Overhead at a Plumbing Contractor: Where the Office Hours Actually Go

The problem: You know what your office staff cost you in total, but not what any particular piece of their week buys, so overhead is a number you carry rather than a number you manage.

The solution: Count the hours behind one job from the call to the payment, price them against real salaries, and the overhead you have been treating as fixed turns out to be mostly a handful of repeated steps.

Ask a plumbing contractor at this size what a service call costs to run and you get a good answer about the truck, the technician, and the material. Ask what it costs to process, meaning everything that happens in the office before the technician arrives and after he leaves, and the answer becomes a shrug and the word overhead.

Overhead is where the unexamined money lives. It is a single line on the profit and loss, it is large, and it is understood as the price of having an office at all. So it never gets taken apart. The result is that a contractor can tell you gross margin by job type to the point and cannot tell you what a dispatcher's Tuesday afternoon was spent on, or what that afternoon cost, or whether the customer paid for any of it.

Overhead is not one thing. It is a few steps repeated thousands of times.

The reason overhead resists examination is that it looks like a person rather than a task. You are paying a dispatcher, an office manager, and someone who does billing and permits. Those are salaries, and salaries feel fixed, so the mind files them next to rent and insurance.

They are not fixed. They are the sum of a small number of steps, each of which happens on every job you run. The call comes in and gets written somewhere. It gets put on a schedule somewhere else. The technician writes down what he did. Someone reads that and decides what to bill. Someone checks whether the parts on the ticket match the parts on the purchase order. Someone types the result into the accounting system. Somebody chases the customer when it does not get paid.

Run 4,000 calls a year and each of those steps runs 4,000 times. A step that takes four minutes and touches every job is roughly 265 hours a year. Nobody has ever timed it, because it is four minutes.

The hours nobody has counted

Here is the specific gap. Payroll knows what your office staff are paid. Your job costing, if you do any, knows what the field cost. Nothing in the business knows how many office minutes attached to job 4187, so nothing can tell you which kinds of work are expensive to administer.

That matters more than it sounds. Two jobs can carry identical field margin and cost wildly different amounts to run through the office. A recurring commercial maintenance account with a standing purchase order and one invoice a month is cheap to administer. A one-off residential job with a permit, a change on site, a warranty callback, and a customer who disputes the bill can consume more office time than the technician spent on the property. Both show up on the profit and loss inside the same overhead line, averaged into invisibility.

So the questions that stay unanswered are:

• Which job types cost the most to process, as opposed to the most to perform.
• How much of the office week goes to entering information the business already has somewhere else.
• What you are actually buying when you hire the next office person, and whether the work would still exist if the information moved on its own.
• Whether your overhead grows with revenue because it must, or because it always has.

Why the work exists in the first place

It is worth being precise about the cause, because it decides the fix.

Most of this work exists because information is captured in one place and needed in another, and a person is the connection between them. The technician writes the job on a paper ticket. The ticket travels in a truck. Somebody reads it and retypes it. The purchase order lives with the supplier portal, the labor lives with the timesheet, the invoice lives with accounting, and none of the three know each other, so a person reconciles them.

That arrangement was correct when the business was small enough to hold in one head. At $9M it is a tax on every job, and it scales with volume rather than being absorbed. Every new customer adds to it. Every new technician adds to it. That is why the office grows a person roughly every time the field grows a truck, which is a ratio most owners have noticed and few have questioned.

What changes when the information moves on its own

The fix is not a faster typist or a stricter process. It is that the ticket, the hours, the parts, and the invoice stop being separate documents and become one connected picture of the job that each of them writes into as the work happens.

The technician closes the job on a phone, and that closes it everywhere: the schedule, the job record, and the draft invoice. The parts used come off the truck stock without a second entry. When something does not match, the exception surfaces on its own rather than waiting for someone to catch it during a reconciliation nobody scheduled. The office manager stops assembling the picture and starts working the exceptions, which is a smaller job and a more useful one.

The point is not that the work is done faster. It is that most of it stops needing to exist, and the office hours per job become a number you can see, per job type, on purpose.

A look at a plumbing contractor

Consider a commercial and residential plumbing contractor doing about $9 million a year: 38 people, five office staff, running service, small projects, and a growing maintenance book. The books are clean. The owner watches gross margin by division and is comfortable with it. Overhead is about what he expects for a company this size, which is to say he has never taken it apart.

Suppose he counts the office hours behind one week of jobs. You would expect the result to be uncomfortable in a specific way. Three of the five office staff, on salaries averaging roughly $62,000, would likely turn out to spend on the order of two days a week each on transfer work: retyping field tickets, matching supplier invoices to job numbers, rebuilding a schedule that already exists on a whiteboard, and chasing the technician who forgot to note the part he used. Two days a week each is about 40 percent of three salaries, so roughly $75,000 a year of payroll that produces nothing a customer would pay for.

Then the second number would appear, and it would be the more interesting one. Sorting the office hours by job type, you would expect one-off residential work to consume several times the administrative minutes of the maintenance accounts, on jobs that carry lower field margin to begin with. That is a repricing decision and a sales decision, and until the hours are counted there is no basis for making either.

What an owner would likely do with this is not lay anyone off. It is stop hiring the sixth office person he had penciled in for next year, move one of the three onto collections and permit follow-up where the time produces cash, and raise the price on the job type that quietly consumes the office. None of that is available while overhead remains one line.

How to start

You can do the first pass yourself, on one week, without buying anything.

1. Pick twenty finished jobs. Take a spread across the job types you actually run, not the memorable ones.
2. Trace each one through the office. From the call landing to the payment clearing, list every time a person handled the information, and roughly how long it took.
3. Price the minutes. Convert to hours a week, then to a share of the real salary. A number in dollars lands where a number in minutes does not.
4. Sort by job type, not by person. The question is which work is expensive to administer, not who is slow.
5. Separate the steps that would vanish. Mark every step that exists only because two records do not talk. That set is the size of the opportunity, and it is usually most of the list.

The takeaway

Overhead in a contractor this size is not a fixed cost of having an office. It is a handful of steps repeated on every job, priced into salaries you have already committed to, and never counted because no single instance is big enough to notice. Count the office hours behind twenty jobs, price them against the real salaries, and sort them by job type. You are likely to find both a payroll number nobody has ever seen and a job type that has been quietly expensive all along.