How to calculate your true overhead rate

There is a number in your estimating template that decides whether you make money, and there is a good chance nobody has checked it in three years.

The short answer: your true overhead rate is total annual overhead divided by total productive field hours. Not divided by revenue. Not a percentage you add at the end. A dollar figure per hour of actual production, recalculated at least once a year.

Most contractors use a different method, and it is wrong in two directions simultaneously.

The method most people use

It goes like this: take last year's overhead, divide it by last year's revenue, get a percentage, add that percentage to every bid.

Say overhead was $420,000 and revenue was $3,000,000. That gives 14%. So you add 14% to cover overhead, then your margin on top.

Clean, simple, and wrong for two separate reasons.

Reason one: overhead is not a percentage of revenue. Your rent does not care what you charged for a job. Neither does your insurance, your software, your estimator's salary, or the truck that never leaves the yard. Those costs accumulate with time, not with sales. When you express a time-based cost as a percentage of a sales figure, the two only line up if next year's revenue matches last year's exactly. It never does.

The distortion runs in a predictable direction. When you price a job cheaply, a percentage-of-revenue method assigns it less overhead — even though a cheap job consumes exactly as many days of your rent and insurance as an expensive one. Under-priced jobs get under-charged for overhead, which makes them look better than they are, which encourages you to take more of them.

Reason two: last year is not this year. You are bidding work that will be delivered in the coming months, against overhead you will incur in the coming months. If you added a truck, a salary, or a software subscription, last year's rate is already stale.

The method that works

Two numbers.

Numerator: total annual overhead. Everything that is not a direct job cost. Rent, utilities, insurance, licences, vehicles that are not on jobs, office staff, your own salary, software, accounting and legal, marketing, bank fees, training, phones, and the estimator's time. If it gets spent whether or not a specific job exists, it belongs here.

Denominator: total productive field hours. This is where it gets interesting.

True overhead rate = total annual overhead ÷ total productive field hours

Productive field hours are hours that can actually be charged to a job. Not hours paid. Not hours on the clock. The hours where someone is doing work a customer is paying for.

Why productive hours is the number that bites

Take a crew member paid for 2,080 hours a year. How many of those end up on a job?

Subtract vacation and statutory holidays. Subtract sick days. Subtract travel between sites, shop time, loading and unloading, yard clean-up, training, toolbox meetings, warranty callbacks, and the two hours lost when a delivery did not show.

What is left, for most trades businesses, is 60 to 75 percent of hours paid.

That range is the single most important thing on this page. If you are dividing overhead by hours paid instead of productive hours, your overhead rate is understated by somewhere between 25 and 40 percent — and every bid you have sent this year carries that error.

Worked through:

Same company, same costs. A difference of $14.43 an hour.

On a job with 240 labour hours, that is $3,463 of overhead you either recovered or did not. On forty jobs a year of that size, it is $138,000. That is the difference between a good year and wondering why a busy year felt tight.

Finding your real productive percentage

You do not need to guess. Two ways to get there.

The direct way. If you run timesheets that code hours to jobs, total the job-coded hours for last year and divide by total hours paid. That ratio is your answer.

The reconstruct way. If you do not have that, build it:

• Start with 2,080 hours per person
• Subtract vacation (say 80 hours) and statutory holidays (say 72)
• Subtract sick and personal time, honestly — most crews run 40 to 60 hours
• Subtract non-productive time: travel between jobs, shop, meetings, training, callbacks

It is the last line that people underestimate. Half an hour a day of travel and loading, across 230 working days, is 115 hours — more than three full weeks.

Add it up and you typically land between 1,250 and 1,550 productive hours per person. Against 2,080 paid, that is 60 to 75 percent.

Applying it to a bid

Once you have a dollars-per-hour figure, the estimate gets simpler, not harder.

``` Job cost = labour (burdened) + materials + subs + equipment + (estimated labour hours × overhead rate)

Price = job cost ÷ (1 − target margin) ```

Note the last line. To hit 25% margin you divide by 0.75 — you do not add 25%. Adding 25% to cost gives you a 20% margin, and that five-point gap is one of the most common silent errors in trades estimating.

Same intention, $4,167 apart.

What to do when the number comes out high

Most people running this for the first time get a figure noticeably higher than what is in their template, and the instinct is to assume a mistake.

Check three things:

Is anything in overhead that is really a job cost? Materials, subs and site labour belong in direct cost, not overhead. Double-counting inflates the rate.

Are you including your own salary? You should. If you are not paying yourself a real wage, your overhead is understated and the business only looks viable because you are subsidising it.

Are the productive hours right? If you used 85% because it felt right, go back and build it from the bottom.

If it survives all three, the number is real — and every bid you have sent this year is light by the gap.

What to do about it

You do not have to reprice everything overnight. But you do need to stop bidding at the old rate.

Change the template first. Whatever is currently in your estimating spreadsheet, replace it.

Rebid the next job at the real number and see what happens. The fear is always that you will lose everything. In practice you lose some, and the ones you win are worth having. There is a whole separate discussion about deliberately pricing to lose bids, and it starts here.

Recalculate quarterly for the first year. Overhead moves when you hire, add a vehicle, or take on a lease. Annual is enough once it stabilises.

Why this one matters more than the rest

Of everything in the weekly numbers, this is the quiet one — the one that does damage without ever showing up as an event.

A job that visibly goes wrong gets attention. An overhead rate that is $14 an hour light does not produce a bad day. It produces a slightly disappointing year, repeatedly, and it is almost impossible to trace back to a cause unless you go looking for it deliberately.

Which is exactly why it is worth an afternoon.

When volume changes, the rate changes with it

One trap catches people the year after they first do this calculation.

Your overhead rate is a division: fixed costs spread over productive hours. Change the denominator and the rate moves, even though nothing about your costs did.

Say your overhead is $180,000 and last year you billed 4,200 productive hours. That is $42.86 an hour, and you build it into every bid.

This year the schedule is thinner and you bill 3,400 hours. The same $180,000 now needs $52.94 an hour to cover it — but you are still bidding at $42.86. Across 3,400 hours that is a shortfall of roughly $34,000, and it does not surface anywhere until the year-end statements arrive.

The under-recovery is invisible because every individual job still looks profitable. Each one covered its direct costs and its allocated overhead at the old rate. The gap sits in the difference between the rate you used and the rate you needed, and no single job report will ever show it to you.

Two practical responses.

Recalculate at least twice a year, using a rolling twelve months of costs and hours rather than last year's numbers. If the business is changing shape — a new hire, a new vehicle, a slow quarter — recalculate sooner.

Bid on realistic hours, not hopeful ones. If you set the rate using the hours you hope to bill, you have built optimism into the price of every job. Use the hours you actually billed over the last twelve months, and if the coming year looks quieter, use a lower number rather than a higher one.

The contractor who raises their overhead rate in a slow year feels like they are making themselves less competitive. They are doing the opposite — they are the only one in the market whose prices are telling the truth.

Common questions

Should my own salary be in overhead?

Yes, at a real market rate for the work you do. If you are not paying yourself properly, your overhead is understated and every price you quote is too low. The business only looks viable because you are subsidising it personally, which is not a business model — it is a slowly depleting personal balance sheet.

How often should I recalculate my overhead rate?

Quarterly for the first year, annually after that, and immediately after any structural change — a new hire, a new vehicle, a new lease, a significant insurance increase. An overhead rate that is a year stale in a business that has grown 20 percent is understating every bid you send.

What if my overhead rate makes me uncompetitive?

Then you are currently winning work that does not pay, and your competitors are either more efficient, carrying less overhead, or making the same mistake you were. Test it before assuming the worst: bid the real number on the next few jobs. You will lose some. The ones you win will be worth more than the ones you lose.