HM Land Registry (HMLR) gives UK PropTech businesses free, commercially usable property data through several open datasets and a RESTful JSON API. The flagship is Price Paid Data: over 24 million sold-price records covering England and Wales back to January 1995, updated monthly and released under the Open Government Licence, so you can build commercial products on it without paying a penny. INSPIRE Index Polygons add indicative freehold boundary extents, each carrying a unique Land Registry INSPIRE ID. The API Catalogue lists around 15 HMLR APIs. Paired with the free EPC register, these three datasets form a viable minimum stack for automated valuation models, development opportunity finders, and portfolio analytics platforms. The traps: data stops at the England-Wales border, there is no rental or asking-price data, 2025 inspection fees rose from £3 to £7, and UK GDPR still applies to any personal data you touch.
Last updated: June 2026
HM Land Registry makes available a layered set of property datasets, ranging from completely open and free to commercial through to restricted records that require a paid, per-title inspection. The open layer is what most PropTech businesses build on, and it is genuinely usable for commercial products. The restricted layer, which includes the actual register entries naming proprietors and charges, sits behind a fee and behind UK GDPR obligations. Understanding which tier a dataset belongs to is the single most important decision you make before writing a line of code, because it determines your unit economics, your compliance burden, and whether your product can scale.
The open tier is licensed under the Open Government Licence (OGL). That licence permits commercial reuse, copying, adapting, and combining with other data, provided you attribute the source. This is the foundation of almost every UK property-tech product you have used: sold-price lookups, valuation estimates, and area analytics all draw from OGL data. The restricted tier, by contrast, is the register itself, which is information about who owns a specific title and what charges sit against it. That is available on a pay-per-view basis and cannot be bulk-scraped or freely redistributed.
Our honest view: founders consistently overestimate how much they get for free and underestimate the licensing nuance. You can build a powerful product entirely on open data, but the moment your roadmap touches "show the owner's name" or "monitor a specific title for changes", you cross into paid, GDPR-heavy territory. Plan that boundary deliberately rather than discovering it after launch.
| Dataset | Coverage | Licence | Cost | Typical use |
|---|---|---|---|---|
| Price Paid Data (PPD) | England and Wales, from Jan 1995 | Open Government Licence | Free, commercial use allowed | AVMs, market analytics, comparables |
| INSPIRE Index Polygons | England and Wales | Open Government Licence | Free | Boundary mapping, plot identification |
| UK House Price Index data | UK-wide aggregates | Open Government Licence | Free | Trend indices, area benchmarking |
| Official Copy of Register / Title | England and Wales, per title | Restricted, fee-bearing | £7 per digital inspection (2025 fee) | Conveyancing, ownership verification |
| Title and Charge data feeds | England and Wales | Commercial agreement | Bespoke licence | Lender and conveyancer platforms |
Note immediately what is missing: Scotland and Northern Ireland. HMLR covers England and Wales only. Registers of Scotland and Land and Property Services in Northern Ireland are separate bodies with their own datasets, formats, and licences. If your pitch deck says "UK-wide", your data layer does not yet support it, and bolting on the other nations is a real project, not a config change.
Price Paid Data (PPD) is HMLR's record of residential and commercial property sales lodged for registration in England and Wales, and it is the single most valuable free dataset for PropTech. It contains over 24 million transaction records reaching back to January 1995, it is refreshed monthly, and it is released under the Open Government Licence so you can build paid commercial products on top of it. Each record carries the sale price, the transaction date, the full address components, the property type, whether it was a new build, the tenure (freehold or leasehold), and a PPD category that flags standard arm's-length sales versus other transfers.
The release history matters for anyone backfilling. HMLR phased PPD out from March 2012 and completed the full 1995-onward backfile by November 2013. Today you can ingest the entire history as a single bulk file or take the monthly update files, which are far lighter and the right choice once you have done your initial load. A sensible architecture loads the complete file once, then applies the monthly "additions, changes and deletions" delta files so your dataset stays current without re-importing 24 million rows every month.
What you can build on PPD alone is substantial. Sold-price comparables tools, area-level price trend dashboards, the comparable-evidence layer of an automated valuation model, transaction-volume heatmaps for investors, and "what did the neighbours pay" consumer hooks all run on PPD. It is the comparables engine of the UK property internet.
Be sceptical of anyone who claims a "complete" valuation purely from PPD. The honest limitation is that PPD tells you the agreed sale price, not the asking price, not the rental value, and not the property's current condition or internal specification. Two identical-footprint houses on the same street can transact 30% apart because one is refurbished and one is a probate sale. PPD is necessary for credible valuation; it is not sufficient on its own.
INSPIRE Index Polygons and the EPC register are the two companion datasets that turn a sold-price database into a genuine property intelligence platform. INSPIRE polygons give you the spatial dimension, indicative freehold boundary extents, and the EPC register gives you the building-attribute dimension, floor area, energy rating, and construction characteristics. PPD tells you what sold; INSPIRE tells you where its boundary sits; EPC tells you what the building is actually like inside. Combine all three and you have moved from "a list of prices" to "a model of properties".
INSPIRE Index Polygons are published under the Open Government Licence and represent the indicative extent of registered freehold land parcels in England and Wales. Each polygon carries a unique Land Registry INSPIRE ID, which is the join key you use to connect a parcel on a map to other datasets. The word "indicative" is doing real work here: these are not the definitive legal title plans, they are an index layer, and they can be imprecise at the edges, overlap, or lag behind subdivisions. For development feasibility and plot-area estimation they are excellent; for legal boundary disputes they are not authoritative.
The EPC register, published by the Department for Levelling Up, Housing and Communities through its open data service, is the dataset most founders forget and most regret forgetting. Every property marketed for sale or let since 2008 needs an Energy Performance Certificate, and the register exposes the certificate data: total floor area in square metres, current and potential energy rating, property type, built form, and main fuel type. Floor area alone is transformative, because price per square metre is the single most important normaliser in any credible valuation model, and PPD does not contain it.
| Dataset | What it adds | Key field | Join key | Limitation |
|---|---|---|---|---|
| Price Paid Data | Transaction price and date | Sale price | Address / postcode | No floor area, no condition |
| INSPIRE polygons | Plot boundary and area | INSPIRE ID | Spatial overlap | Indicative only, freehold |
| EPC register | Floor area and energy rating | Total floor area (m²) | Address / UPRN | Only properties with a certificate |
The hard part is the join. None of these three datasets shares a single clean primary key. PPD uses address components, EPC uses its own address strings plus sometimes a UPRN, and INSPIRE uses geometry. A robust platform invests heavily in an address-matching and UPRN-reconciliation layer early, because the quality of every downstream feature depends on it. This is precisely the kind of unglamorous data-engineering work that separates a credible product from a demo, and it is where many in-house teams stall. A focused business process automation approach to the ingest and matching pipeline pays for itself within weeks.
You access HMLR data programmatically through a mix of bulk file downloads for the open datasets and a RESTful JSON API for the live, query-driven services listed in the GOV.UK API Catalogue, which catalogues around 15 HMLR APIs. For Price Paid Data and INSPIRE polygons, the practical pattern is scheduled bulk ingestion rather than per-request API calls, because you want the whole dataset locally for fast comparables querying. For live register lookups, change monitoring, and per-title services, you use the authenticated REST APIs, several of which require a formal data services agreement with HMLR before you receive credentials.
The core "Use land and property data" service exposes RESTful JSON endpoints, which means standard HTTP verbs, JSON request and response bodies, token-based authentication, and predictable status codes. Any modern stack consumes it comfortably. The realistic engineering shape of a PropTech ingest looks like a small scheduled worker that pulls the latest files, validates them, loads them into staging, runs your matching and enrichment, and promotes to production tables behind a feature flag so a bad release never corrupts live data.
| Access method | Best for | Auth | Update cadence |
|---|---|---|---|
| PPD bulk download (CSV) | Comparables, analytics, AVM training | None (OGL) | Monthly |
| INSPIRE bulk download (GML) | Boundary and plot mapping | None (OGL) | Monthly |
| EPC open data API | Floor area and energy enrichment | Free API key | Continuous |
| HMLR RESTful JSON API | Live register and title services | Data services agreement | Real time |
A minimal ingest worker in pseudocode reads like this, and it is deliberately boring because boring is reliable. Pull the monthly delta over HTTPS. Verify the checksum and row count against the published manifest. Load into a staging table. Run address normalisation against your reference set. Diff against production. Apply additions, changes, and deletions. Swap the live view. Emit a metric for rows changed so you get an alert if a release looks anomalous, for example if 90% of rows suddenly "change". That anomaly check has saved more than one platform from publishing a corrupt month.
For the live REST endpoints, treat HMLR's terms seriously. Per-title inspections are fee-bearing at £7 per digital inspection following the 2025 fee rise, the first in a decade, so a product that fires an inspection on every page view will bleed money. Cache aggressively within the licence terms, batch where permitted, and only hit the paid endpoint when a user genuinely needs a fresh authoritative copy. If you are wiring these calls into a customer-facing workflow, a custom CRM with proper request caching and audit logging is the right home for that logic.
Use free HMLR open data when you have engineering capacity and want to own your data layer; use a paid aggregator such as PropertyData, Searchland, or emapsite when speed to market matters more than margin and you need datasets HMLR simply does not publish. This is not a binary religious choice. The mature pattern is to build your core comparables and analytics on free OGL data, where the marginal cost is zero and the dataset is yours forever, and to buy the gap datasets, rental yields, planning applications, EPC enrichment at scale, ownership intelligence, where someone else has already done the painful aggregation.
The aggregators add real value precisely where HMLR has gaps. PropertyData and similar services blend PPD with rental indices, yield estimates, and area demographics. Searchland focuses on land sourcing and development opportunity intelligence, layering ownership, planning, and constraints over the map. emapsite and similar mapping specialists package boundary, environmental, and survey-grade geospatial data. You are paying them to have already solved the address-matching and multi-source-join problem that would otherwise consume your first quarter.
| Approach | Upfront cost | Ongoing cost | Time to market | Best when |
|---|---|---|---|---|
| Free HMLR open data | High (build pipeline) | Near zero | Slow (8-12 weeks) | You will scale and want margin |
| Paid aggregator API | Low | Per-call or subscription | Fast (days) | Validating an idea quickly |
| Hybrid (free core + paid gaps) | Medium | Targeted | Medium | Most serious products |
Our blunt recommendation: prototype on a paid aggregator to prove the product, then migrate the high-volume, margin-sensitive core to your own free-data pipeline once you have traction. Founders who insist on building everything from raw HMLR files before validating demand routinely burn three months on data engineering for a product nobody wanted. Founders who stay on per-call aggregator pricing forever watch their gross margin evaporate the moment they scale. The hybrid is not a compromise, it is the correct answer for almost everyone.
One more honest caveat on the paid premium tiers. HMLR's INSPIRE polygons for local land charges, for example, sit behind a premium commercial licence quoted around £20,000 plus VAT per year. That is a specialist, high-value licence aimed at search providers, not a casual line item. Read the licence schedule before you assume a dataset is in your budget, because the gap between "free OGL" and "premium commercial agreement" can be five figures annually.
The most realistic and commercially proven PropTech products on UK open property data fall into four archetypes: automated valuation models, development opportunity finders, portfolio analytics platforms, and conveyancing or lender workflow tools. Each maps to a specific combination of datasets, and knowing exactly which datasets an archetype needs lets you scope an MVP honestly rather than promising features your data cannot support. The single most common founder mistake is pitching a product whose core feature depends on a dataset that does not exist in the open tier, the classic being a rental-yield platform built before realising HMLR holds no rental data.
An automated valuation model (AVM) is the most data-hungry archetype and the most valuable when done credibly. It needs PPD for comparables, EPC for floor area to compute price per square metre, the House Price Index for time-adjustment, and ideally INSPIRE for plot context. A development opportunity finder, by contrast, leans on INSPIRE polygons to spot large or under-developed plots, planning data to see what has been permitted nearby, and PPD to value the end product. A portfolio analytics platform aggregates PPD, EPC, and HPI across a landlord's holdings to track value, energy compliance, and exposure. Conveyancing and lender tools sit at the paid end, querying the live register for authoritative ownership and charge verification.
| Product archetype | Datasets required | Paid data needed? | Primary buyer |
|---|---|---|---|
| Automated valuation model | PPD + EPC + HPI + INSPIRE | No (open data sufficient) | Agents, lenders, portals |
| Development opportunity finder | INSPIRE + planning + PPD | Planning data usually paid | Developers, land sourcers |
| Portfolio analytics platform | PPD + EPC + HPI | No | Landlords, asset managers |
| Conveyancing / lender tool | Live register + title data | Yes (per-title fees) | Solicitors, mortgage lenders |
For a first product, our view is to start with portfolio analytics or a focused AVM, because both run entirely on free data, both have clear buyers, and neither requires a regulatory perimeter you might trip over. Development finders are exciting but planning data is fragmented across local authorities and partly paid. Conveyancing tools touch real money and real liability, so they belong to teams who already understand the regulatory and indemnity landscape. Whatever you build, the customer-facing layer, lead capture, onboarding, automated follow-up, can be assembled quickly with a GoHighLevel automation setup so your engineers stay focused on the data engine rather than rebuilding marketing plumbing. Adding an AI chatbot that answers "what's my property worth" by querying your AVM is a proven top-of-funnel hook.
The biggest compliance traps in UK property data are assuming the Open Government Licence covers everything, processing personal data without a lawful basis under UK GDPR, and ignoring third-party rights that the OGL explicitly excludes. The OGL is generous, but it is not a blanket permission, and property data routinely contains personal data the moment it identifies a living individual, for example a named proprietor on a register entry. Get the licensing and data-protection foundation wrong and you risk an Information Commissioner's Office enforcement action, a breach of HMLR's terms, and reputational damage that no amount of clever product can repair.
Start with the OGL exclusions. The Open Government Licence does not grant any rights in third-party intellectual property, personal data that the licence does not cover, or trademarks and logos. Price Paid Data is published as open data precisely because the address-and-price combination is treated as suitable for release, but the moment your product joins datasets in a way that re-identifies and profiles individuals, UK GDPR engages regardless of where the raw data came from. Open licence is a copyright concept; data protection is a separate, parallel obligation.
UK GDPR, enforced by the ICO under the Data Protection Act 2018, requires a lawful basis for processing personal data, transparency about what you do with it, data minimisation, and appropriate security. For a property platform, the high-risk surfaces are: storing proprietor names from register inspections, building profiles of individual owners, and any feature that infers something sensitive about a person from their property. If your product touches those, you likely need a Data Protection Impact Assessment, a clear privacy notice, and a documented lawful basis, usually legitimate interests, with a balancing test on file.
The honest rule we give every PropTech client: treat the comparables and analytics layer (prices, areas, trends) as low-risk open data you can build freely upon, and treat anything that names or profiles a person as high-risk personal data requiring a documented basis and a DPIA. Keep those two worlds architecturally separate. If you would not be comfortable explaining a feature to the ICO in plain English, redesign it before you ship it.
Softomate Solutions builds UK PropTech data platforms through a five-stage process that takes a typical minimum viable product from kick-off to live in roughly eight to fourteen weeks, with a fixed quote agreed before development begins. We are a London-based software and automation agency in Stanmore (HA7), and we specialise in turning open and licensed datasets into reliable, compliant products: ingest pipelines, valuation models, analytics dashboards, and the automated customer-facing layer around them. We quote fixed scope and fixed price so you are never surprised by an open-ended day rate, and we build the data engineering, the application, and the lead-capture automation as one coherent system.
Stage one is discovery and data architecture, where we map exactly which datasets your product needs, confirm the free-versus-paid split, and design the address-matching and join strategy that everything else depends on. Stage two is the ingest pipeline: scheduled workers that pull PPD, INSPIRE, and EPC, validate them, and load them into a spatially indexed database with anomaly alerting. Stage three builds your core product logic, the AVM, the analytics, or the opportunity finder. Stage four wires in the application UI and the automated marketing and onboarding layer. Stage five is compliance hardening, performance, and launch, including the GDPR documentation and attribution your product legally needs.
| Stage | What happens | Typical duration | Output |
|---|---|---|---|
| 1. Discovery and data architecture | Dataset mapping, licensing review, join design | 1-2 weeks | Technical spec and fixed quote |
| 2. Ingest pipeline | PPD, INSPIRE, EPC ingestion and matching | 2-4 weeks | Live, validated data layer |
| 3. Core product logic | AVM, analytics, or opportunity engine | 2-4 weeks | Working product engine |
| 4. Application and automation | UI, dashboards, lead capture, onboarding | 2-3 weeks | Customer-facing platform |
| 5. Compliance and launch | GDPR, attribution, performance, go-live | 1-2 weeks | Compliant live product |
Indicative pricing: a focused MVP on open data, for example a portfolio analytics dashboard or a single-archetype AVM, starts from around £14,000 as a fixed-quote build. A fuller multi-dataset platform with live integrations, custom modelling, and automated marketing workflows typically lands in the £25,000 to £45,000 range, again fixed-scope and fixed-price. Ongoing data pipeline maintenance and monthly dataset refresh handling is available from around £600 per month. We give every prospect a fixed written quote after stage one discovery, so you decide with full cost visibility before any build commitment. Explore how our AI automation agency and software development teams combine on these projects, or talk to us directly through our contact page.
Yes. Price Paid Data is published under the Open Government Licence, which permits commercial reuse, copying, adapting, and combining with other data, provided you display the required attribution. You can build and sell products on PPD without paying HMLR for the data itself.
No. HMLR covers England and Wales only. Scotland is served by Registers of Scotland and Northern Ireland by Land and Property Services, both separate bodies with their own datasets, formats, and licensing terms. A genuinely UK-wide product must integrate all three sources.
Monthly. HMLR publishes a complete file and monthly update files containing additions, changes, and deletions. Best practice is to load the full file once, then apply the monthly deltas so prior-month corrections are honoured rather than just appending new rows.
Following the 2025 fee rise, the first in a decade, a digital inspection of the register or title plan costs £7. That replaced the previous £3 fee. Because every per-title inspection is fee-bearing, cache results within the licence terms rather than querying on every page view.
Yes, credibly. Combine Price Paid Data for comparables, the EPC register for floor area to calculate price per square metre, and the UK House Price Index for time-adjustment. INSPIRE polygons add plot context. The main limitation is that none of these capture a property's current internal condition.
It is an open-data spatial layer showing the indicative extent of registered freehold land parcels in England and Wales, each with a unique Land Registry INSPIRE ID. It is excellent for plot identification and development feasibility but is indicative only, not an authoritative legal boundary plan.
It can. An open copyright licence is separate from data protection law. The moment your product processes personal data, for example a named proprietor from a register inspection, UK GDPR applies and you need a documented lawful basis, transparency, and often a Data Protection Impact Assessment.
Most serious products use both. Build your high-volume, margin-sensitive comparables core on free HMLR data, and buy gap datasets, rental yields, planning, ownership intelligence from aggregators where HMLR publishes nothing. Prototype on a paid API, then migrate the core to your own pipeline as you scale.
HMLR holds no rental data, no asking prices, no internal condition or specification data, and nothing for Scotland or Northern Ireland. It records what sold, not what was listed or what a property could let for. Plan around these gaps rather than discovering them after launch.
A focused minimum viable product on open data typically takes eight to fourteen weeks, covering data architecture, ingest pipeline, core product logic, the application layer, and compliance hardening. The address-matching and dataset-join work is usually the largest and most underestimated portion.
UK open property data is one of the strongest free foundations any PropTech founder can build on: Price Paid Data delivers 24 million transactions back to 1995 under the Open Government Licence, INSPIRE polygons add boundaries, and the EPC register supplies the floor area that makes valuation credible, all combinable into AVMs, analytics platforms, and opportunity finders. The discipline is in the detail. Plan around the England-Wales border, the absence of rental and asking-price data, the £7 per-title inspection fee, and the UK GDPR obligations that the open licence never waives. Most successful products run a hybrid: free data for the margin-sensitive core, paid aggregators for the gaps. Get the address-matching layer right early and the rest follows. The opportunity in 2026 is real, the datasets are richer than founders assume, and the businesses that win will be the ones that treat data engineering and compliance as features, not afterthoughts.
If you are ready to turn HMLR and open property data into a working, compliant product, talk to our team about a fixed-quote build through our London software development service or get in touch for a free scoping call.
Written by Deen Dayal Yadav, Founder of Softomate Solutions, a London-based software development and AI automation agency in Stanmore (HA7). With over 12 years building software, data pipelines, and automation systems for UK businesses, he helps PropTech founders and established property firms turn open and licensed datasets into reliable, compliant platforms. Softomate Solutions is registered at Companies House and works with clients across London and the wider UK. Learn more about our team and approach.
We protect the real names of all clients featured in examples and case studies. Every testimonial is from a real client.
More from the blog
Work with us
Book a free 30-minute discovery call with DD and get a personalised automation roadmap.
Softomate Solutions Limited
Deen Dayal Yadav
Online
We use essential cookies to keep the site running. With your permission, we also use analytics cookies to understand how visitors use our site so we can improve it. No data is sold. Privacy Policy
