Closed Is a Claim, Not a State

A contractor reopened a month after it had already been declared closed.

At a second company, the leadership team sat down to review financials for a month nobody could confirm had finished closing. Both statements were produced by competent people. In neither case did closed mean the same thing to the person saying it and the person hearing it.

What Does It Mean for a Construction Month to Be Closed?

It means a named person has completed a defined list of steps, confirmed the result, and committed that no further entries will be posted to that period. Anything less is a status update. If closed is a feeling rather than a checklist with a signature, the word carries no information.

The practical test is whether anyone can tell you the date. Not the target date — the date it actually happened last month, and the month before. Most contractors at this size cannot, which is itself the finding.

This matters more in construction than in most industries because so many costs arrive late by nature: subcontractor invoices, equipment charges, material tickets reconciled after the fact, retainage adjustments. Every one is a reason to keep a period open, and keeping a period open indefinitely is the same as never closing it.

Why Do Months Get Reopened After Someone Says They Are Closed?

Because something material shows up late and reopening feels more accurate than accruing. It is more accurate, and it is also the decision that makes every prior month unreliable, because a number that can change after the fact is a number nobody can safely build on.

At one construction client, a month was reopened after it had been declared closed, which meant a review already underway was working from figures that then moved. At another, the July financials were reviewed while the team was still uncertain whether the close had finished — the review went ahead anyway on the basis that the direction was useful even if the detail might shift.

Both of those are reasonable decisions in the moment. Made repeatedly, they train everyone to treat the financials as provisional, and provisional financials do not drive decisions.

What Arrives After the Close, and What Do You Do About It?

Subcontractor invoices, mostly, and they arrive late for structural reasons rather than careless ones. The answer is to accrue against the period the work belongs to and reverse when the invoice lands, rather than to hold the period open or push the cost forward.

At one construction client, a subcontractor had gone roughly seven weeks without submitting — two weeks of one month plus all of the next — against work running in the neighbourhood of fifty to sixty thousand dollars. The work was real and the cost was known approximately. The decision was to accrue it and reverse when the invoice arrived, alongside a standing rule that no payment goes out without an invoice, and that payment would be staged rather than released in one lump once the backlog cleared. Two vendor invoices in the same period had arrived late through a general inbox rather than the normal route, one of them only found because somebody went looking.

The accrual is not an accounting nicety. It is the only way the month the work happened in ever shows the cost of that work.

How Long Should a Contractor's Close Take?

Long enough to be right and short enough to be useful, which in practice means statements in hand within about two weeks of period end. Past that, the decisions the numbers would have informed have already been made without them.

The constraint is usually not effort. It is sequence — the close waits on final payroll, which waits on approvals, which wait on someone who is on a job site. A closing calendar that names each step, its owner and its deadline turns a vague two-to-three-week range into a date.

A third construction company we spoke with this year described exactly what it wanted and did not have: a systematic monthly cycle with weekly, biweekly and monthly tasks that culminate in a close landing on a known date. What it had instead was approval bottlenecks that produced late payables, which is the same problem viewed from the operations side rather than the accounting side. A close does not slip because accounting is slow. It slips because the approvals feeding it are.

The other common constraint is that part of the sequence sits with someone outside the company. At one client the close depended on work from an external bookkeeper where promised next-day turnaround routinely arrived late with no status update in between, which meant the internal team could not plan around it. A closing calendar only works if the external steps are on it with the same named deadlines as the internal ones.

One judgement call worth making explicitly: at what dollar value do you stop waiting? At one client the answer landed at roughly a thousand to two thousand dollars — below that, the correction goes to the next period rather than delaying statements another fortnight. Setting that threshold in advance removes the argument.

What Is the Difference Between a Late Close and an Unreliable One?

A late close produces numbers you can trust after the moment you needed them. An unreliable close produces numbers you cannot trust at any point. The first is an operational problem; the second is a decision-making problem, and it is much more expensive.

The clearest symptom of the second is that nobody reviews the output. At one construction client the financials were treated as closed and finalised without anyone stepping through them first, and a request for a closing calendar sat unanswered while the process continued as it always had. The numbers were not wrong so much as unexamined, which produces the same result.

Add a review step before anything is finalised and the character of the close changes. It is the cheapest control available and the one most often missing. Our page on construction-specific CFO work covers how that runs alongside job costing.

When Should You Refuse to Reopen a Month?

Almost always, once a closing calendar exists and a materiality threshold has been set. The discipline is what makes the prior months mean anything.

At one client, an error that had likely landed in the prior month was handled by leaving that month alone and catching the correction through the next physical inventory instead. That is the right instinct: correct it forward, disclose it, and keep the closed period closed.

The exception is a material misstatement that would change a decision — a lender covenant, a bonding calculation, a distribution. Those are worth reopening for, and they are rare. Everything else is a preference for tidiness that costs you the reliability of every period behind it.

Where Does Backbone CFO Start?

With the calendar, because it is the artefact that turns close from a habit into a process. Named steps, named owners, named deadlines, a review step before finalisation, and a materiality threshold that decides what waits and what goes to next period.

Then we look at what is arriving late and why. Late subcontractor invoices usually trace to a submission expectation nobody set rather than to the subcontractor, and the fix is a stated deadline with a consequence, not more chasing.

We also look at what the close is being used for. At one client the reporting pack was deliberately held until the close tasks were confirmed finished, rather than issued early and corrected later. That sequence is worth protecting: a pack that goes out before the period is settled trains its readers to wait for the second version, and once they are waiting for the second version the first one has no purpose.

What we do not do first is change systems. A close that is unreliable in one accounting package will be unreliable in the next one, because the problem is almost never the software — it is that nobody owns the date.