Out of a million dollars, five cents is yours.

Line up every claim on that million dollars, in order, and watch it drain.

The waterfall to what survives

Direct cost goes first and goes biggest: $710,000 of steel, labour and plant — the work you can point at. Then the prelims, $85,200 of site that belongs to no item. Then escalation, then the contingency you may or may not spend, then the $74,051 slice of head office that this job has to carry, and finally the $3,225 it cost you to finance the cash hole from Week 16.

What survives all of that is about $48,163. On a million-dollar contract, under five cents in every dollar sticks to the company. That is not a bad job or a mismanaged one — that is a normal one. Construction runs on some of the thinnest margins in any industry, and everyone new to it is shocked by the number, because a million-dollar project sounds like a fortune and leaves behind the price of a modest car.

Two words that are not the same word

Now the distinction that separates people who understand this from people who merely work in it. There are two ways to measure what a project makes, and confusing them is one of the most expensive mistakes a contractor can make.

Contribution is revenue minus the variable cost — the cost that only exists because you took the job. On this project that is the million dollars minus the $827,008 of direct work, prelims and escalation: $172,992. That is what the job “contributes” towards two things that have to be paid regardless — the head office, and the profit. It is the same $172,992 that was the gap in Week 9 and the profit line in Week 16, now wearing its proper name.

Margin is what is left after head office takes its share too — the net profit over the revenue. Here that is $48,163 on a million, a net margin of about 4.8%.

Contribution asks one question: is this job worth doing at all? Margin asks a different one: did the company actually make money? They are not interchangeable, and the gap between them is exactly the head office cost that Week 20 warned you about.

When to take a job that “loses” money

Here is where the distinction earns its keep. Suppose the order book is thin and a job comes up at $880,000 — below your normal price. Load the full head office onto it and it looks like a loss: it recovers its variable cost and its overhead share and comes out around $24,000 in the red. On margin, you should walk away.

But the head office is already there. The managing director is on salary whether you win this job or not; that cost is committed. The only question this job actually changes is whether its variable cost gets covered with something left over. At $880,000 against $827,008 of variable cost, it contributes $52,992 — and $52,992 towards a head office you are paying for anyway is $52,992 better than an empty yard and idle gangs.

So in a lean spell, with capacity you would otherwise waste, a job that “loses” money on margin can be the right job to take — because contribution, not margin, is the correct test for filling spare capacity. But — and this is Week 20's warning in its final form — do it as a habit, price every job this way, and you never recover your head office at all. You will have a full order book, every site busy, every job contributing, and a company losing money all year. Contribution is the test for the marginal job. Margin is the test for the business.

Why a thin margin forgives nothing

And now, at the very end, the whole track snaps into focus.

Seventy-one cents in every dollar of this contract is direct cost — the steel and the gang and the plant you have spent twenty-one weeks learning to measure and control. About five cents is the profit you keep. Look at what that ratio means.

A five percent overrun on the direct cost — a gang running slightly slow, steel bought slightly dear, the sort of slippage that happens on every job that is not watched — is $35,500. That single, ordinary overrun wipes out seventy-one percent of the entire net profit. And Week 6's overrun, the $61,868 we traced to a slow gang and a dear steel order, is larger than the whole profit on the job. One uncontrolled quarter and the margin is not thin, it is gone.

This is why the whole track exists. On a five percent margin, cost control is not administration, not housekeeping, not a report someone files. It is the difference between a company that survives the year and one that does not. Every unit rate in Week 2, every variance in Week 6, every productivity factor in Week 15 was, in the end, about protecting a five-cent margin from a seventy-one-cent cost. That is the job.

Practical insight

Find your project's net margin — the real one, after head office and financing, not the gross markup in the tender. If it is around five percent, as most are, then work out what a five percent overrun on your direct cost would do to it. The number will frighten you, and it should.

Then, the next time someone treats cost control as bureaucracy — a form to fill, a report nobody reads — show them that arithmetic. A thin margin does not mean the money does not matter. It means every dollar matters more, because there are so few of them between you and a loss.

Key takeaways

✔ After direct cost, prelims, escalation, contingency, head office and financing, about $48,163 of a million-dollar contract survives — under 5 cents on the dollar.
✔ Contribution is revenue minus variable cost: $172,992, what the job hands to head office and profit together.
✔ Margin is net profit over revenue: about 4.8%, what is left after head office too.
✔ Contribution asks “is this job worth doing?” Margin asks “did the company make money?” They are not the same question.
✔ In a lean spell, a job that loses on margin ($880k, −$24k) can be worth taking if it contributes ($52,992) to a head office you pay for anyway.
✔ But price every job on contribution and you never recover head office — a full order book that loses money all year (Week 20's warning).
✔ 71 cents in every dollar is direct cost; ~5 is profit. A 5% overrun on direct cost erases 71% of the net profit.
✔ Week 6's $61,868 overrun alone wipes out the job. On a 5% margin, cost control is survival — the reason the whole track exists.

What is coming next

We have followed one job from an empty drawing to its final five-cent margin. But we have quietly assumed, all the way through, that the job makes money at all.

Some do not. And a project heading for a loss does something strange and specific to the numbers — the profit you booked in the good months has to be given back, all at once, the moment you can see the end. It is the hardest, most honest calculation in the whole of cost management, and it is where a commercial planner earns their reputation or loses it.

Next week, the climax of the track: revenue recognition and the expected loss — what happens to the money when the job is going to lose.