UK building contractors: the profit belongs to whoever owns the land

The 929 UK building companies that publish a full profit-and-loss put up everything from office towers to housing estates, and on paper they book a little over £100bn of combined turnover. But the map splits along one line: who owns the land under the building. Wates Construction, the Wates group’s main contracting arm, turned £2.24bn of work into £1.3M of pre-tax profit — less than 6p kept from every £100 built; even stripping out a one-off £23M pension charge, the underlying figure is about £24M, barely 1%. Bellway Homes, a housebuilder that put up £2.77bn of homes, kept £226.6M — an 8% margin — because it builds on land it owns and sells the finished houses. Same trade on site; opposite businesses in the accounts. Figures are approximate — verify against a company’s own accounts before relying on any single number.

One trade, two profit models — never compare them

Before any number here means anything: a contractor builds someone else’s project for a contracted price. It carries the delivery risk — weather, ground conditions, materials inflation, a client that pays late — and keeps whatever sliver survives. Two to four per cent is a normal year; a single bad job can erase several good ones. A housebuilder is a principal: it buys the land, builds at its own pace and sells the homes, so its margin carries the land’s gain as well as the build. Even in a soft housing market the builders here print 8–12% while the best contractors fight for 4%.

Both models file under “building construction”, so both appear in this map — the volume housebuilders are labelled in the tables and kept out of every contractor-margin comparison. The land-led developers who don’t build at all have their own map.

Two more caveats that change how you read everything below. First, survivorship: contracting is Britain’s most insolvency-prone trade — ISG, a £2bn-a-year contractor, collapsed in 2024 — and a failed contractor exits the numbers entirely, so this map only ever shows the ones still standing. Second, group stacking: holding companies and their trading arms both publish accounts, so the biggest groups appear more than once; we read the operating entity and drop the duplicates.

The giants

(Laing O’Rourke’s holding and delivery entities also file at similar scale and are dropped as duplicates; BDW Trading and Redrow sit inside the Barratt Redrow group.)

*All PBT figures are statutory; several carry one-offs. Wates’ £1.3M is after a £22.9M one-off pension charge (underlying ~£24M, still about 1%). Laing O’Rourke’s −£6.9M is after £31.7M of exceptionals (£24.8M before them). Bellway’s £226.6M is after a £51.8M legacy building-safety charge (underlying £278.4M). Barratt Redrow’s £273.7M carries merger integration and building-safety charges (adjusted £488.3M).

Read the two models side by side. The housebuilders — BDW (8.6%), Bellway (8.2%), Redrow (11.3%) — all cleared eight points of margin in a housing market everyone calls difficult. (Bloor Holdings prints 11.4% too, but it is the family group behind both Bloor Homes and Triumph Motorcycles — roughly a third of its turnover is a global motorcycle maker — so its margin blends housebuilding with manufacturing and sits outside the pure comparison.) The contractors: Wates Construction at 0.06% — about 1% even before its one-off pension charge — Laing O’Rourke at a −£6.9M statutory loss (£24.8M before exceptionals), Galliford Try at 2.3%. Morgan Sindall’s construction and infrastructure arm at 4.1% is about as good as UK contracting gets at national scale — and it still keeps half what the weakest housebuilder does.

Two rows need care. Barratt Redrow’s merger year — the +34% turnover and +20% headcount are the Redrow combination arriving — left statutory profit at £273.7M, a ~4.9% margin dragged well below its own trading arm’s clean 8.6% by integration and building-safety charges; BDW made £389M in the same period. And Balfour Beatty Group is an intermediate group company — 124 employees on £4.66bn of turnover tells you the site workforce is elsewhere — so its 4.2% is a group-level figure, not a clean contracting-margin read. Meanwhile watch Wates: turnover +11%, headcount +21%, profit £1.3M — building more, hiring more, keeping almost nothing. That is the contracting model in one row.

The shape of the market

Scale sorts this market brutally. Below £1M of turnover only 37% of firms make money — a graveyard of marginal operators. From £5M upward, profitability jumps to and stays above 80%: once a contractor has the balance sheet to pick jobs rather than take whatever pays this month’s wages, it survives. The unusually thin £1–5M band is the squeeze in the middle — big enough to carry overhead, too small to absorb one bad contract.

The best-run contractors

The table below takes profitable mid-market operators (£5–100M) clearing at least a 5% margin — a bar most contractors never reach in their best year, so this is the exceptional tier, not the norm. Housebuilders that passed the same screen (Bargate Homes, Rowland Homes, Legal & General’s affordable-homes vehicles) are excluded — their margins are land economics, not contracting. So is Integrated Accommodation Services, a long-term facilities concession whose 26% margin is availability-payment economics, not building work.

The pattern is regional and repeat-work: Robertson and Ashleigh in Scotland, Geda in Northern Ireland, R.G. Carter in East Anglia, John Turner in the North West — established names that win the same public-sector and framework clients year after year, priced with discipline rather than bought with margin. Vascroft at 23.8% and Ashleigh at 17.3% are extraordinary numbers for this trade — margins that high usually mean a premium niche or an unusually kind job mix. Vascroft’s margin was close to 17% the year before too, which starts to look structural; Ashleigh’s prior year was nearer 7%, so we’d want to see it repeated. The quieter story is what the 5% bar leaves out: hundreds of perfectly well-run contractors earning their normal 2–4%, invisible here precisely because their margins are ordinary.

Growth, read with care

The top of the growth table is almost entirely an accounting artefact. A builder recognises revenue when a scheme completes, so a firm that sold nothing last year and finished a site this year prints a four- or five-digit growth number: Bugler Homes (+34,037%), Connolly Homes (+6,198%) and Hillreed Homes (+4,028%) are housebuilders’ sales landing in lumps, not businesses growing at those rates. Nicholas King Homes (+211%, −£4.2M) is revenue bought at a loss.

The genuine signals are the ones backed by hiring. Wates Residential Construction (+875% turnover, +367% staff) is a young vehicle ramping hard into affordable-housing work at an 8.5% margin. Huber Parking (+1,099%, +116% staff) is a specialist car-park builder scaling with real people at a contractor’s 3.8%. Both are hiring ahead of workload — the one growth pattern in contracting worth believing.

The most fragmented top table we’ve mapped

Building is structurally fragmented. The top 5 firms hold just 18.9% of the mapped turnover and the top 10 hold 28.7% — far looser than road freight (top 5: 38%) or temp staffing (31%). Construction is local, every project is bespoke, and the national tier subcontracts most of the physical work anyway, so scale buys procurement clout but no lasting share. Group entities appear more than once near the top of the curve, so read the concentration as directional — the true market is even flatter than it looks.

Ownership and vintage: age is the moat

Ownership splits almost exactly down the middle — 447 corporate-owned against 439 individual-owned — one of the most founder-and-family-held markets we’ve mapped, which fits a trade where the surname over the door is often the brand. About 12% (107 firms) carry a Holdings/Group/Bidco-style name, the structural fingerprint of a buyout or a planned exit — private equity is present but hasn’t rolled this market up the way it has care or staffing, perhaps because 2% margins and lumpy cash flow make poor leverage fodder.

The vintage profile explains why. Pre-1990 is the largest cohort (272 firms) — in most markets age is not a moat, but in contracting it is close to one: clients committing years to a project pick the firm that has visibly survived several cycles, and frameworks are won on track record. The 2021+ cohort is tiny (37 firms) — nobody storms into a 2%-margin, working-capital-hungry trade.

What the map shows

1. The land line is the whole story. Firms that own what they build keep 8–12% even in a weak housing market; firms that build for clients keep 0–4%. Wates turned £2.24bn into £1.3M (about £24M before a one-off pension charge — still barely 1%); Bellway turned £2.77bn into £226.6M.
2. Elite contracting tops out around 4% at national scale. Morgan Sindall’s arm at 4.1% is the benchmark; the regional best-run tier (Robertson, R.G. Carter, John Turner, Geda) earns 6–8% on repeat framework work.
3. Scale is survival. Below £1M only 37% make money; above £5M, more than 80% do. The balance sheet to ride out one bad job is the real barrier to entry.
4. Nobody owns UK building. Top 5 at 18.9% is the flattest concentration curve we’ve mapped — construction is local, bespoke and subcontracted, so share doesn’t compound.
5. Growth numbers here are mostly revenue-recognition noise. Completions land in lumps; trust only the growers hiring ahead of workload, like Wates Residential (+367% staff) and Huber Parking (+116%).
6. Age is the credential. The biggest cohort predates 1990, ownership is half founder-and-family, and almost no one has entered since 2021 — the map only ever shows the survivors of a trade that regularly kills its participants.

Methodology and caveats

This covers the 929 UK building companies — commercial and residential builders, from national contractors to regional firms — that publish a full profit-and-loss; several thousand more are too small to report figures and don’t appear. Volume housebuilders file alongside contractors and are labelled and separated in every margin comparison; land-led developers have their own report. Group holding companies and their trading arms can both appear, so top-end totals carry some double counting; insolvent contractors exit the numbers entirely, so the map has a survivor bias; large one-off losses may reflect mergers, impairments or building-safety provisions rather than ordinary trading. Figures are approximate and business-type labels are directional — verify any specific figure against a company’s own accounts before relying on it. This is analysis, not financial advice.