The problem job costing solves
A construction P&L tells you the company made money last month. It does not tell you which jobs made it, and that is the only question that changes what you bid next.
Job costing assigns every dollar of cost to the job that caused it, so margin can be read per job rather than per month. Without it, a profitable quarter can hide two jobs bleeding badly and one unusually good one, and you will keep bidding all three the same way.
The lag that hurts
Most contractors discover a bad job at closeout, when nothing can be done. Job costing that is current within a week turns that into a mid-job signal you can still act on: re-sequence, re-scope, or file the change order you have been putting off.
Setting up the cost structure
The chart of accounts and the cost code structure are separate things and should stay that way. Accounts answer what kind of cost it was. Cost codes answer which part of the job it belonged to.
- Labor: burdened, not raw wage. Include payroll taxes, workers compensation, and benefits or the margin is fiction.
- Materials: assigned to the job at delivery, not at purchase, or timing distorts the job.
- Subcontractors: tracked against the sub contract value so you can see committed cost, not just spent cost.
- Equipment: owned equipment charged to jobs at an internal rate; rented equipment charged directly.
- Other direct: permits, bonds, dumpsters, temporary utilities.
Committed cost is the number that matters
Spent-to-date understates where a job actually stands, because it excludes purchase orders written and subcontracts signed but not yet invoiced.
Committed cost is spent plus outstanding commitments. Compare that against the budget line and you get an honest projection while the job is still running. Contractors who only watch spent-to-date routinely discover a blown line the week the invoices land.
Percentage of completion, without the jargon
For contracts spanning more than one period, revenue is recognised in proportion to progress rather than when you invoice. The most common measure is the cost-to-cost method: costs incurred divided by total estimated costs.
The estimate in that denominator is the whole ball game. If your estimate of total cost is stale, your recognised revenue is wrong, and the correction lands as an ugly swing in the period somebody finally updates it.
- A current estimate of total cost at completion for every open job, updated monthly
- A WIP schedule reconciling contract value, costs to date, estimated cost to complete, and billings
- Discipline about change orders: unapproved scope in the cost column with no matching contract value distorts every ratio
- A review of over and under billings, which is where cash and revenue diverge
Over and under billing, plainly
If you have billed more than the work completed, you are overbilled. That is a liability, not profit, and it flatters cash today at the cost of a lean stretch later in the job.
If you have completed more than you billed, you are underbilled. That is an asset and usually a billing process problem worth fixing this week, because it is your own cash sitting in somebody else account.
Sureties and lenders read the WIP schedule specifically for this. Consistent large underbillings read as weak billing discipline; consistent large overbillings read as borrowing from future jobs.
What good looks like
- Job cost reports current within one week, not one month
- Committed cost tracked, not just spent
- Estimate at completion refreshed monthly on every open job
- A WIP schedule produced with the monthly close, not annually for the surety
- Change orders logged when raised, not when approved, with status visible