How do I price a building job without losing money?

The principle everything below serves: separate cost from judgement. Cost is what the job will consume, measurable, checkable, improvable. Judgement is what you add for risk and profit, a decision. Prices go wrong in both layers, but you can only learn from a loss if you can see which layer failed, and a single gut-feel number can never tell you.

Layer one: the cost, built up honestly

• Measured work: quantities from the drawings, taken off systematically, priced at rates that reflect your outputs, not a price book’s optimism. Where the design is not finished, price what is drawn and handle the rest as honest allowances, not silent hope.
• Labour at true cost. The classic hole: pricing a man at his wage. True cost carries employer’s NI, holidays, downtime, the van and the phone, and for most firms lands 30 to 50 per cent above the wage. Price at wages and every labour hour sold loses money by design.
• Materials at current quotes with realistic waste, and in a moving market, a decision about who carries the movement.
• Subcontract packages on real quotes, not last year’s rates, with their attendances (your scaffold, your forklift, your clearing up) priced by you, because they are not in the sub’s number.
• Preliminaries, priced as their own section: supervision, welfare, scaffold, skips, insurance, the weekly cost of simply running the site for the whole programme. Prelims have their own page, because leaving them smeared invisibly across rates is the single most common way small jobs lose money.

Layer two: the judgement, added visibly

On top of cost go two numbers, and they are different decisions kept deliberately apart. Risk: what could this job do to me, ground, weather exposure, an unknown client, a hard programme, priced per job, not as a habitual percentage. Margin: what the firm needs to exist plus what the market bears, the markup question has its own page. Keeping them apart matters at the end: a job that ate its risk allowance and kept its margin was priced well and unlucky; a job with no risk left and no margin either was priced wrong, and next quarter’s prices should know that.

The two pricing sins with pages of their own

Winning on scope vagueness: a low number over an undefined scope is not a price, it is a deferred argument, and the was-it-included fight is where it matures. The quote’s scope discipline is its own page. Bidding everything: some jobs should not be priced at all, wrong client, wrong risk, wrong month for your cash, and saying no has its own page too. Nothing here changes for a homeowner client: pricing is client-blind arithmetic, though the consumer quote carries extra drafting duties this series covers elsewhere.

What to do this week

1. Work out your true labour cost per hour, once, properly, with your accountant’s numbers. It re-prices everything you sell.

2. Restructure your next estimate into the layers: measured work, packages, prelims, risk, margin, each visible.

3. Dig out the last finished job’s estimate and final account and compare them layer by layer, one evening, and let what you find set the next quote’s numbers.

Where the information stops

Your outputs, your overhead recovery and your market are yours: no page knows whether your bricklaying gang really lays what your rates assume, and the honest sources of that number are your own job records. Where this page meets its limit, a quantity surveyor pricing or checking a big tender is buying exactly this structure with local knowledge attached.