A slow close is rarely caused by the closing itself. It is caused by work that could have happened during the month being deferred to the days after it ends — then competing for the same people, in the same week, against a deadline.
Compressing the close to five working days is mostly a sequencing problem. Here is the structure we implement with clients.
Move work out of the close and into the month
Anything that does not depend on the month being over should not wait for the month to be over. Supplier invoices posted as they arrive, not batched. Bank transactions reconciled weekly. Payroll inputs agreed before the run, not reconciled after it. Recurring journals — depreciation, prepayment amortisation, lease unwinds — scheduled in advance with the amounts already calculated.
In most teams this alone removes two to four days, because it converts discovery work into confirmation work.
Set a hard cut-off and defend it
A cut-off that moves is not a cut-off. Fix the date the ledger stops accepting entries for the period, communicate it to every department that feeds it, and hold it. Late items go to the following period unless they are material — and “material” needs a number attached, agreed in advance, not a judgement made under pressure by whoever happens to be asked.
The five-day sequence
Day 1 — Close the subledgers. Sales, purchases, payroll, inventory movements and cash all stop and reconcile to the general ledger. Nothing further is posted to them for the period.
Day 2 — Accruals and provisions. Goods received not invoiced, accrued expenses, bonus and leave provisions, expected credit losses. Each on a standing schedule with a named owner, so the question is “has this changed” rather than “what is this”.
Day 3 — Balance sheet reconciliations. Every balance sheet account reconciled to supporting evidence, not to itself. Prepared and reviewed by different people, with differences either cleared or explained in writing the same day.
Day 4 — Review and analytics. The financial controller reviews the trial balance against budget and against the prior month, and asks about the variances before the pack is built. Most closing errors are found here, by someone who knows what the numbers ought to look like.
Day 5 — The pack. Statements, variance commentary, the KPIs management actually uses, and a short list of items needing a decision. Issued — not “nearly ready”.
What makes it hold
A written close calendar. Every task, owner, day and dependency. Without it the close runs on memory, and memory is the first thing to fail when someone is on leave.
Standing schedules rather than fresh spreadsheets. A provision computed in a new workbook each month is a new opportunity for error and an old one for inconsistency.
Review separate from preparation. Self-reviewed work is not reviewed. This matters for accuracy, and it matters again at audit, when the auditor asks who checked it.
An exceptions log. Anything that delayed the close is written down with its cause. Reading the log after three months tells you precisely which two or three things to fix.
What it is worth
Management receives numbers while they can still act on them. Auditors arrive to find reconciliations already prepared and reviewed, which shortens fieldwork and reduces adjustments. And the finance team spends the second half of each month on analysis rather than assembly — the part of the job that produces value, and the part that keeps people.
We implement this as a fixed exercise: we map the current close, rebuild the calendar and the standing schedules, and run alongside the team for two cycles until it holds without us.