Halfway through a job, the report looks great. Materials are in. The estimate is holding. If you pulled the numbers today, you’d feel good about this one.
Then the job finishes, and somewhere between the last week of work and the final numbers, the margin gets thinner. Sometimes a lot thinner. Nobody can quite say when it happened.
Most of the time, I’d bet on labor.
Why labor is the one that hides
Materials leave a trail. There’s a receipt, a bill, a supplier invoice. Even if it gets coded to the wrong job, it exists somewhere.
Labor is different. Labor gets captured for payroll. Payroll cares about one question: how many hours did this person work, so we can pay them? It does not care which job those hours belonged to. So unless something forces the job and the type of work to come along with the hours, they show up as a company cost and never make it to the project.
The accounting can be right and the job report can still be wrong. The number may not be wrong. It may be incomplete.
Tracing the trip
Here’s the path labor has to take to show up on a job, and where it usually falls off:
- The crew works. Real hours, on a real job.
- Time gets recorded. On an app, a paper sheet, a text to the office, or memory on Friday. If the job and the type of work aren’t captured here, they’re already gone. You can’t add them back accurately later.
- Time goes to payroll. Usually this part works, because people care a lot about getting paid.
- Payroll goes to accounting. Often as one lump wage expense. The job link, if it existed, can get dropped right here.
- The job report pulls costs. It shows what’s tied to the job. Labor that never got tied doesn’t show. So the job looks better than it is.
- Someone uses a labor rate. If it’s just the hourly wage, the report still understates the cost, because it’s missing payroll taxes, workers’ comp and other burden.
Any one of those breaks makes the job look more profitable than it was, until the end, when it’s too late to do anything about it.
What the owner thinks is wrong
Usually, nothing, until the end. Then the guesses start: the crew was slow, the estimate was off, the customer was difficult. Some of those might be true. But if labor wasn’t reliably tied to the job, you can’t tell which. You’re arguing about a number that was never complete.
What I’d investigate
I’d take one finished job and follow the labor. Who worked on it? How many hours does payroll say they worked that week? How many of those hours landed on this job in the job report? What rate was used?
The gap between those answers tells you which step is broken.
What I’d change
- Capture time by job and cost code at the source. At clock-in or clock-out, in whatever tool the crew will actually use. Simple enough that a crew can do it in seconds.
- Use a burdened labor rate, so the job carries the real cost of an hour, not just the wage.
- Make sure the job link survives the trip from time tracking to payroll to accounting.
- Look at labor against budget weekly while the job is still running. That’s the whole point. Know where each job stands while it’s still running, not after.
What I’d leave alone
I wouldn’t automatically change the payroll provider, the estimating method, or the accounting software. Most of the time, the tools can do this. They’re just not set up to, or the habit at the source isn’t there.
The tradeoff
Crews do a little more at clock-in. Someone in the office has to review labor every week and follow up when something’s missing. That’s the cost. In exchange, you find out a job is going sideways while you can still do something about it.
The bigger point
Job profitability without reliable labor is a guess with a spreadsheet around it. Your gut can tell you a job feels off. A system can show you where.
This is the core of what Job Cost Ready is built for: getting labor, materials, subs and change orders to land on the right job, so the numbers mean something while the job is still open. If you want to start smaller, pick one finished job this week and follow the labor.
