What a home is actually worth (and why the headline number lies)

There is a number on every property listing, and it is almost always the least honest thing on the page. The asking price is a negotiating position. The portal's "estimate" is an average of averages, sanded smooth until it means very little about the specific house in front of you. And the "average price for the area" — the figure that ends up in newspaper headlines and dinner-party arguments — is a statistic doing a magic trick: it hides more than it shows.

This is the value pillar. Everything filed under it answers one question from different angles: what is this specific home actually worth, and how would you know? Not what someone hopes to get for it. What it is worth, read from evidence you can check yourself.

The average is the first thing to distrust

Start with the number people quote most and understand least: the area average.

Take one ordinary postcode district — CT11, in Ramsgate, Kent. The public record contains 3,250 sales there since January 2021. The median sale was £260,000. The mean was £290,482 — about 12% higher, dragged upward by a single £3.12m sale at the top of the range. Already, before we've looked at a single house, "the average price in CT11" can be £260k or £290k depending on which average someone chose, and nobody tells you which one they used.

Now zoom in one more level, to the street. Across the 106 streets in CT11 with at least ten recent sales, the median price ranges from £121,000 to £597,000 — a spread of nearly five to one, inside the same postcode district, the same schools, the same council, the same "£260k average". Wellington Crescent sits at the bottom; Foreland Heights, a cluster of detached new-builds, sits at the top. They are a short walk apart and £476,000 apart.

Go one level deeper still and even the street average dissolves. On Wellington Crescent the flats sold at a median of £107,500, while a terraced house on the very same crescent changed hands for over £400,000. Same address on an envelope; nearly four times the value. An "average for the street" would describe neither home.

None of this is a Ramsgate quirk — we picked CT11 because it's unremarkable. The lesson is general: every average you're shown is a bundle of very different homes, and the value you care about is the one it averaged away. We're extending this same analysis to the full national picture in a companion piece to follow; the takeaway won't change, and it's why the honest place to start valuing a home is a sale, not a summary.

A sold price is a fact; an estimate is a guess

His Majesty's Land Registry records what actually changed hands: 31.3 million transactions in England and Wales, going back to 1995. A sold price is not modelled, smoothed or optimistic. It is the number two people agreed on, in the real world, on a real day.

An automated valuation — the "estimate" a portal shows — is a model's opinion about a house it has never seen. It cannot know that the kitchen was gutted, that the lease has 79 years left, or that the garden backs onto a railway. It leans heavily on those same misleading averages. Treat it as a starting hypothesis, never a verdict.

The asking price is a third thing again: what the seller and their agent decided to ask. The gap between asking and achieved is where negotiations live. If you want to understand a specific home's value, work in this order — recent nearby sold prices first, then the specifics of this house, and only then the asking price as the number to argue with. (More on how to do that in how much is my house really worth? and sold prices vs asking prices.)

Two homes that look identical can be worth wildly different money

Once you've anchored on real sold prices, the job becomes explaining the differences — the reasons two lookalike homes on the same street carry different value. The biggest one is invisible in a photograph.

Tenure. A leasehold flat with a short lease is a fundamentally different asset from the same flat held freehold or with a long lease. As a lease falls toward the 80-year mark, the cost of extending it climbs steeply, and below it the maths gets punishing — a discount that can run to tens of thousands of pounds and is completely absent from the listing photos. If you're weighing a flat, freehold vs leasehold is not a technicality; it's often the single largest swing in value. (The whole-of-life cost of that difference is big enough that we gave it its own study.)

Then there is the tax you pay to buy at all. Stamp Duty Land Tax is banded, so a home priced just over a threshold can cost you meaningfully more in tax than one priced just under — a real part of what the home "costs" that never appears in the price field.

The running costs are part of the price

A cheaper house that costs £2,000 a year more to heat is not, over a decade, the cheaper house. Energy performance is value, deferred.

The EPC band is the public shorthand for this — a modelled estimate of how efficient the home is, and how much it costs to run. It is imperfect — a modelled figure that can misjudge a specific home, and something we take apart properly in a companion piece — but a band F or G is a genuine warning: expensive to run today, and increasingly expensive to sell or let tomorrow as minimum-standard rules tighten. When you compare two homes at the same price, the one two bands better is quietly worth more. Start with what an EPC band actually means.

The map decides part of the price too

Some of a home's value is set before anyone laid a brick — by where it sits.

A house in flood zone 3 carries a discount and an insurance premium that a visually identical house 200 metres uphill does not. A home in a conservation area may enjoy a "prettiness premium" but comes with real restrictions on what you can change — and those two effects don't cancel out neatly. And the value of a home is partly the value of what you're allowed to do to it: whether there's realistic scope to extend under permitted development can be worth as much as an extra bedroom.

These are the things a portal estimate can't price because it doesn't look at the map. They are also, conveniently, all in the public record — which is the whole point of being able to check them.

How to value a specific home, honestly

Put it together and there's a repeatable method, none of which requires trusting a single headline number:

1. Pull the recent sold prices on and around the street — the median, not the mean, and ideally split by property type.
2. Adjust for the specifics the sold prices can't see: tenure and lease length, EPC band and running costs, condition, and what the map says about flood, conservation and planning scope.
3. Only then look at the asking price, and treat the difference as your negotiating room.

That's the sequence AskHouse runs for any UK address — sold prices, tenure, energy, flood, planning and building footprint, joined against a specific home rather than an area average. You can ask about any property directly, or read how it's assembled in Explorer. If you want to go deeper on a single theme, the answers library and the property glossary break each of these factors down on its own page. Two companion reads go further on the sticker-price problem: the postcode average is a lie, on how far two streets in one postcode can diverge, and why you research a £300 phone harder than a £300,000 house.

The information a seller and their agent hold about a home has always existed. For most of the history of buying property, the buyer simply wasn't shown it. The value pillar exists to close that gap — to make "what is it actually worth?" a question you answer with the record, not the sticker.

Methodology and limitations

Source and scope. HM Land Registry Price Paid Data, analysed on 18 July 2026.

• National scale: 31,299,627 transactions, covering January 1995 to March 2026.
• CT11 headline: 3,250 sales since 2021; mean £290,482, median £260,000, top sale £3,120,000.
• Street-level spread: among the 106 streets with at least ten sales, median prices range from £121,000 on Wellington Crescent to £597,000 on Foreland Heights.
• Within-street variance: Wellington Crescent flats have a median of £107,500 across 52 sales, while its terraced houses sold at £401,000–£418,000; Foreland Heights is detached new-build stock at £555,000–£625,000.

Limitations, stated plainly. Price Paid Data covers England and Wales only — there are no Scotland or Northern Ireland sales in these figures. It includes "Other" property-type records (commercial units, part-shares, land transfers) that inflate means, which is exactly why we lead with the median and split by property type. The CT11 numbers are a deliberately small, illustrative slice of one district since January 2021 — they demonstrate how averages hide variance; they are not a national market claim. Land Registry data lags real completions by roughly one to two months, hence the 2026-03-31 ceiling. Street and postcode text is used as recorded, with no address-matching applied for these particular cuts.