Fractional CFO and Controller Services for Construction Companies

Construction is the only industry where you finance your customer's project with your own payroll. Labor goes out every Friday. The pay application goes in on the 25th, gets approved on the 15th of the next month if the architect is prompt, and pays 30 days after that if the owner is. Add 10 percent retainage held until substantial completion, and a healthy job can run cash negative for five months before it turns. Multiply by six jobs and the line of credit is not a safety net. It is the business model.

The forecast that manages this is not a budget. It is a weekly schedule of cash receipts by job and by pay application, set against disbursements by payroll date, by subcontractor payment date, and by supplier due date. You update it every Monday. When a general contractor slides a payment two weeks, you know on Tuesday which supplier gets a phone call and which sub payment moves. Most contractors run this in their head. It works until the head is on vacation.

Cash does not follow revenue in construction. It follows the pay application, the retainage release, and the supplier's due date. Forecast those, not the P&L.

Your estimator built the bid by cost code. Your PMs think in phases. Your accounting system posts costs to whatever code the invoice approver picked on a Tuesday afternoon. When those three views disagree, the estimate-versus-actual report becomes wallpaper, and your PMs go back to managing from memory. The job still gets built. It just gets built without anyone knowing, until closeout, whether the labor on the second floor came in at the number the estimator promised.

Good job costing starts with the cost code structure and ends with committed costs. Every subcontract and purchase order is entered at award, not at invoice, so the cost-to-complete report shows what you have promised to spend rather than what has happened to arrive in the mail. Labor posts daily from the field, burdened at a rate someone has actually calculated. Change orders live in the job from the day they are priced, flagged pending, so the PM sees the exposure before the owner's representative does.

Whether the system is Sage 300 CRE, Foundation, Procore with a QuickBooks back end, or QuickBooks alone with a job list that has become a filing cabinet, the discipline is the same. The software matters less than the person who insists the field uses it. We have run this in Timberline for two decades, and we are not precious about the platform.

An over-billed job is not a win. It is a liability on your balance sheet called billings in excess of costs and estimated earnings, and it represents cash you have collected for work you have not yet done. Your surety knows this. An under-billed job is worse, because it usually means one of two things: the billing is behind, or the estimate is wrong and the profit you have been booking is not there.

A monthly WIP schedule lists every open contract with contract value, estimated cost, cost to date, percent complete, revenue earned, billings to date, and the over or under position. Done right, it is the single most honest document in the company. It exposes profit fade job by job, and it forces a cost-to-complete conversation with each PM every month rather than at closeout, when the only thing left to do is explain.

The mechanics are not hard. The discipline is. The WIP is the report that turns the field's opinions into numbers a lender can underwrite, and it is the first thing we build. Once it exists, it gets produced every month whether or not anyone outside the company has asked for it, because the month nobody asks is usually the month something changed.

Under-billings are the earliest warning you will get that a job is going sideways. Read them before the PM reads them to you.

Your bonding agent wants three things every quarter: a WIP schedule that reconciles to the financial statements, working capital that supports the backlog you are asking to bond, and gross profit that does not fade between the estimate and closeout. Your bank wants the same, plus a borrowing base that treats retainage receivable with appropriate suspicion and a debt service coverage covenant you are not going to trip in the slow quarter.

A buyer, whether a private equity group or a strategic acquirer, wants to see that the profit is repeatable: consistent margin by job type, a backlog with margin in it, customer concentration that does not depend on one developer's mood, and add-backs that survive a quality of earnings review. We have sat on the company side of that review, and the difference between a clean WIP history and a reconstructed one shows up directly in the price.

The way to be ready is to produce these documents monthly, not when someone asks. The bank meeting stops being a scramble and becomes a status update. The surety renewal becomes a conversation about growth rather than a defense of last year. And when a buyer's letter arrives, the answer to their first request is already sitting in a folder with 24 months of history behind it.

Most contractors between $3M and $50M run some combination of a construction ERP, a project management platform, a payroll service, and a spreadsheet nobody admits to. Procore knows the change orders. Sage or Foundation knows the costs. Payroll knows the hours. QuickBooks, if that is the general ledger, knows what the bookkeeper got to. Each is right about its piece, and the month-end close is where the differences surface.

The fix is a close checklist with owners and deadlines: labor posted and burdened by Wednesday, AP cut off and accrued, subcontract commitments updated, billings matched to schedules of values, WIP calculated, and the general ledger tied to the job cost ledger to the penny before anyone prints a P&L. Where an integration exists we use it. Where it does not, we build a controlled import rather than a retyping exercise.

The goal is a close that finishes by the 10th business day with numbers you would show your surety. Once the systems reconcile, the reports become believable, and believable reports are what change behavior in the field. A PM who has watched the job cost report be wrong three times will ignore it forever. One who has watched it be right will start reading it on Monday before you ask.

You do not need a 40-page binder. You need a one-page dashboard and four schedules: the WIP, the cash forecast, an AR aging that separates retainage from current billings, and a backlog report with gross profit remaining by job. Below that sits a P&L by division or job type, so you can see whether the service work is subsidizing the bid work or the other way around.

The package is reviewed in a monthly project meeting with the PMs, where each open job gets its cost-to-complete challenged in the open. That meeting is where profit fade stops, because a PM who knows the number will be questioned tends to find the change order that was never submitted. This is not accounting. It is management, and finance runs the agenda.

Every schedule in the package is documented so a successor can produce it. Fix it, run it, document it, hand it off. That is the whole engagement in one line. The package does not depend on us being in the building, on a particular spreadsheet, or on one person's memory. It depends on a written process, a close calendar, and a set of reports your team has produced with us enough times to own.