Fractional CFO for Construction: Cash Flow Control

Reading Time: 22 minutes
Material delivery tickets and a notebook on a folding table inside a job site trailer

One contractor's December looked like a 19% month against a year that ran between 44% and 50%. Nothing had gone wrong in December.

Key Takeaways

  • A month that looks bad may just be the month a full year of depreciation was booked into.
  • You buy material in one period and consume it in another, and if the cost hits on purchase, every job in between reads wrong.
  • Work performed and work billable are different dates, and treating them as the same one turns your receivables report into fiction.
  • You cannot forecast cash off a P&L whose months do not reflect what happened in them.

Why Does One Month Look Terrible When Nothing Went Wrong in It?

Because a cost that belongs across twelve months got recorded in one of them. A specialty contractor doing close to $29M booked roughly $1.2M of depreciation once, in December. Every other month that year carried none of it. December read at 19% while the rest of the year sat between 44% and 50%.

The owner's instinct in that situation is to go looking for what broke in December. Nothing did. The fix was to book depreciation monthly and move it below the EBITDA line so the operating months could be compared to each other honestly.

This is worth naming plainly because it is the cheapest problem on this list to solve and one of the most common. It costs nothing but a change in how one entry is made, and it removes an entire category of false alarm from the reporting.

How Do Material Costs Land in the Wrong Period on a Construction Job?

When they are expensed at purchase rather than when they are consumed on a job. The same specialty contractor was expensing explosives as they were bought. That ran to $1,000,009 year to date, roughly 16% of revenue, with approximately $200,000 sitting in the wrong place and needing to move to inventory.

The consequence is not an accounting technicality. Every job that used material bought in a prior month showed a cost that was not its own, and every job that stocked material for future work absorbed cost it had not yet consumed. Job-level margin was unreliable in both directions at once.

The correction was mechanical. Reconcile usage back through the year against the blast records that logged it, move unused purchases to the balance sheet, expense consumption monthly against the job that used it, count inventory weekly, and reconcile the difference once a month. That contractor also carried $151,000 of year-to-date payroll allocated to no job at all, part of it crews being carried between phases of work.

Take the Financial Control Score Quiz 7 questions. About a minute. See where your business actually stands.

Why Are My Receivables Wrong When Every Invoice Was Entered Correctly?

Because work performed and work billable happen on different dates, and entering the first as if it were the second creates receivables that do not exist yet. The contractor above recorded work daily as it was completed, but pay applications could not be submitted until the 20th of the month. Entered-but-unsent invoices had been sitting in accounts receivable the entire time.

That produces a receivables balance nobody can collect against and a cash forecast built on it. The correction was to treat performed-not-yet-billable work as accrued revenue on the balance sheet, and only create a receivable when the pay application actually goes out. They renamed the line unbilled revenue rather than underbillings, deliberately, so it would not trigger a lender request for a full work-in-progress schedule before they were ready to produce one.

The scale of what this hides is easy to underestimate. A service contractor doing $7M to $7.5M found unbilled labor running somewhere between 9,000 and 12,000 hours a year. At their $75 hourly charge rate, the low end of that range is about $675,000 of work performed and never invoiced.

What Happens When Someone Edits a Closed Month?

Everything downstream of it stops tying, and usually nobody finds out for weeks. A development and construction group we work with had a bookkeeper making corrections back to January in periods that had already closed. At one point the books were off by $160,000, and it took an hour and a half just to find out why.

One specific instance shows the mechanism cleanly. An invoice dated July was entered when the payment cleared in October. Under accrual accounting that changes July, a month that had been closed, reported, and used for decisions. The fix was a rule rather than a conversation: the bookkeeper books only clearly project-specific expenses, anything ambiguous stays in the register for reconciliation, and closed periods do not reopen.

The same group's percent-complete line was adding $454,000 of income with no attributable expense against it, which is the same period-matching failure wearing different clothes. Revenue recognized without its cost is a month that looks better than it was, followed by a month that looks worse.

What Does a Fractional CFO Change About Construction Cash Flow?

It puts someone in the seat whose job is to ask whether the number landed in the right month before anyone forecasts off it. Every problem above was found by looking, and none of them would have surfaced in a normal close.

The pattern across all three contractors is the same: capable people, functioning systems, real work getting done, and a reporting layer that nobody owned end to end. The service contractor was mid-conversion from cash to accrual and described it as not happening quickly enough, while the owner said plainly that he could not see where he was. Our fractional CFO for construction companies work usually starts by making the months comparable before anyone tries to build a forecast on top of them.

Order matters here. A thirteen-week cash forecast built on a P&L with a year of depreciation in one month and receivables that were never billable is a confident-looking document that will be wrong. Fix the periods first.

What Should You Check Before Building a Cash Forecast?

Check whether last month is comparable to the month before it. If it is not, a thirteen-week forecast built on top will be confidently wrong, and it will be wrong in a direction nobody can predict.

The specialty contractor above ran that check and found three distinct problems in one pass: annual depreciation compressed into a single month, material expensed on purchase rather than consumption, and performed-but-unbillable work sitting in receivables. Each one on its own would distort a forecast. Together they made the monthly trend unreadable.

There is a sequencing point buried in how they fixed it. They deferred moving their accounting system over to a new platform for two to three weeks specifically because a transaction was still unsettled and no opening balance sheet existed yet. Migrating first would have carried every unresolved question into the new system and made it permanent. They also explicitly deprioritized the thirteen-week cash forecast during that window, which is counterintuitive advice from a finance team and exactly right.

The same instinct showed up in how they treated their reporting rebuild. Rather than reconstructing the prior twelve months to match a new structure, they set a forward date after which everything would use the new setup and accepted a variance in the range of five to ten percent on the historical comparison. Rebuilding a year of history is a project that never finishes. Drawing a line and moving forward is a decision that takes an afternoon.

Your Next Step

If Your Bad Month Has No Explanation, It Probably Has an Entry

Before you diagnose an operating problem, check whether the month is telling you the truth. The Financial Control Score Quiz scores cash visibility as its first dimension, and period-matching is usually where the score comes apart.

Common Questions

Frequently Asked Questions

What Do Fractional CFO Services for Construction Companies Actually Include?

At minimum: making reporting periods reflect the work that happened in them, building cash forecasting on top of that, and owning job-level profitability. That means monthly depreciation rather than annual, material expensed on consumption, and revenue recognized when it is genuinely billable.

Why Is My Construction Company Profitable but Short on Cash?

Usually because profit and cash are measured on different timelines. Work performed months ago may not be billable yet, material bought this month may belong to next month's jobs, and receivables may include invoices that were entered but never sent.

How Should Construction Companies Handle Materials Bought in Advance?

Carry unused material as inventory on the balance sheet and expense it to the job when it is consumed. Expensing at purchase distorts every job in both directions: jobs using older stock absorb cost that is not theirs, and jobs stocking ahead absorb cost they have not used.

What Is Unbilled Revenue in Construction Accounting?

It is work performed that cannot yet be invoiced, usually because a pay application window has not opened. Recording it as accrued revenue keeps the PandL honest while keeping it out of accounts receivable, where it would inflate a balance nobody can collect against.

Should a Construction Company Use Cash or Accrual Accounting?

Accrual, if you want to make decisions from the numbers. Cash basis obscures job profitability by design, because it records money movement rather than work performed. Contractors converting mid-year should expect the transition to surface problems rather than create them.