How England and Wales record a property sale

England and Wales publish more sale prices than almost anywhere and no floor areas at all. Every transfer of a registered title becomes a public row carrying the price, the date, the address, the property type and the tenure. What it does not carry is how big the property is, which is the one field a price per square metre cannot do without.

Why is the price in one register and the floor area in another?

Because size lives in a different register, every rate published for this market is the product of a join between two datasets that were never designed to be joined: sale records from the land registry and floor areas from energy certificates, matched on address text.

That join is the defining constraint of the market, and it produces two consequences the other two registers here simply do not have. A sale with no matching certificate is counted but cannot be priced per unit of area, so this market has to publish a match rate. And because matching is by address string, it succeeds unevenly: a flat with a number and a postcode matches more reliably than a named house on a rural lane.

The current build carries 609,912 sales in the last twelve months, of which 482,172 found a floor area, across 2,279 postcode districts. Which sales go unmatched, and whether they differ from the rest is the market’s own article on the subject.

What exactly does one Price Paid row carry?

Seven fields do almost all of the work: the price paid, the date of transfer, the address, the property type, a new-build flag, the tenure and a transaction category. There is no floor area, no room count, no condition and nothing about the parties beyond the transfer itself. Everything a rate needs that is not on that list has to come from somewhere else.

The transaction category is the quiet one. The register separates standard market sales from a second category that holds repossessions, buy-to-let purchases where they can be identified and transfers to non-private owners, and a median built over both is a different statistic from a median built over the first alone. Two published figures for the same district can disagree for no reason other than this choice, so which categories a figure includes is the first thing to check about it.

Why are Scotland and Northern Ireland not in it?

Because the register follows the jurisdiction, not the state. HM Land Registry covers England and Wales; Scotland’s sales are recorded by Registers of Scotland and Northern Ireland’s by its own registry, each publishing under its own terms and on its own cycle. A “UK house price” therefore stitches together at least three registers, which is why this network’s market is named England & Wales and covers exactly that.

What does that separation make possible?

The separation is not all cost. Because the certificate register is independent of the sale register, it contains floor areas for homes that have not sold, and it contains more than one measurement for many addresses. That makes it possible to check the join rather than assume it: an address assessed twice, years apart, is the same dwelling measured twice, and the difference between those measurements bounds the error a mismatched date introduces.

A register that publishes its own floor area, as Singapore’s does, offers no such check. It is more convenient and less auditable.

How does it compare with the other two registers?

A reader comparing a French rate with an English one is therefore comparing a figure with no match step against one with a match step, on top of two different definitions of a square metre. Neither figure is wrong. They are not the same measurement.

What does the licence allow?

Price Paid Data is published under the Open Government Licence, which permits commercial reuse with attribution. That is why a dozen independent sites can publish sold-price analysis for this market, and it is the reason the market is competitive in a way that registers with restrictive terms are not. The condition is attribution, and it is stated in full on every page that uses the data.

For what this register contains field by field, and how to download it yourself, RealScout’s own guide is the fuller treatment. What each of the three registers records sets all of them side by side. Two constraints apply to this register as much as to the other two: how many sales a local figure needs before it means anything, and whether the figure is in cash or in today’s money.