How to price a construction job
without guessing.
A method rather than a feeling: measure, build the rate, add the site costs that carry the job, recover overhead, then decide margin. The order matters.
Published ·Updated ·6 min read·Written by Darshan Parmar, Founder
Six steps in order: measure the work, build a rate for each item from labour, plant and materials, price the preliminaries that carry the job, add overhead recovery, decide the margin, then test the total against what you know about the client and the risk. The two steps firms leave out are both near the end.
Pricing is a sequence, and most of it is not controversial
Ask five contractors how they price and you get five answers, but the disagreement is narrower than it sounds. Almost everyone measures, and almost everyone builds rates. Where firms genuinely diverge is what happens after the rates are built, and that is where the money is won and lost.
So this article moves quickly through the first three steps, which are craft, and slows down on the last three, which are business decisions that are frequently not made at all.
One. Measure the work
Quantities from the drawings and specification, in the units you will be paid in. The discipline that matters here is recording where each quantity came from, because in four months somebody will ask, and because a measure with a documented origin survives a remeasurement argument that a number in a spreadsheet does not.
Where the drawings do not tell you something, that is not a gap to be estimated over. It is either a question to the client, a provisional sum, or a stated exclusion. Absorbing an unknown silently at tender is how a job is lost before it starts.
Two. Build the rate
For each item, from first principles rather than from last year's number.
- Labour. The gang, the output per unit, and the all-in hourly cost of the people in it. All-in means what they cost you, not what they are paid: employer National Insurance, pension, holiday cover, and the on-costs of putting somebody on a site. That figure is usually higher than people assume, and the savings calculator puts employer National Insurance and the pension on it.
- Plant. Hire or owned, with standing time included rather than only working time. Plant that is on the job and idle is still on the bill, which is the subject of off-hire discipline generally.
- Materials. Quoted, with waste, and with a validity date on the supplier quote that you actually check against your own.
Rates built this way survive scrutiny and can be adjusted when something changes. Rates carried forward from a previous job carry that job's assumptions, including its mistakes.
Three. Price the preliminaries
Everything that makes the job possible and appears on no drawing: supervision, welfare, temporary services, access and scaffolding, insurances, security, cleaning and the time either side of the work itself.
This is the first of the two steps firms habitually get wrong, and the error is usually structural rather than careless: prelims are priced as a percentage of the measured work when most of them are time-related. Get it right and a delay costs you the delay. Get it wrong and every week the programme slips comes straight out of margin. It has an article of its own on pricing preliminaries, because it is worth the space.
Four. Recover the overhead
The second habitual omission, and the one with the strangest failure mode: the job makes money and the company does not.
Your business has costs that no job carries on its own. The office, the estimator, the accounts function, the vehicles, the software, the insurances that are not project-specific. Those are paid for out of the work, which means every job has to carry a share, and the share has to be calculated rather than hoped for.
The arithmetic is not difficult. Take last year's fixed overhead, divide it by the volume of work you expect to turn over, and you have the percentage each job needs to contribute. What makes it uncomfortable is that the number is usually larger than the margin people think they are making, which means a firm winning work at cost plus a thin margin has been running its office at a loss and funding it from somewhere else.
A job priced at cost plus ten per cent margin, on a business needing twelve per cent to cover overhead, loses money while looking profitable on the sheet.
Five. Decide the margin
Decide, rather than discover. Margin is a number you choose before the quote goes out, based on the risk you are taking and how much you want the work, and then defend.
Three things that legitimately move it. The risk in the job: an unfamiliar client, an unclear scope, a tight programme, or work you have not done before. The payment position: a client who pays at ninety days is charging you for the privilege, and that should be in the number. And how busy you are, which is a genuine commercial input rather than a shameful one.
Two things that should not move it, and constantly do. What you think the competition will bid, which is a guess about somebody else's cost base. And wanting the work, which is a feeling rather than an input, and the one that produces the jobs that were never going to make money.
Worth being precise here about margin and markup, because they are not the same number and the confusion is expensive. Markup is added to cost; margin is the share of the selling price that is profit. A twenty per cent markup produces a margin of about sixteen and a half per cent, and pricing as though they were interchangeable quietly loses the difference on every job.
Six. Test the total
Before it goes out. Three questions, and they take five minutes.
What is this job's cash profile, not just its profit? Retention, CIS and payment terms can make a profitable job unaffordable, and that is worked through in the gap between doing the work and having the money. What happens to the total if the two riskiest items are twenty per cent worse than priced? And is this a quotation or an estimate, which decides who carries what you did not price, and is set out in quote or estimate, and which one binds you.
In Unibuild, Advanced Estimating holds the item database that step two draws on. Each item carries its cost, charge rate, unit and fit time, and kits are built once with fit times and mark-up inside. The priced version then stays connected to what happens next. Where a project was quoted through the platform, the works schedule draws on the priced sections of that quotation with their tendered values. The job is measured and applied against the rates it was won on, not against a spreadsheet somebody has since edited. Labour arrives from clock-ins at dated cost rates, so the actual against the priced is arithmetic rather than an exercise. Orders become committed cost from the moment they are issued. Because Unibuild is built as bespoke software for each contractor, the estimate follows the firm's own build-up rather than the firm pricing to suit the software.
Where to start, on Monday
Work out your overhead recovery percentage. Last year's fixed overhead, divided by last year's turnover. It takes ten minutes with the accounts open, most firms have never calculated it, and it is the number that tells you whether the margin you have been adding is margin at all.
Then take the last job you priced and check whether that percentage was in it. If it was not, you now know something useful about every quote you have sent this year.
The follow-up questions.
Every term used here is defined in the construction glossary.
How do you price a construction job?
What should the labour rate in a construction price include?
What is overhead recovery and how do I calculate it?
What is the difference between margin and markup?
What should not influence the margin on a job?
Should the price change if the client is difficult?
How long should it take to price a job?
How do I price work I have never done before?
Should I show a client the breakdown of my price?
What should I do if my price comes out higher than the competition?
Measure against what you priced.
A build-up is a prediction. It teaches you nothing unless the job's real cost comes back to sit alongside it.
- Thirty minutes, weekdays, from tomorrow.
- Nothing to prepare. Bring a job number and we mock that job up.
- You drive it. There is no slide deck.
- You keep what you saw as a 14-day trial. No card.
More from Insights.
How to tell whether you are underpricingYou will not find it in the jobs that went wrong, but in the ones that went fine and returned less than they should. A test you can run in an afternoon.Read it →
Preliminaries, the cost most firms forget to priceEverything that makes the job possible and appears on no drawing: supervision, welfare, access, plant standing, insurance, and the time before and after. Why it gets under-priced and how to stop.Read it →
Quote or estimate, and which one binds youOne becomes a fixed price the moment it is accepted. The other is an informed guess. The word at the top of the page is not what decides which you sent.Read it →All 11 articles on commercial.