Mark up your costs by 20 percent and your margin is 16.7 percent. Most founders price on markup and carry that number in their head as the margin. The difference is real money, and it compounds on every job you win. Below is how to find your gap in about a minute, and how to set a price floor that holds when a customer pushes on it.

Three moves this week

You finish a strong year. Trucks ran all week, crews stayed booked, the phone never went quiet. Then you open the year-end numbers and the profit is smaller than the year felt. Nothing went wrong that you can point to.

Two words are doing it, and most founders use them as if they mean the same thing.

Markup is what you add on top of cost. Margin is what you keep out of price. A job costs you 1,000 dollars and you add 20 percent. You charge 1,200. Your markup is 20 percent. The 200 you kept is 16.7 percent of what the customer actually paid. Margin is always the smaller number, and the bigger the job, the more dollars sit in the difference.

Residential HVAC company. Good crews, steady work, an owner who prices every install at a 20 percent markup and calls it a 20 point business.

Two million booked. In his head that is 400,000 dollars of gross profit. The math says 16.7 percent, which is 333,000. Sixty-seven thousand dollars he meant to keep and never put in the price.

That is a technician's salary. It is the second truck he decided to hold off on buying. It came out of the good jobs, a few points at a time, while he was watching for the bad jobs that never showed up.

Check your own number. Markup divided by one plus that markup:

To keep 20 points you price at a 25 percent markup. Run the same year through that number and the HVAC company bills 2.08 million on the same work and keeps 417,000. The customer pays about four percent more. The owner keeps 84,000 more.

Seeing the gap is the easy part. The number you defend comes from a floor, and the floor takes about an hour at the kitchen table with the bank statements and the job list open.

1) Your overhead. What it costs to keep the doors open for a year before a single job earns a dollar. Rent, trucks, insurance, your own pay, everyone who never touches a billable hour. Install crews stay out of it, because their hours land on the jobs. Call it $300,000.

2) Your job count. The number you run in an ordinary year, not the best one. Call it 300 installs.

3) Overhead per job. Divide the first number by the second. Three hundred thousand across 300 installs is 1,000 dollars. That $300,000 arrives whether or not you win any particular job, so every job you run owes $1,000 toward it.

The floor. Now you have a price you can write down. The lowest number you quote is what the job costs you, plus 1,000. Quote exactly that and the job pays for itself, pays its share of overhead, and makes you nothing.

Now follow one real install through it. $2M across 300 jobs averages 6,667 dollars. At a 20 percent markup, $5,556 of that goes to the equipment and the crew hours on that job. Another 1,000 goes to overhead.

What is left is 111 dollars.

That is the whole business. Thirty-three thousand dollars of profit on two million of work, and the owner thought the year had left him three times that. He was running a company with no room in it, and the pricing never told him.

A builder calls with twelve units and leans on the price the way builders do, landing on a round number that sits just under the line. The owner running on the 20 point story takes it to keep the crews busy and finds out in March what it cost him. The owner with a floor knows those twelve units have to clear 12,000 dollars over their cost, and holds.

The floor bends once, and you should know where before someone argues it at you. When the calendar has real holes and the crews are getting paid anyway, a job clearing 600 dollars beats a job you did not run. You make that call with your eyes open in a slow month. The builder does not get to make it for you.

One caveat on the arithmetic. Splitting overhead evenly per job works when your jobs are similar in size. If a 60,000 dollar retrofit sits in the same count as a 3,000 dollar changeout, divide overhead by billable field hours instead and load the floor on by the hours each job takes.

All of this is arithmetic, and arithmetic is now cheap. Export last year's job list, every job with its cost and its price, and give an AI tool three asks.

Copy and paste this with your job list attached:

Prompt Design by: The Next Rung

Ten minutes later the 67,000 is no longer a feeling about a thin year. It is a list of named jobs, with the builder work at the top, priced under the floor three years running.

Set it up once and it keeps going. Paste in each week's completed jobs and it flags anything under the floor while the customer's check is still in the mail. The year-end reckoning becomes a Tuesday habit that takes less time than a coffee.

Then know where the machine's job ends. It counts, converts, and flags. The three numbers underneath the floor are judgment calls. The honest volume number is you deciding to stop planning on the best year ever. Your own pay sits in the cost line because you put it there. And when the builder leans on twelve units, the machine has no seat in that conversation. The floor holds because a person with authority decided it holds.

Trucks still ran all week. The phone still did not stop. This time the profit is the size of the year, because every job cleared a floor the owner set on purpose and priced to a margin he can say out loud.

The number is boring, and boring is the point. It holds when the room gets tense and the job is one you want to win.

Mark up your costs by 20 percent and you keep 16.7. Decide whether that is the number you meant to keep.

If you run a founder-led business and you are not sure the margin you keep is the margin you planned, LightPath Advisory rebuilds the pricing floor and the reporting that keeps it honest. If that is the work in front of you, we are glad to have a conversation.