A three-week close is a data problem, not an accounting problem.

Ask an owner when they see last month's numbers and the answer is often "around the twentieth", sometimes later. Ask why, and the answer is usually some version of "that's how long the accountant takes."

I spent most of my career measuring processes before trying to improve them, and when you measure a month-end close, the accounting itself is a small fraction of the elapsed time. Posting, reconciling and reviewing a month for a small or mid-sized business is a few days of actual work. The rest of the calendar is spent waiting: for bank statements, for invoices, for payroll detail, for someone to explain a transfer nobody recognises.

Where the days actually go

When you trace a slow close day by day, the delays fall into a short list:

• Documents collected by hand. Expense invoices arrive by email, WhatsApp or on paper, and someone gathers them at month-end rather than as they come in.
• Bank movements retyped. Statements are downloaded and entered line by line, instead of flowing into the ledger as they happen.
• Payroll handed over as a summary. The accountant receives totals, not the detail, and has to ask for the breakdown before anything can be posted.
• Supplier payments made one at a time. Each transfer has to be matched afterwards to an invoice that may or may not have been recorded.
• Unexplained items. A deposit or a charge that nobody can identify waits in a queue until the one person who knows is available.

Every item on that list is a handoff. Information is created in one place, in the bank, on an invoice, in a payroll run, and has to be moved by a person into another place before the month can be closed.

Why a faster accountant does not fix it

The natural response to a slow close is to push the person doing it. That rarely works, for the same reason a faster cashier does not shorten a queue that is waiting on the card machine. If the information reaches the accountant on the twelfth, the books cannot close on the sixth, however good the accountant is.

The length of your close is set by the last piece of information to arrive, not by the speed of the person waiting for it.

This is also why changing firms often changes less than expected. A new provider inherits the same handoffs, the same email threads and the same unexplained transfers. Unless the way information reaches the ledger changes, the calendar does not.

What a fast close actually requires

The fix is to remove the handoffs, so that the ledger is current before the month ends rather than rebuilt after it. In Costa Rica, much of what is needed already exists:

• Electronic invoices are already structured data. Every supplier invoice is issued through Hacienda's system as a file that can be received and posted without being retyped.
• Bank movements can flow in daily. When the ledger is connected to the bank, reconciliation becomes a matter of confirming matches, not entering lines.
• Payroll can post itself. If payroll runs in a system rather than a spreadsheet, the journal entry comes from the same record that paid the employees.
• Supplier payments can go out in batches, each tied to the invoice it settles, so the match exists from the moment the money leaves.

None of this removes the accountant. It moves their time from collecting and typing to reviewing and explaining, which is the part of the work that was always worth paying for. In our own practice, clients who run payroll, supplier payments and bank movements through connected systems close in an average of six business days.

What a late close costs

The direct cost is easy to see: hours spent chasing and retyping. The larger cost is harder to see, because it shows up as decisions made without numbers.

When September's results arrive on October 22, most of October's decisions have already been made. Pricing, hiring, a supplier negotiation, how much cash to hold for December: all of them were decided on instinct and last quarter's memory. A margin problem that started in August is discovered in late September and confirmed in late October. By then it has run for three months.

A close that finishes in the first week does not make anyone a better manager. It simply means the information arrives while it can still change what you do.

A test you can run this week

For last month's close, write down two dates: the day the last piece of information reached your accountant, and the day you received your financial statements. Then list what that last piece of information was.

If the gap between the two dates is short and the list is long, your accountant is not the bottleneck. The handoffs are. That is good news, because handoffs can be designed out, and it costs far less than adding people to wait for them.