Every job made money and the year did not

A flooring contractor on Kishwaukee Street in Rockford, four in the afternoon on a Monday, the warehouse still smelling of adhesive from the morning pull. Beside the desk, on the floor, stood the carton the stair nosing had shipped in, packed upright with manila job folders. Somebody had written TO COST down one side of it in grease pencil.

I lifted the folder off the top. That job had finished eight months before we sat down.

Doyle has been laying floor in Winnebago County for twenty-nine years. Twenty-four people, five crews, a hardwood man half the builders in Belvidere ask for by name. Four point eight million last year, his best.

I asked what his gross margin was. He said thirty-one percent, and he said it without pausing, the way a man says something he has said a hundred times.

I asked where the number came from. He said that is what we bid.

The margin he bid and the margin he got

Thirty-one is what the estimating sheet returns when you type a job into it. A forecast, made before anybody drove out there. Doyle had never sat down with a finished job and worked out what it cost by the time the last strip went down.

A few he knew about. The slab on Auburn that came up an inch out over eleven feet, where the crew burned a day and forty bags of leveler nobody had bid. What the ordinary jobs cost him, he could not say.

Ordinary jobs are most of a year.

So the year runs. The board books four months out, the crews are good, the callbacks are few. Then the books close with a number that does not match how it all felt, and it gets explained on the spot. Material went up. That builder paid slow, and we bought the van.

Some of this will read like your place:

You can say what your gross margin ought to be. Saying what it was last month means looking somewhere, and you are not sure where.

Ask what the company has to bill in a month before a dollar is yours, and the honest answer is a call to your accountant.

A pallet runs short on a Tuesday, somebody drives to the supplier, and the ticket lands on whichever job was open on the screen.

No finished job has been laid beside its own estimate, line for line, and read down both columns together.

You know which customer buys the most from you. You could not say which one leaves the most behind per man-hour on site.

The sentence every man in the trade says

Doyle did not think he had a costing problem. He thought he had a flooring business. Thin margins are the trade. Everybody runs on three or four points of net, the builders squeeze you, and that is how it goes. Every man in that trade says a version of it, and there is enough truth in it to be dangerous.

Once you believe the industry set your margin, one lever is left. Sell more. So the move is a production builder, forty houses a year, priced tight on the understanding that volume covers the overhead. Doyle signed one two years back. It paid for the fifth crew.

Volume helps only if the margin on it clears the cost of your building. Below that line, every extra house adds revenue and subtracts cash, and a busy warehouse still feels healthy. The year you grow fastest is the year you are shortest of money.

The next move a man makes is to job cost. The bookkeeper codes material to job numbers and hours off the timecards, and inside a few months there is a report. Almost every job lands within two points of the bid, and he feels better.

That report is built out of whatever was easy to code. The drive to Machesney Park goes to shop time, and so does the second trip for the stair nose. A Saturday punch day sits on an overtime line and never reaches the house it was worked on. The plank cut wrong is a write-off in a bin, on no job at all.

If every job made its margin and the company did not, your job costing is wrong. The bank statement is the one telling the truth.

Timing does the rest. A job costed eight months after the crew drove away tells you about something dead, and the same detail gets bid again on Thursday.

Eleven folders on the conference table

We pulled eleven folders out of the carton, spread across the kinds of work he does. Three builder houses, four retail jobs out of the showroom, two apartment turns, a church hall in Loves Park, a small commercial job in Cherry Valley. Then we costed them the slow way, off supplier tickets and real hours with the driving in, the subs, the leveler, every trip back.

The eleven came in at 25.5 percent against a bid of thirty-one. Two beat their number. Three were bad, and he could have named those from memory. The other six were quietly short, each for a reason too small to raise on a Friday.

Five and a half points of margin on four point eight million dollars of installed work is $264,000. His net profit that year, the whole company, was a little over seventy thousand. The hole was more than three times what the business kept.

He beat his own number all year and finished tight

Then the other piece of arithmetic. Add up everything the company spends that does not belong to a job: warehouse rent, office people, insurance, trucks and fuel, phones, advertising, his own pay. Doyle came to ninety-six thousand a month.

Divide that by your gross margin and you have the revenue you must bill before a dollar is yours. At the thirty-one he believed in, about $310,000 a month. At the 25.5 he was getting, $376,000. He had been running four hundred thousand a month, beating his own target by ninety grand, and ending every quarter with nothing.

Put the profit inside the overhead figure before you divide. Decide what the company has to earn, add it to the ninety-six thousand, divide the total. Do it the other way and profit stays whatever the year leaves behind.

Doyle could sell and Doyle could install. The company still could not buy a van without a conversation at the bank, still had nothing put by for the slow weeks between seasons, still lost a good tile man over three dollars an hour. Everything it wanted to do next needed money the jobs were supposed to bring home.

One job, one sheet of paper, one conversation

Pick a job that finished in the last three or four weeks. Not your biggest, not the disaster everybody argues about. An ordinary one, recent enough that the invoices and the timecards are findable in one sitting.

One sheet of paper, turned sideways, a line down the middle. Left column is what you bid. Right column is what it cost. Six rows: material, installation labor, subcontractors, freight and handling, return trips, and a last row for everything the job used that has always been treated as overhead. The measure, the delivery, the warehouse pull, the punch day.

The left column takes four minutes off the estimate. The right column is the work. Supplier invoices with that job on them, the timecards for those dates, the sub invoice. Then the slow part, asking out loud where the rest went. How many hours of driving. Who went back and why. What came out of shop stock and never got billed. Write those hours down even when nobody recorded them at the time.

At the bottom, three lines: sell price, cost, and the two margins side by side.

Now hand the sheet to the one person who bid the job. Just the sheet. Do not walk him through it, do not open by saying the margin came up short. Ask him to read down the right side and tell you what happened on each line.

Then one question. Which of these lines happens again on the next job like it. That separates a bad week from a bad price. A slab nobody could have seen is a bad week. Stairs bid at flat footage, a waste allowance nobody has touched in ten years, a haul-off you quit charging because one customer complained, those are prices.

Change nothing this week. Do a second job the same way next week, and pick a different kind of customer on purpose. By the fourth sheet you are looking at which work pays you.

What the carton looks like eight weeks on

The surprises go first. The books stop closing with news in them. You stop hearing your own voice explain a thin year with material prices and one slow payer. The argument about a job at a number nobody likes gets shorter, because somebody can say what the last three of those returned.

What starts is quieter. Your estimator changes a line before you ask him to. Somebody turns down a house. The carton by the desk holds four folders instead of thirty, and the oldest one is three weeks old.

Doyle kept the builder. He repriced the stairs, started charging the haul-off, and let two of the forty houses go to somebody else. A finished job is the only honest thing a company produces about itself, and most owners file it before they read it. The floors going down in Rockford were always good. Nobody had gone back to see what they cost.

Margin, Break-Even, and Job Costing is seven statements in the Business Check-in, and this article works through most of them. A finished job answers several of the other statements too, and the check-in is free and takes about ten minutes.