The ledger is not the bottleneck
Ask a finance team what makes the close slow and they will describe the ledger: the journals, the accruals, the reconciliations. Watch one for a fortnight and you see something different. The ledger work takes about a day.
The rest is waiting. Waiting for the delivery team to confirm what was completed. Waiting for someone to say whether an invoice should be raised. Waiting for expenses that are sitting in a coat pocket. Waiting for a spreadsheet from the person who owns the spreadsheet, who is on leave.
Six days of asking for numbers that already exist
The uncomfortable part is that almost none of what finance waits for is unknown. The hours were logged. The milestone was completed. The expense was incurred and the person knows it. The information exists — it is simply somewhere finance cannot read, held by somebody who has no deadline of their own.
So the close becomes an internal collections exercise. A finance person spends six days chasing colleagues, and the skill that matters most in the role turns out to be politeness under repetition.
Why closing faster usually fails
The standard responses are a checklist, a deadline, or a project to "streamline the close". These help at the margin and they do not touch the cause, because the cause is not disorganisation. It is that the systems do not share a record.
When delivery lives in one system and the ledger in another, somebody has to carry information between them, and that somebody is finance at month end. A checklist makes the carrying more orderly. It does not remove it.
What actually shortens it
The close gets shorter when postings arrive from the source document as it happens rather than being assembled afterwards. An invoice posts when it is issued. A pay run posts when it is run. Delivered work is delivered work, in the same record, so nobody has to be asked what was completed.
What is left is genuine judgement: accruals that need an estimate, a provision that needs a decision, an exception worth a conversation. That is a day of skilled work, which is what a close should be.
There is a useful test. Count how much of your close is arithmetic and how much is asking. If the second number is larger, the problem is not in the ledger and no amount of process design in finance will fix it, because the fix is not in finance.