What Is a Job Cost Sheet? Components, Math, and How Trade Shops Use One

Construction employers paid an average of $48.58 per hour in total compensation in March 2025, and $14.79 of that (30.5 percent) was benefits rather than wages, according to the BLS Employer Costs for Employee Compensation release. In our customer base, the shops that price work off the hourly number they write on a paycheck are the same shops that finish a profitable year with no idea which jobs paid for it.

This gap is where the cost sheet for a job closes.

Contractors who post on trade and construction subreddits keep asking one question: How can you create a G702 as well as a G703 pay application without having to purchase software? The responses are the same every time. Spreadsheets. Word templates are rebuilt each billing period. Arithmetic on the Continuation Sheet is reconciled manually, and one transposed figure can cause the entire package to be returned. The issue is not always about formatting. The issue is that the numbers entered into the pay application were not recorded at the job level from the beginning, so each billing cycle becomes a reconstruction of the project.

What follows is the definition, with the operational portion of the definition: what a cost sheet looks like in terms of accounting and, in reality, what is on one, how cost buckets change from trade to trade, and the point at which the sheet breaks within a functioning shop.

In accounting, the job cost sheet is a record of all costs incurred for a particular job within a job-costing system: direct labor, materials, and applied overhead, all totaled and compared against the estimate as well as the amount that is ultimately paid. One sheet per job. This is what the rest of the sheet is attached to.

Textbooks have accounting compile the sheet after the job closes and hand it to management to judge whether the bid was right. That works in a plant where the job sits on the floor for six weeks. It does not work in service, where the invoice usually goes out the same day the tech leaves the driveway. By the time an after-the-fact sheet exists, the price is already set and the customer has already paid it. That timing difference is the whole argument for keeping job costing inside the same system that dispatches and invoices the work.

Many businesses that begin using job cost sheets with us are moving from spreadsheets or paper folders, or even a QuickBooks-only configuration, where job costing was, in fact, invoices compiled after the fact with no per-job breakdown of labor, material costs, or overhead. QuickBooks will tell you that your business is profitable. A job cost sheet will tell you which jobs were not profitable.

CFMA's Construction Financial Benchmarker put average gross profit margin for specialty trade contractors at 22.4 percent in fiscal 2024, with net profit before tax in the 6.2 to 6.7 percent range (CFMA executive summary). Owners respond to that benchmark one of two ways: they assume they are roughly at it, or they know they are under it and cannot name the work dragging them down. Both answers come from a P&L that aggregates everything into one column.

Here's what attribution per job provides that an income statement structurally cannot.

• Which job types actually pay. A quarter of your time can generate profit, while a 90-minute diagnostic call may be worth only forty dollars if three installation jobs are responsible for the entire result.
• Bid feedback with a number attached. A closed spreadsheet with an ambiguous column shows you exactly how much you didn't know and whether the mistake resulted from hours, materials, or a cost bucket you omitted from the estimate entirely.
• Change-order exposure while it is still fixable. Subreddits for trade contractors describe the entire failure surface: changes made at the wrong time, change orders delivered months after the work was completed, and customers not paying for unexpected circumstances. Scope is moving faster than paperwork. The job cost sheet is where the additional hours show up, and there's plenty of time to record a change order.
• A defensible labor rate.** **It's not the wage. It's the burdened, non-billable-time-adjusted cost of putting a truck in a driveway.
• Estimating memory. The sheets from last year are the only accurate pricing information your shop has.

A customer from an enterprise who was looking over Field Promax on the QuickBooks App Store explained the application's rollout across four of its companies, with plans for more, and the customization options that other out-of-the-box products were unable to match. A costing system that is flexible across different trade companies beats four spreadsheets that do not reconcile at the end of each month.

The canonical formula is short:

Total job cost = direct materials + direct labor + applied overhead

Everything else on a job cost sheet (job number, customer, dates, subcontractor lines, equipment lines, the estimated-versus-actual comparison) is either an identifier or a subdivision of those three.

Direct materials. Everything consumed during the job can be traced back to it: the condenser or line set, the forty-foot length of PEX, and two coupling boxes pulled from the stock of trucks. In a plant, the primary document is a material requisition. In a service shop, it will be the scanner or the selection of parts through the mobile app before leaving the shop. Materials that are not scanned are among the two expenses that silently reduce job margins.

Direct labor. Not the wage rate. BLS OEWS May 2024 data puts median hourly wages at $28.75 for HVAC mechanics and installers, $30.27 for plumbers and pipefitters, $29.98 for electricians, and $18.82 for landscaping and groundskeeping workers (BLS OEWS May 2024). Those are the numbers going onto the labor line, when the number that belongs there is the wage grossed up for the benefit load, then divided across hours you can actually bill. Unlogged travel time is the other silent margin killer, which is why the labor line should be fed by timecard and GPS data rather than a Friday recollection.

Applied overhead. This is where the majority of sheets are faulty, and this is the part that template downloads don't provide. The overhead is calculated, but not tracked, based on a predetermined rate before the start of the period:

Predetermined overhead rate = Estimated total indirect cost for the period ÷ Estimated allocation base for the period

The base for allocation is whatever driver is most effective in tracking how jobs use indirect costs. Machine shops have a base called machine hours. In field service, it is nearly always billable technician hours. A shop with $412,000 in annual indirect costs (rent, office payroll, insurance, software, fixed truck expenses, and non-billable hours) and 18,600 hours of billable technician work would have an overhead rate of approximately $22.15 per billable hour, and each job would be charged based on the amount of time the work takes.

Make the base incorrect, and the rate is incorrect for every job at the same time. Select a base that some jobs do not reach, and those jobs absorb none of the overhead while the rest absorb too much. This represents an allowance, not necessarily a measure, and that's why it should be reviewed each year instead of allowing a decade of inertia.

I read every support ticket that comes in, and the job costing tickets follow one shape. A shop sets overhead at a flat ten percent because someone once told them ten percent is the number, then wonders why the day fills up with short service calls that pay nothing. Ten percent of a $180 drain call is $18. That does not cover the truck, the phone line that booked it, or the office time that invoiced it. The rate that matters is annual indirect cost divided by the billable hours you actually sell, and most shops have never run that division once. Conventional advice says tighten your estimating. I would start one step earlier and fix the burden rate, because a tight estimate built on a wrong overhead rate is just a precise wrong answer.

Field Promax supports roughly 24 service verticals. The generic three-component sheets stop being helpful after 10 minutes of the setup process for the majority of them. A lawn-care routing sheet as well as an HVAC changeout sheet are based on the same formula, and there's not much else.

• Lawn care and landscaping: route time between properties, mower and blower engine hours, fuel, seasonal labor rates that change twice a year
• HVAC installs: equipment cost, permit fees, lift or crane rental, a warranty labor reserve, callback hours booked back to the original job number
• Plumbing service: truck stock consumption, after-hours premium pay, jetter and camera time, disposal and dump fees
• Electrical: takeoff versus wire actually pulled, permit and inspection fees, prevailing-wage splits when public work sits alongside residential
• Pest control: chemical cost per application, license and state reporting time, recurring contract revenue amortized across scheduled stops
• Construction and GC work: schedule-of-values line items, retainage held, subcontractor billing against approved change orders

This is the last area where the biggest debate takes place. General contractors frequently debate how to create a schedule of values, how precise they can make the schedule, and how to charge for the work. There isn't any common agreement, which is why each contractor creates one and then defends their approach to the owner, who would like to split the work differently. The most practical solution isn't glamorous and involves creating the SOV using similar cost codes to those the job cost sheet already uses. If the cost breakdown and the cost codes are both on the same page, the pay application ceases to function as a translation tool.

Two buckets are trade-specific in name only. Vehicle cost and drive time show up in every vertical and get left off the sheet in most of them. The IRS 2025 business standard mileage rate is 70 cents per mile, while AAA's 2025 Your Driving Costs study puts average real ownership and operating cost near 77 cents per mile, with half-ton crew-cab pickups at roughly 98.5 cents. Route-heavy work looks more profitable on the sheet than it is precisely because those two lines are missing.

Job cost sheets fail for structural reasons more often than for mathematical reasons. The responsibility to keep them current belongs to everyone, so they are maintained for three weeks and then neglected. The clearest and best-documented version of this pattern across our customers is the same one we see with review responses: nobody owns the process, and the mechanism is identical.

It's most frequently seen in an owner-operator running an unassuming multi-trade business with fewer than ten technicians covering plumbing, HVAC, and electrical work within a medium-sized metro. Review responses fall into an unnamed seam that nobody claims. The HVAC director assumes that plumbing has handled its own. Plumbing assumes that the office handles the issue. Google notifications go to a shared email inbox that is managed alongside dispatch escalations and parts-vendor emails for days.

At the end of the cooling season, this store had a number of unanswered one- and two-star reviews piled up, with several of them naming the same issue: callbacks involving one furnace installation team. The proprietor was unable to determine which specific complaints were causing problems and which were also generating follow-up revenue.

The solution was to spend 30 minutes each Friday afternoon personally responding to each review from the prior week and logging every review in the job notes to conduct a walkthrough of the root cause during the Monday meeting. Version one didn't work. Friday afternoons were constantly consumed by emergency dispatches at the end of the week or parts collection. In the second month, they shifted the block to 7 a.m. on Mondays, before the meeting. This was a success.

In the course of two years, the tone of the updated reviews changed noticeably and became more positive. Additionally, a couple of customers who had previously been unhappy were able to book again following personal calls, one of whom signed an agreement for maintenance. It was a mixed adoption. The plumbing lead detected root-cause tags quickly. The senior HVAC technician argued that the reviews were unfair without context. Therefore, the owner took on those tasks with him personally instead of having him join the group meeting. The pace was slower and prevented the system from falling apart.

This is a composite case based on the most common version of this pattern among the small multi-trade companies we've worked with since 2019.

The transferable component is the mechanism rather than the review. A recurring data task without a named owner and a scheduled time slot is dead within two weeks, and a job cost spreadsheet is precisely that type of task. It is also important to note that the proprietor was unable to identify which complaints generated follow-on revenue. This is a job-level attribution problem that is unanswerable without a per-job report.

Seven steps, in the order they take place during a week.

1. Open the sheet when the job is created, not when it closes. Customer number, job number, details, scope, promised dates, and the estimated sums divided by bucket. The estimate column is the one that controls. Without it, the sheet becomes just a receipt and not a tool for management.

2. Capture labor at the truck. Clock-in and clock-out per job from the mobile app, including travel legs. Mobile-app adoption by technicians is the single biggest predictor of whether a rollout succeeds, because a sheet fed by end-of-week memory is fiction with decimal places.

3. Log materials at consumption, not at purchase. Parts pulled from the truck's inventory are assigned to the job when the technician records them in the system. A supply-house run for one job is assigned to that job before the receipt is put into the glovebox.

4. Record equipment and vehicle use. Days of rental, hours of operation, or mileage. Rental time for idle equipment still costs money because the task holds the machine.

5. Apply overhead automatically at your rate. Hours consumed during a given period, determined when the timecard is closed. No one does this manually for long.

6. Reconcile what was dispatched, performed, and billed. The most requested feature from our clients is more efficient dispatch-to-invoice automation. The reason is that one of the most significant real-world sources of cost is the gap in the handoff, where additional work does not become an additional line item. Pay applications that come back with an arithmetic or formatting error are not just a minor inconvenience; they can mean another month of waiting. The payroll process isn't waiting for the rest of your time.

7. Close with a variance review. Actual versus estimated costs by bucket, with a one-line explanation for each bucket that is off by more than 10 percent. Incorporate the results into the next quarter's pricing.

Step one is easier when the estimate itself is structured by cost bucket rather than written as a single lump price. If the proposal the customer approves already separates equipment, materials, labor hours, and permits, the estimate column populates itself and the variance review has something to compare against. A one-line quote gives you nothing to measure. This also connects the estimating side to the accounting side, and shops running QuickBooks integration get the closed job cost pushed through without a second data entry pass.

Here’s an illustration using an HVAC shop and a house furnace-and-coil switchout that costs $6,900. The figures represent illustrative math based on the labor-burdened rates from the table above, as well as the $22 billable-hour overhead rate.

Estimated: Equipment: $3,400; other materials: $620; labor: 16 hours at $41.37 ($661.92); permit: $185; overhead applied: 16 hours at $22 ($352). The total estimated cost is $5,218.92, which represents a 24.4% gross profit based on the quoted price.

Actual: Equipment cost: $3,400; materials: $780 (an additional pipe set and fittings); labor: 22.5 hours at $41.37 ($930.83) following a return trip to repair a leaky flare; permit: $185; overhead: 22.5 hours at $22 ($495); and vehicle cost: 74 miles of vehicle cost at the 70 cent IRS rate ($51.80). The total cost is $5,842.63, resulting in an approximate 15.3% gross margin.

That’s a 9-point margin loss in a situation that no one would have considered a major issue: one return trip and one additional line set.

The process for totaling the costs involves four steps and is identical whether you execute it on a spreadsheet or use software:

1. Sum the direct materials consumed on the job, including truck stock.

2. Sum direct labor at the burdened rate, including travel and any additional hours.

3. Include subcontractor invoices, rental or equipment charges, permits, and vehicle costs.

4. Apply overhead at a predetermined rate multiplied by the number of hours actually used for the job, then add everything together.

For a job that covers multiple identical units, divide the total cost by the number of units to calculate the cost per unit. This number will tell you whether the multi-unit installation was priced as a bulk job or as four separate service calls.

The next step is the variance analysis because “we lost nine points” is not actionable, but “we lost 6.5 labor hours” is. The material price variance is calculated as (actual price - standard price) times actual quantity. The labor efficiency variance is calculated as (actual hours - standard hours) times standard rate. The additional 6.5 labor hours at $41.37 represent $268.91 of the variance, and the additional line set accounts for $160. That tells the whole story and points toward the flare joint rather than the estimator.

Five methods separate companies that still have job cost sheets one year later from those that have a missing tab in an office book.

Give the task a name and a slot. The same principle applies to the multi-trade shop mentioned previously. Someone should own the weekly variance check, and it should take place at a set time that isn’t Friday afternoon.

Stop treating receipts as a filing problem. Take a photo and attach it at the time of purchase, then tag it with the job number. A receipt that is delivered to the office without being attached to a job is an expense that you should be able to allocate appropriately; otherwise, you have no way to accurately assign the expense.

Put retainage on the sheet. Retainage causes more friction in conversations with owners than almost any other billing topic. Companies may fail to hold it correctly, clients may hold payments past practical completion, and subcontractors may seek payments that are months overdue. The recurring detail is that no single location clearly shows what is in the account and when it is due to be released. A retainage line with a release trigger makes the amount visible, even while the job is still being completed.

Review the overhead rate annually. Rent, insurance, and non-billable hours can shift. A cost structure established three years ago may apply yesterday’s costs to today’s work.

Push the closed sheets into reporting, not into storage. Sorted by job type, the sheets tell you which service categories to reprice, which crews need a process fix, and which equipment is earning its keep. Standing reports and dashboards beat a quarterly spreadsheet excavation.

A long-time user of this platform through the QuickBooks App Store mentioned that requested changes to business-specific requirements were created specifically for users. Job costing is where this flexibility makes money, since most businesses require different cost buckets, and no standard template comes with fields for permit fees, disposal charges, and callback labor that gets charged to no one.

Should the sheet be filled in during the job or after it closes?

During. An after-the-fact sheet is compiled using receipts that may have been lost and a number of hours that no one remembers. By the time it is completed, the invoice may have already been paid based on whatever amount the estimate stated. Only the live option allows you to identify and address deviations from the estimate.

Is Excel good enough?

For a small business with more than three jobs per week, with one person handling all the data, it is possible. However, it is not enough when technicians are the ones generating the information, since spreadsheets are unable to reliably capture a timestamp from a driveway or a part number from a truck. The problem usually appears as a debate about Friday’s hours rather than as an obvious spreadsheet error.

Who should own the job cost sheet?

The person who has visibility into every aspect of the process is usually the office manager or operations manager responsible for dispatch and invoicing. The bookkeeper can help move the information to the next stage, where it can be reconciled and finalized.

Continue with:

• Job Costing for Small Commercial Contractors: Phases and Margin — the pillar guide: phase structure, committed cost, scheduled vs actual, pay applications, and honest margin.
• Labor Burden: What an Hour of Field Labor Actually Costs You — the burden buckets, the truck-and-tool hourly adder, and the billable-hours denominator that sets the labor line on this sheet.