Site area times density is how people get the number wrong. Three times over.

The honest answer is a range, and the range is wide enough that the question has to be narrowed before it means anything. An acre is 0.405 hectares, which under the published net developable scale sits in the 90% band — so about 0.36 hectares of buildable land after roads, footpaths, drainage and incidental open space. Apply typical measured densities to that and you get roughly 7 detached houses, 13 semi-detached, 24 terraced houses, 40 low-rise flats, or 72 in a mid-rise block. Those densities are not invented: post-war semi-detached estates measure at about 35 dwellings per hectare, Victorian terraces at 63, three-storey townhouses at 72, and low-rise apartment schemes at around 110. But the arithmetic is the easy part and it is not what decides the answer. What decides it is the local plan, the character of the surrounding streets, access and highways, trees, flood risk, ecology, daylight, overlooking, parking, and whether the council has a five-year land supply. The same acre in two different wards supports genuinely different schemes. There is also a threshold worth knowing: at 0.5 hectares — about 1.24 acres — a site becomes major development regardless of how few homes are on it, which brings affordable housing into play. If you are anywhere near that line, it is worth knowing which side you are on before you draw anything.

Because a published build rate answers a narrower question than almost anyone realises. The standard measure is cost per square metre of gross internal floor area, and it covers the building itself including preliminaries and the contractor's overheads and profit. It excludes, in terms, external works and site infrastructure, contingency, professional fees, finance charges and VAT. So the roads, the drainage, the services, the landscaping, the architect, the engineers, the surveys, the planning application, the contingency and the interest are all outside the number you were given. Add them back and the all-in figure typically lands 45% to 60% above the headline rate. On top of that sit the things that are not construction at all: land, stamp duty, affordable housing, the community infrastructure levy where one applies, marketing and sales, and the developer's profit — which planning guidance itself puts at 15% to 20% of gross development value. There is a second reason the number has moved, and it is recent. The house builders' federation calculates that policy and cost changes since 2020 have added about £76,000 per home — materials, labour, the Future Homes Standard, wider building regulations, biodiversity net gain, the building safety levy, landfill tax — with a further £22,000 on apartments for building safety compliance. Against an average new home value of £365,000 that is more than a fifth of the value of the house. A spreadsheet built before 2022 is describing a different market.

Net developable area is the part of a site you can actually build on, once roads, footpaths, public open space, drainage attenuation, landscaping and buffers have come out. The professional standard defines it as the extent of the site on which buildings and their ancillary space can be built, excluding public realm, open space, sustainable drainage systems and roads not strictly necessary for servicing the properties — but it deliberately gives no percentage. The percentages live in local authority housing land assessments, and they converge on a banded scale: 100% below 0.4 hectares, 90% from 0.4 to 2 hectares, 75% from 2 to 10, and 50% above 10. So no, 60% is not right as a general figure — it is a large-site number. Applying it to a one-hectare site understates the capacity by roughly a third, and that error compounds through every subsequent calculation. The banded scale is also better evidenced than most rules of thumb: one authority tested it against fifty completed major developments and measured 98.0% for sites up to 0.4 hectares, 91.4% for 0.4 to 2 hectares and 78.8% above 2 hectares — slightly better than the assumptions, not worse. One caveat, raised by an authority publishing its methodology in 2025: no council has yet revised these percentages to reflect biodiversity net gain and nutrient neutrality, both of which can take real land. On a constrained site the published figures may now read a little high.

Two answers, and the first one changed very recently. On 6 August 2026 a new exemption came into force for sites of 0.2 hectares or less within the red line boundary, unless on-site priority habitat is negatively affected. Government expects it to take roughly half of residential planning permissions out of the regime entirely. It is an area test, not a major-or-minor test — the 2024 commencement dates split by major and minor, but applicability never did, and keying it off dwelling numbers is a common and expensive misunderstanding. The exemption applies only to applications submitted on or after 6 August 2026; earlier ones stay under the old rules. The self-build and custom build exemption was removed at the same time, and temporary permissions of five years or less became exempt. For sites above the threshold, the second answer is less satisfying but more useful: nobody publishes a land-take percentage, and you should be wary of anyone who quotes one. The land take is driven by the baseline habitat condition, not by site size. A species-poor arable field or a hardstanding yard can frequently reach 10% net gain within the landscaping and drainage the scheme needs anyway, at close to no additional land. A site with existing good-condition grassland, mature hedgerow or woodland can be impossible to satisfy on site at any density. What has changed helpfully for smaller schemes is that off-site gains now rank equally with on-site creation for minor development, which converts a land problem into a cost one. Off-site units traded at roughly £24,000 to £29,000 for grassland and hedgerow in early 2026; statutory credits, which are a deliberate last resort, run from £42,000 to £650,000 and you must buy two for every one unit needed.

Because eighteen metres is a regulatory cliff rather than a gradient, and three separate regimes land on the same line. First, and most immediately, second staircases become mandatory for buildings with a storey at or above 18 metres, for new applications from 30 September 2026. The amendment was confirmed on 29 March 2024 after being consulted on at 30 metres, and transitional cover applies only where a building control application went in before the deadline and work either has sufficiently progressed or does so within eighteen months — with 'sufficiently progressed' meaning foundations poured or permanent piling started. The government's own impact assessment puts the additional capital cost at £0.5m to £1.3m per building in the 18 to 30 metre band, rising to £2.3m–£11.7m above 50 metres, and calculates a net present value of minus £2.67 billion against monetised benefits of £9.1 million. Second, it takes floor area: roughly 7% of the saleable area, so the revenue falls at the same moment the cost rises. One analysis prices the sixth storey above a fifth at £630,000, against £14,400 for an ordinary additional floor. Third, at 18 metres or 7 storeys a building becomes a higher-risk building and enters the gateway regime. In the first quarter of 2026 the median time from a Gateway 2 application to approval was 34 weeks, with a 67% approval rate. Performance has improved markedly through 2026, but at 7% finance a scheme sitting still for six months is losing real money. The practical implication is that five storeys and six storeys are different businesses, and a scheme that only just crosses the line is usually worth redesigning to sit under it.

Ten or more dwellings, or a site area of 0.5 hectares or more. The second limb is the one that catches people, because it applies regardless of how few homes you are proposing — a low-density scheme of six large houses on 0.6 hectares is major development. The definition comes from the development management procedure order and is repeated in national planning policy. What it triggers is affordable housing: national policy says provision should not be sought on non-major development, other than in designated rural areas where a lower threshold of five units or fewer may apply. Designated rural areas means National Parks, National Landscapes and areas designated as rural under the Housing Act 1985. There is no national affordable housing percentage — that is set locally and varies widely, so treat any figure quoted at you as a local plan question rather than a national one. The only nationally specified figure is the Green Belt 'golden rules' cap of 50%. Three things it is worth knowing major status does not control. Biodiversity net gain now turns on the 0.2 hectare area test, not on major status. The community infrastructure levy is not tied to major development at all — it is charged per square metre and can be levied on a single house of any size. And environmental impact screening runs on entirely different thresholds: more than 150 dwellings, or a site over five hectares. Also worth checking on brownfield: the vacant building credit deducts the floorspace of an existing vacant building from the affordable housing calculation, and it can materially reduce the requirement.

It is a calculation rather than a principle, and the calculation is frequently closer than landowners expect. Selling with consent almost always achieves a higher headline price, because the buyer is purchasing a known quantity rather than a hope. Against that: promoting a site through planning takes twelve to eighteen months at best, costs real money in consultants, surveys and application fees, and carries the risk of a refusal that is then a matter of public record and damages the site for years. There are four structures rather than two, and the middle ones are where most small sites land. An unconditional sale is fastest and cheapest and leaves the uplift with the buyer. A sale subject to planning keeps you as owner while the buyer promotes it at their cost and risk, usually with a long stop date. An option agreement gives a developer the right to buy at a discount to market value, typically 10–20%, once consent is obtained. A promotion agreement aligns you with a promoter who takes a percentage of the eventual sale proceeds and is incentivised to maximise price rather than minimise it — often better for a landowner than an option, and less often offered. The thing that most commonly costs landowners money is not the choice between these but overage: the clause dealing with what happens if the buyer obtains a better consent than the one assumed. Get that drafted by a solicitor who does this specifically. One further thought worth weighing before any of it: a pre-application enquiry with the council costs a few hundred to a couple of thousand pounds and will tell you more about what the site will actually support than any consultant can infer from a plan.

Because it does not know what the finished homes would sell for, and without that the land value cannot be calculated in any honest way. Land is valued residually: you take the gross development value of the completed scheme, subtract every cost of building it — construction, externals, abnormals, fees, contingency, finance, policy costs, marketing, and the developer's profit — and whatever remains is what somebody can afford to pay for the land. Change the sales value by 5% and the residual can move by a great deal more, because it is the difference between two large numbers. That sensitivity is precisely why the number has to be built on real comparable evidence for that particular street rather than a regional average or a portal asking price. It is also why we would rather show you a cost you can check line by line than a land value we have effectively invented. There is a harder point behind this. On constrained brownfield sites, once abnormals and the current policy costs are properly included, the residual land value is frequently negative — meaning the scheme cannot pay for the land at all, whatever the owner has been told it is worth. That is genuinely useful information, and it is much cheaper to discover before a planning application than after one. Our £650 report does the value side from completed transactions, and the £1,450 appraisal runs both sides together with the sensitivities. If the answer is that the scheme does not work, we will say so.

The most expensive mistake on a small site is buying professional advice in the wrong sequence — commissioning a full set of drawings before anyone has asked the council what it thinks, or a cost plan for a scheme that will never be permitted. A sensible order: first, a pre-application enquiry. It costs a few hundred to a couple of thousand pounds depending on the authority and the scale, it produces a written officer view, and it will tell you more about real capacity than any amount of desk work. Second, a planning consultant, if the pre-app is encouraging or ambiguous — for the local plan position, the strategy and the application itself. Third, an architect for a capacity study and design; note that a capacity study is a different and much cheaper thing than a planning application drawing set. Fourth, a quantity surveyor for a real cost plan, once there is something to price. And us for the value side — what the finished homes are worth on that street, what the land is therefore worth, and the sale when it comes. We are property agents. We are not planning consultants, architects, engineers or quantity surveyors, and we will not pretend to be any of them. What we will do, at no charge, is tell you which of them you need first, and say when the honest answer is that you do not need us yet. If a site is genuinely marginal, the cheapest professional advice you can buy is the advice that stops you spending more.