How to Value a Flat Using Sold Prices in the Same Building

Valuing a terraced house is relatively straightforward. You find a similar street, a similar size, a similar condition, and you have a decent anchor. Flats are trickier. Two flats in the same postcode can be worth meaningfully different amounts because one is on the ground floor facing a car park and the other is on the fourth floor with a south-facing balcony. Location alone does not cut it.

That is why experienced buyers and surveyors lean hard on same-building comparables. If a flat two floors directly above yours sold six months ago, that single data point tells you more than a dozen sales from a nearby block. The building's service charge history, its cladding status, its lease length, its management quality, all of that is already baked in. You are comparing like with like in a way that cross-street comparisons simply cannot match.

This guide walks you through how to find those comparables, how to read them honestly, and how to avoid two specific traps that catch out a lot of first-time buyers.

HM Land Registry publishes every registered sale in England and Wales. The quickest free route is the Land Registry's own Price Paid Data tool, or you can search through Rightmove's sold prices tab or Zoopla's equivalent. Type in the full postcode of the building and filter by flat. You will usually get a list going back to the late nineties.

For a new-build block or a large conversion, you might find dozens of sales. For a small Victorian conversion with four flats, you might find only a handful over twenty years. Both situations are useful, just in different ways, and we will come back to that.

One thing worth knowing is that Land Registry data has a lag. Sales typically appear on the register somewhere between two weeks and three months after completion, sometimes longer if there are legal complications. So the most recent sale you can see might actually be from several months back even if you are searching today. Bear that in mind when you are trying to gauge what the market is doing right now.

Once you have your list of sales, do not just average them. That is the lazy approach and it will mislead you. Instead, work through a short mental checklist for each comparable.

First, is it the same flat type? In most blocks, flats are built to a small number of layouts. A one-bedroom flat and a two-bedroom flat in the same building are not comparable. A ground-floor flat and a top-floor flat of identical square footage can still differ by ten or fifteen percent in value depending on the building and the market, so note the floor where you can.

Second, how long ago did it sell? A sale from three years ago is not useless, but you need to adjust for what the market has done since. A sale from three months ago is gold. Recency matters enormously, and we will explain exactly why in the next section.

Third, was it a cash sale at full market value? This is where it gets interesting.

Here is a trap that catches out a surprising number of buyers. When a developer sells a new-build block, or when a housing association disposes of a tranche of shared-ownership or right-to-buy properties, multiple flats can change hands in a short window, sometimes on the same day. These transactions show up in the Land Registry data just like any other sale.

The problem is that bulk sales are often not arms-length market transactions. A developer might sell ten flats to a single institutional landlord at a negotiated bulk discount. A housing association might sell a group of flats under a specific scheme with conditions attached. If those prices feed into your comparable analysis, they will drag your sense of market value downward in a way that does not reflect what a willing buyer would actually pay on the open market.

How do you spot them? Look for multiple sales on the same date, or sales where the buyer name repeats across several transactions (Land Registry data sometimes shows the buyer entity). If you see a cluster of sales at suspiciously round or similar figures all registered on the same day, treat them with real caution. They are not the same thing as a private buyer paying full price after a competitive viewing.

Some buyers, and even some online valuation tools, take all the sales in a building over several years and calculate an average price per square foot. Then they apply that to the flat they are looking at. It sounds rigorous. It is not.

Property markets move. Interest rates change. Leases get shorter. A building that had no cladding issues three years ago might now be sitting in the middle of an EWS1 dispute. A block that was scruffy in twenty-nineteen might have had a full refurbishment since. A multi-year median smooths over all of that and gives you a number that reflects the past, not the present.

The most recent comparable sale in the same building, assuming it was a genuine open-market transaction, reflects all of those current realities. The buyer who paid that price had access to the same lease information you do, faced the same mortgage market you do, and presumably did their own due diligence. That is enormously valuable information.

If the most recent sale is more than about twelve months old, you can still use it, but you should sense-check it against broader market movement in that postcode over the same period. Rightmove and Zoopla both publish area-level price trend data that can help you do a rough adjustment.

Even within the same building, you need to make adjustments. Here are the main ones to think about.

Floor level generally adds value as you go up, because of light, noise reduction and views, though the top floor can sometimes attract a premium of its own. The ground floor is typically the least desirable in a flatted building unless it comes with private outside space, in which case the equation can flip entirely.

Aspect matters in buildings where one side faces something pleasant, like a park or a quiet garden, and the other faces a main road or a service yard. If your comparable was on the quiet side and your flat is on the noisy side, the sold price is an optimistic starting point, not a ceiling.

Condition is harder to read from sold data alone because you do not know what state the comparable flat was in when it sold. If you can, look at whether it was listed on Rightmove before it sold, which would give you photos. A flat sold with an original nineteen-eighties kitchen is a different proposition from one that had just been refitted.

None of these adjustments need to be precise to the pound. The point is to think directionally. Is your flat better or worse than the comparable on each dimension? That shapes whether you should be pitching your offer above, at, or below the comparable price.

A sensible approach is to build a simple shortlist of comparables before you offer. Aim for two or three sales from the same building within the last two years if you can find them, noting the floor, the flat type, and the date. If the building is small and recent sales are scarce, widen your search to the immediate street or a neighbouring block of the same era and construction type, but be honest with yourself that these are weaker comparisons.

Then ask yourself what the most recent genuine open-market sale tells you. Does the asking price on your flat sit above that, below it, or roughly in line? If it is above, what justifies that? A higher floor, a longer lease, a recent renovation? If it is below, is there something the seller knows that you do not yet?

Do not be shy about sharing your research with your estate agent. A good agent will engage with it. A bad one will waffle. Either way, you learn something useful. And if you are getting a mortgage, your lender's surveyor will do their own comparable analysis. Going into that conversation with your own numbers means you are far less likely to be caught off guard if the valuation comes in below the agreed price.

Knowing your comparables is not just about negotiating a few thousand pounds off. It is about understanding what you are actually buying and making a decision you will still feel good about in five years.