Your Construction Business Is One Big Math Problem

A few years ago I ran an experiment that sent me down this path.

I pulled five jobs from our completed work. Not random jobs but our “best” jobs. The ones we had pointed to when someone asked how the year was going. Good clients, clean scopes, crews that knew what they were doing. Jobs we were proud of. Jobs that made money. Or so we thought.

Then I asked a simple question: what if we ran these five jobs on repeat, every year, indefinitely?

The answer was that we would lose money.

Not dramatically. Not in a way that would show up immediately. But steadily, year over year, the company would fall behind. And the reason wasn’t the jobs. The jobs were fine. The reason was that our margin, the number we added to our cost of goods to arrive at a price, did not cover what it actually cost to operate the company.

There was indirect overhead that wasn’t being captured anywhere. Not in the burden labor rate. Not in the OH&P we were applying to each job. It was just missing from the equation.

Instant slip before work started.

Most construction companies set their margin the way we did. They look at what the market will bear. They look at what they charged last year. They look at what a competitor might charge or what a client might tolerate. They land on a number that feels reasonable, apply it to their costs, and call that a price.

That is not a math problem. That is a guess with a percentage attached.

The actual math works differently. The overhead and profit you add to a project — the OH&P, or the plus in cost-plus work — is not an arbitrary margin. It represents the total gross profit your company needs to produce. Once you know that number, you can divide it by the days in one year.

OH&P ÷ 365 = gross profit per day.

That is the number that tells you whether the work you are doing is moving fast enough, financially, to support the company you are running. Not whether a job feels profitable. Not whether the client seemed happy. Whether the company is producing what it needs to produce to cover its costs and leave the owner with the result the business is supposed to deliver.

If the OH&P number is wrong or if something real is missing from it every estimate built on top of it is wrong too. The math looks like math. It has a percentage attached. But it is not solving the right problem.

There is another version of this problem that stalks small operators. One that stalked me when I ran my shop in NY.

The owner still thinks of themselves as a direct cost.

It makes sense on the surface. They came up as tradespeople. They are in the field sometimes. So they put their labor in cost of goods sold, apply a margin on top, and assume the equation is complete.

It is not complete.

Even if you occasionally swing a hammer, you are the general manager of your company. That is your primary function. And general managers are overhead. Are you going to lay yourself off if work slows? No, you are going to put on your marketing and sales hat and hit the streets to find work. And as the general manager you need to pay yourself for this work.

Greg Crabtree makes this distinction clearly: you earn a wage for what you do, and a return for what you own. The wage — what it costs to have you running the company — belongs in overhead. The return — your share of what the business produces after it covers its costs — comes out of net profit.

These are two different things.

Conflating them is how the math breaks before you ever open an estimate.

If your compensation as the owner-operator is not in overhead, your OH&P is understated. If your OH&P is understated, your gross profit per day target is wrong. If that target is wrong, your margin was never built from the actual cost to run the company. And you can do your best work, on your best jobs, year after year, and still wonder where the money went.

In both cases — missing indirect overhead, owner comp in the wrong bucket — something real was left out of the equation. The margin looked like math. But it was not built from the actual question the business needs to answer:

Can this company generate enough gross profit, with this labor, over this amount of time, to pay for itself and leave the owner with the result they need?

That is the equation. The variables change. Jobs stretch. People leave. Overhead grows. The owner wants a different life than they wanted five years ago.

So you solve it again. And again. And again.

But you cannot solve an equation with variables missing. That is what the experiment taught me. The math was there. It was always there. We just had not put the right things into it yet.

The math is not separate from the work. It is the shape of the work made visible.