How do you do job costing for a service business?
Capture five inputs against every job: labor hours per employee, a fully loaded labor rate that includes burden and non-billable time, materials at landed cost, an equipment or vehicle allocation, and any subcontractor spend. Then run gross margin per job = invoiced amount minus (labor + materials + allocated overhead), divided by the invoiced amount. Review the closed jobs weekly, flag anything under 20% gross margin, and ask whether it was priced wrong or executed wrong.
Ask a service business owner which jobs made money last month, and most will name their two biggest invoices. Ask them which jobs lost money, and you'll get a shrug. That gap is where margin quietly dies.
The five inputs a job costing system needs
- Labor hours per employee, tied to the specific job
- Fully-loaded labor rate (wage + burden + benefits + non-billable time)
- Materials and parts used, at true landed cost
- Equipment and vehicle allocation (per-hour or per-job)
- Any subcontractor spend tied to the job
Miss any one of these and the number is a guess. Get all five and you have a decision-grade profitability figure per job.
The formula you actually run
Gross margin per job = invoiced amount minus (labor cost + material cost + allocated overhead). Divide by invoiced amount for margin percentage. Sort your last 90 days of closed jobs by that percentage.
What the sorted list tells you
The bottom quintile is almost always the same shape: undersized jobs booked at the standard rate, jobs that required a second visit, or work for a customer whose scope creep isn't priced in. The top quintile is usually the opposite: right-sized work, first-time-right, and clean scopes.
Run it weekly, not monthly
Monthly job-cost reviews are hindsight. Weekly reviews are steering. Every Friday, pull the week's closed jobs, flag anything below 20% gross margin, and ask two questions: was this priced wrong, or executed wrong? Different answers lead to different fixes, either pricing changes or training.
"The week we started flagging sub-20% jobs on Friday, we found a single customer eating a quarter of our labor budget. We re-priced them the following Monday."
Where the software matters
Job costing is only as good as the labor and material data flowing into it. If techs are logging hours from memory at the end of the week, your job costs are fiction. Real-time clock-in on the mobile app, materials scanned or logged at the truck, and a fully-loaded rate maintained centrally. That is the setup that makes weekly reviews possible.
Job costing is not a report. It's a habit. The service businesses that grow past their first plateau are almost always the ones that made this habit weekly, non-negotiable, and boring.
Common questions
- What is a good gross margin on a service job?
- Most field service businesses target 45% to 55% gross margin on labor-driven work before overhead, and treat anything under 20% as a job that needs a reason. The exact target depends on your trade and overhead load, which is why the useful exercise is ranking your own closed jobs rather than chasing an industry number.
- What is a fully loaded labor rate?
- It is the real hourly cost of putting a person on a job: base wage plus payroll taxes, workers compensation, benefits, paid time off, training, and the non-billable hours they are paid for. It is commonly 25% to 45% above the base wage. Costing jobs at the base wage is the most common reason a business shows profit on paper and not in the bank.
- How often should we review job costs?
- Weekly. Monthly reviews are hindsight; weekly reviews are steering. Every Friday pull the jobs that closed that week, sort by margin, and look at the bottom of the list while the crew still remembers what happened on those jobs.
- Can we do job costing without software?
- You can, but only if hours and materials are captured at the time they happen. Job costing built on timesheets reconstructed on Friday is fiction, because the errors all run in the same direction. The software matters less than the capture point.