How to close the books in six days

Companies that close in three weeks are not doing three weeks of work. They are doing about five days of work spread across three weeks, because nothing starts until the thing before it finishes and nobody owns the queue.

The cost is not the accounting hours. It is that management spends the first half of every month making decisions on last quarter's information, and by the time the numbers land there are only ten days left to react before the cycle repeats.

The fix begins with a written close calendar: every task, an owner, a deadline expressed in business days from month end, and its dependencies. Most companies have never written this down, and the act of writing it exposes that four tasks are waiting on one person who did not know they were blocking.

Then move work before month end. Accruals for recurring items can be templated in advance. Vendor invoices can be chased in the last week of the month rather than the first week of the next. Prepaid schedules and depreciation can be rolled forward before the period closes. In most companies a third of close work does not require the month to be over.

Then set a materiality threshold and honour it. Teams routinely spend two days chasing a variance smaller than the rounding on the board deck. Write the threshold down, get the CEO to agree to it once, and stop relitigating it every month.

Then separate the close from the analysis. The books closing and the variance review are different deliverables with different audiences. Waiting to publish because the commentary is not written keeps the whole company waiting on a paragraph.

Six days is achievable for most venture-backed companies under two hundred people. The constraint is almost never accounting capability. It is that nobody has been given the job of making the close fast, so it stays exactly as slow as it has always been.