Property data products. Portals, mortgage tools, insurance quoting and conveyancing all want a sold price history keyed on an address. This is the source they are all ultimately reading.

Homeowner and agent research. The last sale price of a specific address, found by postcode and house number, in one call.

Academic and journalistic analysis. The dataset is open government data, so an analysis built on it can be published and reproduced by anyone.

🚀 How do I use UK Land Registry Sold Prices?

1. Click Try for free.
2. Set an area. A full postcode for one street corner, street plus town for one road, town or district for a whole area, county for a region.
3. Set a date window with soldAfter and soldBefore, both YYYY-MM-DD, and a price window with minPrice and maxPrice if you want one.
4. Narrow further with propertyTypes, estateType, newBuildOnly or transactionCategory.
5. Set maxResults to cap what the run returns and therefore what it costs.
6. Click Start, then download the results as JSON, CSV or Excel, or read them straight from the API.

⬇️ Input

Every filter is optional and they all combine. A run with no filters at all is allowed but returns an arbitrary slice of the whole register, so it logs a warning and it is almost never what you want.

{
  "town": "BATH",
  "soldAfter": "2024-01-01",
  "minPrice": 300000,
  "propertyTypes": ["detached"],
  "sortBy": "newest",
  "maxResults": 500
}

Field	Type	Default	What it does
postcode	string		A full UK postcode such as BA1 2XY
street	string		Street name without the house number
town	string		Posting town or city
district	string		Local authority district
county	string		County on the address
soldAfter	string		Earliest sale date, YYYY-MM-DD, inclusive
soldBefore	string		Latest sale date, YYYY-MM-DD, inclusive
minPrice	integer		Lowest price in whole pounds, inclusive
maxPrice	integer		Highest price in whole pounds, inclusive
propertyTypes	array	all	Any of detached, semi-detached, terraced, flat or maisonette, other
estateType	string	any	freehold or leasehold
newBuildOnly	boolean	false	Keep only new builds
transactionCategory	string	any	standard or additional
sortBy	string	newest	newest, oldest, highest-price, lowest-price or fastest
maxResults	integer	500	Hard cap on rows returned, and on cost. Maximum 10000
maxRecordsScanned	integer	20000	Safety stop when the property filters are narrow

Case and spacing do not matter. bath becomes BATH and ba1 2xy becomes BA1 2XY before the request goes out, because the register stores everything in upper case and matches exactly.

⬆️ Output

Table view

Results arrive as a Sales table you can sort and filter in the Console, with the date, price, address, type and tenure lined up for scanning. There is a second view, Address parts, which splits the address into the Land Registry's own fields for anyone matching against another dataset.

JSON

A typical row:

{
  "transactionId": "7663DF8A-3554-4438-96E4-6483101AE00A",
  "transactionDate": "1996-05-17",
  "pricePaid": 54000,
  "address": "FLAT 5, GROSVENOR COURT, 8, PARSONAGE ROAD, STOCKPORT, STOCKPORT, SK4 4JZ",
  "saon": "FLAT 5",
  "paon": "GROSVENOR COURT, 8",
  "street": "PARSONAGE ROAD",
  "locality": "STOCKPORT",
  "town": "STOCKPORT",
  "district": "STOCKPORT",
  "county": "GREATER MANCHESTER",
  "postcode": "SK4 4JZ",
  "propertyType": "flat-maisonette",
  "estateType": "leasehold",
  "newBuild": false,
  "transactionCategory": "standardpricepaidtransaction",
  "recordStatus": "add",
  "landRegistryUrl": "https://landregistry.data.gov.uk/data/ppi/transaction/7663DF8A-3554-4438-96E4-6483101AE00A/current",
  "retrievedAt": "2026-08-16T09:41:02.118Z"
}

newBuild is true, false or null. It is null when the register published no value for that sale, because a missing flag is not the same statement as a sale that was not a new build.

Download it from the run as JSON, CSV or Excel, or read it straight from the API.

Every run also writes a RUN_SUMMARY record to the key-value store with the search that ran, how many records were read, how many were saved and why the run stopped: complete when the result set was exhausted, max-results-reached, scan-limit-reached, time-limit-reached, charge-limit-reached or error. That is how you tell a search that found everything from one that was capped part way.

Coverage and currency

Price Paid Data covers England and Wales only. Scotland is held by Registers of Scotland and Northern Ireland by Land and Property Services, and neither appears here.

The record starts on 1 January 1995 and runs to the most recent monthly publication. A sale enters the register some weeks after completion, so expect roughly a two month lag at the leading edge. The newest few weeks of any area will look thin, and they fill in later. That is the publication cycle rather than a gap in this Actor.

Sales are registered transactions of residential property sold for value. Transfers that were not sales, sales that were never registered, and commercial property outside the residential definition are not in the dataset at all.

How the filters behave

The address, date and price filters are applied by the Land Registry query service, so narrowing on those makes the run faster as well as cheaper. Two details are worth knowing because both are silent when you get them wrong.

Address matching is exact and upper case. The register stores BATH, not Bath, and a lower case request is accepted and returns nothing. This Actor upper cases every address field for you, so it is only a problem if you go to the API directly.

Only whole postcodes match. The index is on full postcodes, so BA1 on its own selects nothing rather than selecting all of BA1. Give a full postcode, or use town, district or county for an area. A partial postcode is rejected with an explanation instead of running and returning an empty dataset.

Dates must be YYYY-MM-DD. Any other format is accepted by the API and matches zero rows, which is indistinguishable from a street that never sold. The Actor validates the format before the request goes out and stops with the reason.

propertyTypes, estateType, newBuildOnly and transactionCategory are applied to the rows as they arrive rather than by the API. The query service times out when those are combined with an address filter, so doing it this way is the difference between an answer and a ninety second failure. It costs nothing extra, but a very narrow choice means reading more records to fill the run, which is what maxRecordsScanned bounds.

Sorting and speed

sortBy is handled by the query service and adds several seconds per page. newest is the default because a sold price history is almost always wanted newest first. If you are pulling a whole area and intend to sort it yourself afterwards, set sortBy to fastest and the run drops to a fraction of the time.

An unsorted page of 200 sales comes back in about a second. The same page sorted takes anything from two seconds to the ninety second point where the query service gives up, depending on how wide the area is. That is why maxResults tops out at 10000: it is what fits inside the 300 second default run timeout. A run that runs out of time stops on its own, keeps every row it collected and records time-limit-reached in the run summary, so it never dies mid page with nothing to show. For more than that in one go, raise the run timeout in the Console, split the search by date window, or switch to fastest.

Data source

HM Land Registry Price Paid Data, published at landregistry.data.gov.uk under the Open Government Licence v3.0. This Actor reads the public Linked Data API, which needs no key and no account.

Contains HM Land Registry data © Crown copyright and database right. Reuse is permitted under the Open Government Licence, and the licence asks you to carry that attribution wherever the data is published.

💰 How much does it cost?

You pay per sale returned. A run that matches nothing returns nothing and costs nothing beyond the standard Actor start.

maxResults is the cap on both rows and cost, so setting it is how you bound a run in advance. Narrowing with soldAfter, minPrice and a tight area also keeps you from paying for records you were only going to discard afterwards.

Paid Apify plans get a lower rate per sale. The current price and the plan discounts are shown on this Actor's pricing tab.

🔌 Integrations

Send results straight to Google Sheets, Slack, Airtable, Zapier, Make or your own webhook using Apify integrations. Runs can be scheduled, so a monthly job that pulls the newest sales in your patch and appends them to a sheet takes about a minute to set up. You can also trigger a run from another tool, which is the usual way to attach a sold price history to a new record in a CRM.

🔗 Using UK Land Registry Sold Prices with the Apify API

curl -X POST "https://api.apify.com/v2/acts/spookyweb~uk-land-registry-sold-prices/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"town": "BATH", "soldAfter": "2024-01-01", "minPrice": 300000, "maxResults": 500}'

Or with the Apify client:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_TOKEN' });
const run = await client.actor('spookyweb/uk-land-registry-sold-prices').call({
  town: 'BATH',
  soldAfter: '2024-01-01',
  minPrice: 300000,
  propertyTypes: ['detached'],
  maxResults: 500,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Full detail is in the Apify API reference, and every run is also callable from the Python and JavaScript clients.

❓ FAQ

How far back does the sold price data go?

To 1 January 1995. Every residential sale in England and Wales registered since then is in the dataset, which is roughly 30 million transactions. Anything before 1995 is not held in Price Paid Data at all.

Does it cover Scotland or Northern Ireland?

No. HM Land Registry covers England and Wales. Scottish sold prices are held by Registers of Scotland and Northern Irish sales by Land and Property Services, and neither is part of this dataset.

How recent are the newest sales?

There is roughly a two month lag between a sale completing and the transaction appearing in the register, because the data is published on a monthly cycle after registration. The last few weeks of any area will look sparse and will fill in over the following months.

Can I get the sold price history for one address?

Yes. Set postcode to the full postcode and leave everything else open, then pick out the rows whose paon matches the house number or building name. A postcode usually covers a handful of properties, so the whole history comes back in one small run.

What is the difference between a standard and an additional transaction?

Standard covers ordinary sales at full market value between private individuals. Additional covers everything that does not meet those criteria: repossessions, buy to let purchases, transfers under a power of sale, and sales to or from a company. Both are real prices, but if you are building comparables you usually want transactionCategory set to standard, because an additional sale is often below market value.

Why does my search return nothing?

Four usual causes. The area is spelled differently in the register than in everyday use, most often the posting town rather than the council. The postcode is partial, and only whole postcodes match. The date window is inside the two month publication lag. Or the property type and tenure filters between them exclude everything in the area. The run log names which of these applied.

What do saon and paon mean?

They are the British Standard address terms the Land Registry uses. paon is the primary addressable object, meaning the house number or the building name. saon is the secondary one, meaning the flat or unit inside that building. Most houses have a paon and no saon, and flats have both.

Do I need a Land Registry API key?

No. Price Paid Data is open government data, published without quota. The only practical limits are the ones this Actor works around, which are documented above under how the filters behave.

⚖️ Is it legal to use Land Registry sold price data?

Yes. HM Land Registry publishes Price Paid Data deliberately, as open government data, so that the property market is transparent and anyone can see what a property actually sold for. It is transaction and address information about property, published by the registry itself.

The data is released under the Open Government Licence v3.0, which permits commercial use, adaptation and redistribution provided you carry the attribution: contains HM Land Registry data © Crown copyright and database right, licensed under the Open Government Licence v3.0. Apify's ethical scraping guide covers the wider picture.

👍 Your feedback

Found a bug, or want a field that is not here yet? Open an issue on the Actor's Issues tab. Requests that make the data more useful get built, and problems get fixed quickly.

🔎 You might also like

Actor	What it does
UK Planning Applications	Planning applications from every UK council, tagged by the type of work proposed
UK Food Hygiene Ratings	Official FSA hygiene ratings for every UK food business, searchable by council or postcode
Website Contact Scraper	Company websites in, one contact record per domain out, with emails and social links