Since land registry records became available to download, there has been a significant limitation.  The data contains millions (31.5m) of records, and each sale tells us the address, the sale price, the date of sale, the type of property and so on.  

What we haven't had is a reliable, permanent identifier telling us exactly which property was sold.

That has changed, but only since July 2026.  From this month, the normal data file does not change; what the Land Registry now provides is two separate files.  One with the property UPRN and another with an INSPIRE ID.

So why is this important but also not as beneficial as it first appears?

The address problem

Suppose a property sold in 2005, 2012, 2018 and 2026.

It should be straightforward to find all four sales, right?  Not always.  Addresses can be unreliable identifiers.  For example, one property sold four times could show as:

• Flat 8, 10 High Street
• 10, flat 8 High Street
• 8, 10 High Street
• Apartment 8, 10 High Street

They are all the same property, but not really.  From a data point of view, they are four different addresses.  So for a site like this one, I need to write logic that matches address details like this to ensure, as much as I can, that it treats the four addresses above as the same property.

Add to that, names change, buildings are converted, data input errors at the Land Registry, and so on, all add complications.

Welcome UPRN!

There is a solution to the problem; it's called the Unique Property Reference Number or UPRN for short.  The point of a UPRN is that every property has one.  It's not intended to describe a property; it identifies its location.  It's a permanent reference number for a property.

UPRN already exists; it's not new at all. EPC certificates all have a UPRN, and I used them on this site.  If you go to the EPC search page and zoom right in on the map, you will see that each marker representing a property sits right on top of the actual property, showing you exactly which one it is.  If you do the same for the property search page, it only centres on the postcode and shows a less accurate position. Why?  Because there is no UPRN in the Land Registry data.

Benefits of UPRN

Having a UPRN is great, it allows sites like this to find a property you are searching for and then find a matching UPRN in all sales of the property, no more trying to match addresses that could have slightly different formats.  It also has the advantage of improving property price analysis, but that is not going to be possible just yet.

Limitations

And here we come to the limiting factor.  The Land Registry is not providing a UPRN for all 31.5m records in the dataset.  They are only providing it from July data onwards.  So today, it's really not that useful until there is enough data, and that will take years.  Until then, we are still stuck using fuzzy matching to be sure all sales for a property have been identified.

The Land Registry has been clear: it will not be providing a UPRN for any sales prior to July 2026.

What can PropertyResearch use it for now?

There are ways I can use the data from now, such as accurate mapping of property location.  For example, if you look at the monthly property Dashboard for July (here) I have updated the map to include a property sale view option.  This allows you to zoom in on the map and find properties that sold in July 2026, and the map markers are placed directly on the property location, not just nearby.  Example below.

This allows you to view property sales from July 2026 onwards simply by scrolling around the map, but with precise property locations and a link to that property page with more detailed information.  That will be a big help to many who are browsing without a specific street or postcode in mind.

The same also applies to the high-value property map; from a research perspective, you can simply zoom around the map and get precise locations for the high-value properties selling each month.

It will get better over time as the months go on; it's a victory for data, but it will build slowly into something more meaningful.