Close · August 1, 2026

Month end close process: the step by step checklist, the accounting activities, and how long the close should take

The month end close is the recurring set of accounting steps that finalizes one month's books: recording every remaining transaction, reconciling accounts to outside statements, posting accruals and adjustments, producing financial statements, reviewing variance against budget, and locking the period so the numbers stop moving. Its purpose is to produce figures somebody can make a decision on.

Most close problems are not accounting problems. They are sequencing and chasing problems. The journal entries take an afternoon; waiting for a receipt from a salesperson who has been on the road for two weeks takes eleven days. Below is the full sequence with an owner and a purpose for each step, an honest look at the benchmark everyone gets measured against, and the specific places the close reliably stalls.

What are the steps in the month end close?

Eleven steps in order: confirm the close calendar, send cut-off reminders, collect outstanding receipts and invoices, record remaining transactions, reconcile bank and card accounts, review accounts receivable and accounts payable, process payroll and related accruals, post accruals, prepayments and depreciation, prepare the financial statements, run variance analysis and management review, then lock the period and document what happened.

The order is not cosmetic. The close is a dependency chain, not a to-do list. Reconciliation cannot start until transactions are recorded. Statements cannot be produced until accruals are posted. A team that works the list in parallel because it feels faster usually ends up redoing the last three steps.

The month end close checklist, step by step

Each row names who owns it and what it is actually there to catch. The third column is the one worth reading, because a step nobody can explain the purpose of is a step that gets skipped in a busy month.

Swipe the table sideways to compare all columns.

Two of these steps are deep enough to deserve their own treatment. The card reconciliation mechanics in step five are covered in corporate card reconciliation, and the arithmetic and thresholds behind step ten are in budget vs actual variance. This piece is the sequence they both sit inside.

How long should the month end close take?

APQC, surveying roughly 2,300 organizations, found a median of 6.4 calendar days, with top performers finishing in 4.8 days and bottom performers taking 10. Ventana Research reported in 2022 that 59 percent of companies close within six business days at month end, and that only 43 percent manage six days at quarter end. Five to seven days is a reasonable target for most mid-sized US companies.

Now the caveat nobody attaches when they quote those numbers. The Ventana figure comes from a survey of 48 companies, roughly 70 percent of which had over 1,000 employees. That is a small sample skewed toward large enterprises with dedicated close teams. If you are a 60-person company with one controller and a part-time bookkeeper, you are not underperforming a peer group when you take nine days. You are being compared to a different kind of organization.

The more useful measure is your own trend. Track days-to-close month over month and, more importantly, track how many of those days were spent waiting for other people rather than doing accounting work. That second number is the one you can actually change, and in most teams it is well over half the total. PwC research cited across the FP&A field puts about 60 percent of FP&A data as still needing manual manipulation, with the large majority of the work happening in offline spreadsheets, which is a fair description of where the days go.

Where the close actually stalls

In practice, four bottlenecks account for most slipped closes, and only one of them is accounting.

Missing receipts. A charge posted on the 27th with no receipt and no explanation cannot be coded, and it blocks the reconciliation it sits inside. Chasing it takes days of calendar time and about four minutes of actual work. Teams that fix this fix it at the source, by asking for the receipt within a day of the charge rather than during close week.

Unrecorded liabilities. A vendor delivered in July and invoices in mid-August. If nobody accrues it, July looks cheaper than it was and August looks worse. This is the most common cause of a month that reads fine and then gets restated.

Receivables that were never chased. Step six regularly turns into an archaeology exercise: invoices that went out late, invoices that went to the wrong contact, credits applied to the wrong account. Teams that keep aging under control during the month rather than at close usually do it by automating the follow-up, so that every open invoice gets chased on a schedule instead of when somebody remembers. The close then just confirms a number rather than discovering one.

Surprises in the variance review. Step ten is where a department discovers it is 40 percent over budget, three weeks after the money went out. The review is doing its job by finding it. The problem is that finding it in the second week of the following month means the only available response is an explanation, not a decision.

Can the month end close be automated?

Parts of it, meaningfully. Bank and card feeds, transaction matching, recurring journal entries, depreciation schedules and statement generation all automate well and are where most of the reclaimed days come from. Judgment steps do not automate: whether an accrual estimate is reasonable, whether a variance signals a real problem, and whether an unusual transaction is legitimate all still need a person who understands the business.

Worth being clear about what automation actually buys, though. Automating step five makes reconciliation faster. It does not make the 27th-of-the-month charge appear any earlier in somebody's awareness. If a specific transaction was going to be a problem, an automated close finds it on day two instead of day six. That is better, and it is still after the fact.

The close reports the month. It does not change it.

This is the structural limit of every close process, however well run. By design, the close looks backward. It tells you accurately what happened to money that has already left. A perfectly executed four-day close still delivers the news that a team overspent by 40 percent four days after the month it happened in, by which point the vendor has been paid, the trip has been taken and the subscription has renewed for another year.

The complement is watching spend as it posts rather than only when it closes. That is what budget alerts and anomaly detection are for: a warning to the budget owner at 80 percent, an escalation to finance at 100 percent, and a flag on the duplicate or the unfamiliar merchant the day it lands. The honest limit is worth stating too, because it cuts both ways: monitoring cannot decline a transaction, and it does not close your books. It moves the moment you find out, which is usually the part that mattered.

It also takes pressure off step five. Most of what card reconciliation uncovers is not fraud, it is unexplained charges, and each one costs a round of messages to resolve. Resolving them in the moment, when the person still remembers the charge, is the cheapest version of that conversation. If you are weighing platforms that promise a faster close as part of the pitch, the trade-offs between the two most common ones are laid out in Ramp vs Concur.

What is the difference between month end close and year end close?

The month end close finalizes one accounting period for internal reporting. The year end close does everything the month end close does and adds annual work: full-year adjusting entries, closing revenue and expense accounts to retained earnings, the tax provision, fixed asset and inventory verification, and preparation for external audit. Year end is slower mostly because more of it gets examined by people outside the company.

The practical implication is that a disciplined monthly close makes the annual one cheaper. Every reconciliation left unfinished in March is a question your auditor asks in February, when the person who could have answered it easily has forgotten the details or left.

Getting started without rebuilding anything

If you do not have a written checklist today, and by most accounts a sizeable minority of companies do not, start by documenting what you already do. Run one close with a timer and a notepad, write down every step in the order it actually happened, note who did it, and mark where you waited. That document is a better checklist than any template because it describes your business.

Then fix the waiting, not the accounting. Pull the receipt deadline earlier, put a name against each step, and get visibility on card spend before close week rather than during it. Most teams find that removes two to three days without changing a single journal entry.