Ask a finance team why the close takes three weeks and you will hear about volume, systems and headcount. Watch the close instead and you will usually see something different: three days of waiting for supplier invoices, a bank reconciliation left until the end, a revenue cut-off argument that recurs every month without ever being resolved, and a management pack rebuilt by hand in a spreadsheet because nobody trusts the report in the ledger.
What a slow close actually costs
The obvious cost is decision latency. If March numbers arrive on 22 April, every decision made in the first three weeks of April was made on instinct. For a business changing quickly, that is most of the decisions that matter.
The less obvious cost is that a slow close crowds out everything else. A team spending fifteen working days on month-end has five left for forecasting, analysis, business partnering and controls. That is why finance functions get stuck: not because the people cannot do the higher-value work, but because the calendar never lets them start.
And the compounding cost: a business that cannot close quickly cannot forecast credibly, which means it cannot raise efficiently, which means it does the whole thing under time pressure with worse information.
Most of the work happens before day one
The single largest shift is psychological. A five-day close is not a fast three-week close. It is a close where most of the work was already done during the month.
Anything that can be done before period end must be done before period end. If a task is only started on day one, it is already a bottleneck.
- Reconcile daily, not monthly. Bank reconciliation performed every day takes minutes and is current at period end. Left to month-end it becomes a multi-day archaeology exercise.
- Code as you go. Invoices captured, coded and approved on receipt. An unapproved invoice sitting in someone's inbox on day two of close is a self-inflicted delay.
- Standing journals prepared in advance. Depreciation, amortisation, prepayment releases, lease interest — schedules built once and released automatically.
- Cut-off communicated and enforced. Every budget holder knows the date by which accruals must be submitted, and the close does not wait past it.
- Balance sheet reconciled monthly at pace. A reconciliation performed on day three every month never accumulates. One left for the year-end becomes a project.
The five-day sequence
Sequenced by dependency, not by department. The rule is that nothing waits on something that could have run in parallel.
| Day | Work | Depends on | Complete when |
|---|---|---|---|
| Day 1 | Close subledgers. Final bank reconciliation. Payroll journal posted. Revenue cut-off applied. Accruals from the pre-agreed list posted. | Cut-off enforced at period end | Subledgers agree to the general ledger |
| Day 2 | Standing journals released. Prepayments and deferred revenue rolled. FX revaluation run. Intercompany matched and eliminated. | Day 1 | Trial balance is stable, no further postings expected |
| Day 3 | Balance sheet reconciled in full, every line supported by a schedule. Exceptions listed with owners and dates. | Day 2 | Every balance is substantiated or explicitly flagged |
| Day 4 | Management pack drafted. Variance analysis against budget and prior period. Commentary written. Consolidation run for groups. | Day 3 | Pack is complete and internally consistent |
| Day 5 | Chartered Accountant review, challenge and sign-off. Distribution. Close file locked. Post-close review of what slipped. | Day 4 | Signed, issued, period locked |
Two details make this hold. First, the period is locked on day five — no post-close postings without an approved exception. A ledger that stays open invites late adjustments, and late adjustments mean the pack you circulated is no longer true. Second, day five includes a short post-close review: what slipped, why, and what changes before next month. Without it, the same bottleneck recurs indefinitely.
The substantiation standard
This is the part most teams skip, and it is the part that determines whether your first audit is expensive.
Every balance sheet line must be supported by a schedule that agrees to the ledger, prepared by a named person and reviewed by another. Not a plausible explanation — a schedule that ties.
- Cash — bank reconciliation with no unexplained items over a month old.
- Receivables — aged listing agreeing to the control account, plus a provision assessment.
- Inventory — quantity and valuation with the costing basis documented.
- Prepayments — a schedule showing amount, period and monthly release.
- Fixed assets — register agreeing to the ledger, with additions and disposals traced.
- Payables and accruals — listing agreeing to the control account, with each material accrual's basis noted.
- Deferred revenue — contract-by-contract release schedule.
- Tax accounts — GST and PAYE reconciled to filed returns.
- Debt — agreeing to the lender statement, including accrued interest and fees.
The discipline is simple and it is what a first audit costs you most for when it is absent. If you can hand an auditor a substantiation pack on day one of fieldwork, the audit is a review. If they have to build the schedules themselves, it is a reconstruction — and you pay for every hour.
Five bottlenecks, and how to remove each
1. Waiting for supplier invoices
The most common single cause. You cannot control when a supplier invoices, so stop trying. Maintain a standing accrual list for recurring suppliers based on purchase orders and known run-rates, post it on day one, and true it up next month. Precision on day twelve is worth less than a reasonable estimate on day one.
2. Revenue cut-off arguments
If the same recognition question recurs every month, the problem is not the month — it is that the policy has never been written down. Document the treatment for each contract type once, get it agreed, and apply it. The debate should happen annually when a new contract type appears, not monthly.
3. Bank reconciliation left to the end
Reconciling thirty days of transactions in one sitting is slow and error-prone. Reconcile daily. With automated bank feeds and rule-based matching this is minutes a day, and the period-end position is already clean.
4. Rebuilding the pack by hand
If the management pack is assembled by exporting to a spreadsheet and re-formatting, you have added two days and introduced a category of error that is very hard to detect. Fix the underlying reporting so the pack generates from the ledger. If the pack cannot be generated, the chart of accounts is usually the real problem.
5. Sequential review
A reviewer who only looks at anything on day five will find issues that require reopening earlier work. Review continuously: day one work reviewed on day two, day two on day three. The day-five review is then a challenge of judgement and narrative, not a hunt for errors.
What to automate first
In order of return, measured as hours saved per unit of implementation effort.
- Bank feeds with rule-based matching. Highest return, lowest effort, and it enables daily reconciliation.
- Invoice capture and coding. Extraction plus a coding suggestion against your chart, with a human confirming anything below the confidence threshold.
- Approval routing. Rules-based, with escalation on delay. Removes the "waiting on an inbox" failure mode entirely.
- Standing journals. Depreciation, prepayments, accruals and lease entries released on a schedule.
- Reconciliation preparation. Schedules assembled automatically, with a human reviewing exceptions rather than building the pack.
- Pack generation. The management pack drafted from the ledger with variance commentary proposed against budget, then edited and signed by a Chartered Accountant.
Sequence matters. Automating the pack while the ledger is still reconciled manually produces a beautifully formatted document containing numbers nobody has checked. Fix the base first.
A five-day close is not about speed. It is about the twenty working days a month it gives back to the people who could be doing something more useful.
XLCFOFive working days from period end to a signed management pack is a service commitment in our engagement letters, not an aspiration.
