Your overhead rate is total annual indirect cost divided by an allocation base, usually total direct job cost, direct labor cost, or field labor hours. The result tells you how much overhead each job has to carry before it earns a dollar of profit.
Plan reviews connect field conditions, drawings, and the next decisions on a job. AI-generated illustration.
To calculate a construction overhead rate, add up every cost the company pays that does not belong to a single job (office rent, office salaries, insurance not charged to jobs, software, trucks that never get coded, the owner's salary), then divide that total by an allocation base. The most common bases are total direct costs on jobs, direct labor dollars, or field labor hours. A company with $780,000 of annual overhead and $5,200,000 of direct job cost has a 15 percent overhead rate on direct cost.
That number matters because every estimate either recovers overhead or quietly eats it. Bid a job at cost plus 8 percent when your overhead rate is 15 percent and the job loses money even if the field crew hits every budget line.
Overhead rate formula
Overhead rate = annual indirect costs ÷ annual allocation base. The base should be something every job has, that you track reliably, and that moves with how much of the office's attention a job takes. Federal cost accounting rules describe the same idea in plain terms: group indirect costs logically and spread each group with one base that reflects the benefit each job receives (FAR 31.203). You do not need to follow federal rules on private work, but the logic holds for any contractor.
Common allocation bases and who tends to use them
Base
Formula
Fits best when
Total direct cost
Overhead ÷ total job cost
You sub out a large share of the work and mostly manage
Direct labor cost
Overhead ÷ field wages
You self-perform most work and labor drives the office load
Field labor hours
Overhead ÷ field hours
Crews work at different wage rates, or you price service and T&M work by the hour
Revenue
Overhead ÷ contract revenue
Quick check only; it hides the effect of markup
How to calculate it step by step
1. Separate job costs from company costs
Go through the chart of accounts and mark each expense as direct (charged to a job) or indirect. Be strict. Job-site trailers, a project superintendent, dumpsters, and permits are job costs, usually under general conditions. The estimator, the bookkeeper, the office lease, and the CPA are overhead. If a cost gets coded to jobs some months and to overhead other months, pick one home and stick with it.
2. Total twelve months of overhead
Use the last full fiscal year as a starting point, then adjust for known changes: a new project manager starting in March, a lease renewal, a jump in insurance. Include a market salary for the owner even if the owner takes draws instead. Leaving it out makes the rate look healthy while the owner works for free.
3. Pick the base and total it for the same period
Pull total direct cost, direct labor dollars, or field hours from the same twelve months. Mixing a forward-looking overhead budget with last year's smaller base inflates the rate. Mixing last year's overhead with next year's hoped-for volume deflates it, which is the more dangerous mistake.
4. Divide and sanity check
Run the division, then check the answer against a second base. If the rate on direct cost says 15 percent and the rate on labor says 60 percent, both should recover the same total dollars on an average job. If they don't, your job mix is uneven and the base choice matters more than you thought.
5. Load the rate into estimates and job reports
Add overhead as its own line in the estimate, before profit. Keeping it separate lets you see how low you can go on a must-win bid without pretending overhead disappeared. The same rate should show up in your job cost report so finished jobs are judged on net, not just gross.
6. Recheck every quarter
Volume swings change the rate fast. If revenue drops 20 percent and overhead stays flat, the rate climbs by a quarter. Compare actual overhead and actual base each quarter and adjust estimates before the gap shows up in year-end numbers.
Worked example: a $6 million general contractor
Say a commercial GC closed last year with these numbers. The figures are illustrative, but the math is the same at any size.
Example company, last fiscal year
Item
Amount
Annual overhead (office, admin, owner salary, insurance not on jobs)
$780,000
Total direct job cost
$5,200,000
Direct field labor cost
$1,300,000
Field labor hours
52,000
Rate on direct cost
15%
Rate on direct labor
60%
Rate per field hour
$15.00
Now the estimator prices a $240,000 direct-cost tenant improvement. At 15 percent on direct cost, the job carries $36,000 of overhead, so break-even is $276,000. Profit goes on top of that number, not on top of $240,000. If the same job has only $40,000 of self-performed labor because most scopes are subbed, a labor-based rate would assign just $24,000 of overhead. That gap is why subcontract-heavy GCs usually pick total direct cost as the base.
Mistakes that throw off the rate
Leaving out the owner's pay. If the owner draws profit instead of salary, the overhead rate understates what the company costs to run.
Using a hoped-for revenue number. Spreading overhead over volume you have not won yet makes every bid look cheaper than it is.
Coding the same cost two ways. A superintendent's truck charged to jobs in summer and overhead in winter distorts both the rate and job margins.
Adding overhead on revenue but pricing on cost. Pick one base and apply it the same way in the estimate and the job report.
Never updating it. A rate set three years ago at a different volume is a guess, not a rate.
Overhead rate vs markup
The overhead rate is a cost. Markup is what you add to cost to cover overhead and profit together. If your overhead rate on direct cost is 15 percent and you want 8 percent net profit, your markup has to clear both. The difference between markup and margin trips up a lot of estimators, so read markup vs margin in construction and how to set construction markup and profit margin before you set the final number. For a forward-looking version of this calculation, see how to budget overhead in a contracting business.
Smart Business Automator keeps job costs coded consistently and shows job costing and WIP views next to company totals, so the inputs for your overhead rate come from the same data your PMs already use. See job costing and WIP automation.
Frequently asked questions
What is a typical overhead rate for a construction company?+
It varies widely by trade, size, and how much work is subcontracted, so there is no single right number. Calculate your own from your books rather than borrowing an industry figure, because your bids need to recover your costs, not an average company's.
Should overhead be calculated on revenue or on cost?+
Most contractors calculate it on cost (total direct cost, labor cost, or labor hours) because estimates are built from cost. A revenue-based rate is fine as a quick check but it mixes in markup, which makes the rate change whenever pricing changes.
Are general conditions part of overhead?+
No. General conditions are job-specific costs like the site superintendent, temporary facilities, and dumpsters, and they belong in the job's direct cost. Company overhead covers costs that support all jobs, like the main office and accounting.
How often should the overhead rate be updated?+
Recalculate it at least once a year when you close the books and check it every quarter. A big change in volume or a new office hire can move the rate enough to matter on the next bid.
A job can be overbilled, cash-positive, and still be moving toward a weaker result. This briefing uses a documented $2 million WIP example to show how a six-point margin fade can hide behind $184,000 of overbilling—and maps the field-to-finance workflow needed to surface that warning early.
Starts muted when the player comes into view.
Read the full episode transcript
Published transcript from Scaling Legends. It may contain transcription errors. View the source transcript.
You have a job that is currently overbilled by $184 ,000. Okay. It is bringing in strong cash. I mean, the bank account looks incredibly healthy. Right. But on that very same job, the actual profit margin has silently plummeted from 20 % down to 14%. Yeah. And the consequence here is, well... Your leadership team looks at the bank balance and thinks this project is a resounding success. Right, because the cash is there. Yeah, the cash is there, while the underlying economics of the deal are actively failing. Yeah, that is a dangerous place to be. It really is. So this is an AI -assisted Sailing Legends decision briefing. It's produced by Smart Business Automator from the cited sources in our show notes. And Smart Business Automator, they design AI and
automation systems for construction companies. Right. And before we dive into the data today, please note that this briefing is for educational and informational purposes only. Yeah, not legal or financial advice. Right. Not professional advice. You should always consult your CPA. for your specific circumstances. So our mission today, we are going to look at the work in progress or the WIP report. The famous WIP report. Yeah, but not just as a historical financial document. We need to look at it as an operational warning system. I like that. I mean, think of it like driving a heavy duty truck. You glance at the dashboard and the fuel gauge shows a full tank. Okay. That full tank is your overbilling. It's your positive cash flow. Right. But what you can't see from the driver's seat is this hidden
leak in the fuel line. Oh, man. And that leak is your margin fade. So you were going to run out of gas long before you reached the destination. Yeah, you'll be stranded on the highway. Stranded, yeah. All because the dashboard told you everything was fine. That fuel leak analogy, I mean, it really captures the visceral disconnect between cash and true profitability. Because to understand how a margin fade hides so effectively behind positive cash flow, we have to trace the actual deal math. It's remarkably easy to get lulled into this false sense of security. and the checks are clearing, and the working capital is flush. Well, so if I'm auditing this tomorrow, I don't want to chase ghosts. Right, you need specifics. Give you the specific smoke signals. I want to look at the mechanics of a verified scenario.
Okay, let's do it. So this breakdown, it comes from the Construction Financial Management Association, or CFMA. Right. In their July 6, 2026 guide. Okay, $2 million. The original estimated cost to complete this job was $1 .6 million. Got it. So that gives us an original estimated margin of 20%. But then the reality of the muddy job site hits the clean spreadsheet. It always does. Fast forward a bit, and the actual costs incurred to date are $960 ,000. Okay. However, the revised estimated total cost has jumped. It went from $1 .6 million up to $1 .72 million. Okay. Yeah, so hold on. Let me make sure I'm following the specific cost -to -cost mechanics here. Yeah, let's break that down. If we take the $960 ,000
incurred and divide it by the new $1 .72 million total expected cost, what does that give us? So you get a percent complete of 55 .8%. Okay, so we take that 55 .8 % and apply it to our $2 million contract price. Right. So they have actually billed $1 .3 million to date. And that right there, that is where the trap snaps shut. Wait, really? Okay, so doing the subtraction here. $1 .3 million build minus the earned revenue of $1 ,116 ,000. That leaves us with a billing position that is $184 ,000 overbilled. Spot on. So play devil advocate for me here. The cash is flowing in. The customer is paying ahead of the work. Why is this a crisis? It's a crisis
because of the profit calculation you're missing. I mean, remember that new total cost estimate of $1 .72 million? Right. It went up. Yeah. So subtract that $1 .72 million from your $2 million contract. What? Your total profit. is now only $280 ,000. Oh, wow. Yeah, that brings the current margin down to 14%. Wow. So a full six -point margin fade has occurred right under the nose of strong cash flow. Just hiding there. Yeah. And overbilling itself is not a bad thing. In fact, it's a strategic way to support working capital. Sure, you want the cash. You want to use the project's cash, not your own reserves, to fund the ongoing work. That makes sense. But the systemic danger is treating that cash as evidence that the job economics are healthy.
Right. Especially when the current cost estimate is screaming otherwise. Which, I mean, that brings up an obvious friction point. What's that? The math we just did is basic division and subtraction. Yeah, it's not calculus. No, so why does this specific trap catch highly seasoned project executives and CFOs by surprise? at the end of the month? It's a great question. I mean, these are intelligent people who look at numbers all day long. We miss the math because of where the data lives and how it travels from the job site to the CFO's desk. Ah, okay. The calculation is simple, yes, but the data collection process is deeply flawed. Got it. We really have to look at the human behavior driving the workflow. Which is funny you say that. A WIP report is supposed to be a structured conversation between the field and the office.
A conversation, yes. It's not supposed to be just a static accounting artifact. Right. The decision makers who actually own this problem are the CFO, the controller, the owner, the project executive, and the operations leader. That's a lot of people who need to be on the same page. Yeah. They need this to be a living dialogue. But in reality, that dialogue is completely fragmented. I mean, let's map the broken failure chain that a lot of companies are operating with right now. Okay. Lay it out for me. Picture this. It's 4 p .m. on a Friday. Oh, no. Yeah. The WIP review meeting is already 20 minutes over time. Everyone wants to go home. Yep. The project manager says they are 80 % done, but accounting says they've spent 95 % of the budget. Oh, boy. Now everyone is just staring at a spreadsheet arguing over who typed the wrong number. I can feel the blood
pressure rising in the room just hearing that scenario. It happens every month. And it happens because the crucial field information arrives in unstructured formats. Like what? You know, daily reports, text messages, quick phone calls from the truck, long email threads. Right. None of that flows naturally into a financial system. No, it's just a mess of data. And then look at the actual uncommitted costs. Those are siloed over in the accounting or purchasing departments. Completely separate. Yeah. Meanwhile, pending changes, which... I mean, are the lifeblood of profitability on a complex job? Oh yeah, change orders are huge. They sit in a completely separate log. Or, even worse, in a project manager's inbox
waiting for a signature from the owner. And we can't ignore the psychology here either. What do you mean? Well, project managers are updating their costs to complete on different dates throughout the month, using varying assumptions. That is so true. Like, one PM might factor in a looming material price escalation, while another just carries the original budget forward because they were rushed. Yeah, they just want to get back out into the field. Right, or maybe they're just delaying. Delivering bad news. Which leaves the controller at the end of the month playing the role of a forensic investigator. That sounds awful. It is. They are left to manually stitch together a monthly WIP spreadsheet by pulling from all these disconnected silos. So they're just chasing people down. Yeah. The result is that leadership meetings devolve into arguments over data integrity rather than taking decisive
action on the actual margin fade. So putting a shiny software dashboard over this broken process? does not fix it. Not at all. It only creates a polished, highly unreliable forecast. Right. Garbage in, garbage out. Yeah. The report is late because the workflow is fragmented. Automating a final dashboard without fixing those underlying handoffs will just make a bad forecast look pretty. Okay. So since spreadsheets and fragmented ERP systems cause these late warnings, we clearly need a different approach. We do. How do we engineer a system that surfaces that $184 ,000 trap? Instantly? Yeah. Like before the end of month meeting even happens? Well, we have to build a functioning WIP automation workflow. Okay. And this requires a specific blueprint. First, we need to standardize the inputs. Which inputs? The system must automatically
pull the original estimate, the current contract value, the actual and committed costs by job and cost code. Okay. Plus the build -to -date figures, the pending change value, schedule progress, and critically, the prior period WIP values. So we are bringing all the disparate silos into one central rules engine. Yes. And that rules engine is where the analytical heavy lifting happens. How so? Well... A standard system might rank exceptions by gross profit percent. Right. But a smart rules engine ranks exceptions by projected gross profit dollars. Oh, that is a subtle but vital distinction. It really is. Because a 2 % fade on a $10 million job is a much bigger crisis than a 5 % fade on a $50 ,000 job. Yes,
exactly. I mean, you want to prioritize the actual cash impact to the business, not just a ratio. Right. So what other... rules need to be in place? The engine needs to require an explanation when a value crosses a specific company -defined tolerance. And here is a critical control. It must prevent pending changes from entering the contract value until their treatment is formally approved by the company's accounting and contract process. Oh, because you can't recognize revenue on hope. You really can't. It'll sink you. So once the rules engine identifies a variance, what does the automated action look like? The system must automatically route these exceptions directly to the project manager and the project executive. Fucking alert? Yeah, but it can't just send a generic alert. It must specifically ask for the
cost code driver, the cause of the variance, the corrective action plan, the owner of that plan, and a firm... due date. That's a lot of accountability. It has to be. At the same time, it must notify finance if this margin movement creates a cash exposure, just like we saw in our opening example. Okay, I have a strategic question here, looking at this from the operations side. Sure, go ahead. What stops a project manager from just typing weather delay every single month to clear the alert? and get back to work. Oh, the old weather delay. Right. I mean, we all know that is the industry's favorite get out of jail free card. It really is. And that is the Achilles heel of most reporting workflows. So how do you fix it? The solution is memory. Memory. Yeah. The automated system must retain
the previous month's explanation and forecast version and display it during the next review. Oh, I see. It must explicitly show leadership whether that prior corrective action actually improved the situation, stabilized it, or worsened the projection in the current period. Oh, wow. So you can't just say weather delay three months in a row. Right. Accountability is built into the workflow, making it impossible to hide behind a recurring excuse. That's powerful. So armed with an understanding of what a healthy system looks like, we need concrete ways for leaders to audit their current reality this week. Yes. Action items. The sources highlight three specific red flag combinations. And we need to be very clear on this point. What's up? Do not present these as universal proof of a loss. They are
triggers for a thorough review. Right. They're starting points. Walk us through the first combination. Flag number one is overbilling plus margin fade. This is our opening scenario. Yep. The fuel leak. Right. The cash is ahead while the projected economics are deteriorating. That's the classic false positive. Yes, it is. Flag number two is growing underbilling coupled with pending changes. Ooh, that's a bad one. Yeah. In this scenario, the contractor is essentially acting as a zero interest bank. Basically. They are funding the work out of pocket while the approval and billing processes lag behind. And that is a severe cash flow drain that can cripple a growing company if left unchecked. I can imagine. It means the physical work is happening, but the paperwork is stalled and you are just bleeding capital.
Right. And flag number three. Okay. The percent complete rises while the estimate to complete does not change. I like to think of this as a broken thermometer. That's a good way to put it. A stable forecast might actually just be a stale forecast that no one is updating based on field reality. Yeah, no one is actually checking the numbers. Right. The project is advancing. Money is going out the door. The schedule is progressing. But the financial projection is sitting in a freezer from three months ago. Those three triggers are really powerful indicators. Yeah, they are. To operationalize this, we need three concrete actions the listener can take immediately. Okay, what's action one? Action one. Select the top five active jobs with the largest remaining cost. Just five. Yeah. Do not
try to boil the ocean and look at everything. Just focus on those top five. Makes sense. Compare their original margin, current margin, and prior period margin. Okay. And when they find a material discrepancy during that comparison. That leads to action two. Right. For every material change you find, you must hunt down the exact cost code, the field evidence, the pending change status, and the responsible recovery action. No vague summaries. Right. Do not accept a vague summary from the team. Track it down to the root cause. Action 3 is fascinating because it focuses on the internal cost of the broken workflow itself. Yes. It is a time audit. A time audit, yeah. Physically, time, how long it takes your finance team to assemble a supportable current forecast. Just sit there with a stopwatch. Basically, record
every single spreadsheet involved, every inbox check, every manual reentry, and every chase down approval. Wow, that's going to be a big number. The friction in this process is often shocking when you quantify it, which is the root cause of why the data is always weeks behind reality. But you know, once you start auditing and speeding up this process... Okay, so keep the AI out of the final call. Right. Automation
is highly effective at collecting data, calculating variances, comparing periods, routing alerts, and preserving evidence. It's a logistics engine. Exactly. Yeah, it is a logistics engine for information. The Financial Accounting Standards Board, or FASB. has very clear rules here. They sure do. Under topic 606, which governs revenue from contracts with customers, there are guidelines addressing the nature, timing, and uncertainty of revenue. And we are not offering an accounting conclusion here. But the principle is clear. The automation cannot make these nuanced calls. No, it can't. The workflow must route revenue recognition and contract modification judgments to the contractor's qualified accounting professionals. Human judgment remains the final gate always yeah the system
tees up the problem with all the supporting evidence but a qualified person must approve the forecast changes and the accounting treatment you're just speeding up the prep work yeah you automate the friction of gathering the data not the financial judgment itself got it you still need human experts evaluating whether a recovery plan is credible or whether an issue requires communication with a surety, a lender, or the customer. This brings us to the ultimate directive for your next WIP meeting. The big takeaway. The decision -maker move. So here's what you do. Sit down with the team. If they cannot clearly explain a margin movement, and if they cannot produce the supporting inputs... Like the field evidence and cost codes. Right, the real evidence. If they can't produce that, then you must physically mark that line
item as... And verified. Wow. Just unverified right on the sheet. Do not accept a guess. Assign a specific owner to track down the truth and set a hard deadline before the next WIP meeting adjourns. Because a forecast built on unverified assumptions is just a liability. 100%. Yeah. Now, if you are the CFO, controller, owner, or project executive of a construction company, and your WIP review still depends on spreadsheet chasing and late explanations, smart... So as you review those top five jobs this week, I want to leave you with one final question to ponder.
Okay, what is it? If your system cannot tell you the difference between a project that is truly profitable and a project that is simply cashing in early, who is really running the business, you or the calendar?
