---
title: "Zero-Day Close: What a Short Month-End Close Can Hide"
description: "A three-day close can still hide a week of manual work. What a zero-day close means, where close work hides, and how to measure close hours."
url: https://emptyspaces.ai/resources/blog/faster-month-end-close
language: en
---

[Blog](https://emptyspaces.ai/resources/blog) Analysis

# Zero-day close: Why a short close can still hide a week of manual work

Published October 15, 2026 8 min read

Key takeaways

- Days to close measures elapsed time, not effort, so a three-day close can still sit on top of a week of reconciliations, catch-up, and overtime.
- OpenAI's CFO set a zero-day close as a goal in August 2026, and her proposed scorecard adds automatic reconciliation, exceptions, and the time to explain a variance.
- Track close hours alongside close days, including work before day 1 and after sign-off, to tell a better process from a team absorbing the same work.

A zero-day close is a worthy goal, but close days only show how much time passed, not how much work the close took. Getting this right gives you an accurate view of your team's capacity, catches work that slips outside the close window, and keeps a fast close from depending on overtime. This guide covers where the hidden work sits, what a zero-day close really means, and how to measure close effort, with a worked example.

## Is days to close the right measure of an efficient close?

Days to close is the right measure of speed, but it's incomplete as a measure of efficiency. It counts elapsed time and ignores the hours your team spends before, during, and after the close. Separating the two shows you whether your close is getting easier or just more compressed.

- **Close time** Business days from month-end until the statements are signed off
- **Close effort** Total hours your team spends on close tasks, from the first pre-close task to the last reconciliation
- **Hidden work** Close effort that days to close does not reveal, such as pre-close prep, overtime, and post-sign-off clean-up

Major benchmarks already treat speed and effort as separate measures. In The Hackett Group's June 2025 research, Digital World Class finance teams had “35%-57% shorter close cycles than peers.” They also needed “up to 42% fewer full-time equivalents across key finance functions.” APQC's general accounting benchmarks likewise track the total cost of the work next to the cycle time for the monthly statements.

Speed and effort can move in different directions. In Ledge's 2025 survey of 100 finance professionals, 18% of teams closed in 1–3 business days. Yet cash reconciliation alone took 20–50 hours a month on average, and 37% of respondents named understaffing or capacity gaps as a blocker. A fast close and a heavy workload can both be true.

For the basics of the close itself, see [what month-end close is](https://emptyspaces.ai/resources/blog/month-end-close).

## Where does the manual work hide in a fast close?

The manual work in a fast close usually moves to the edges of the close window, where days to close doesn't count it. Knowing where to look tells you which hours to measure first.

| Where it hides | What it looks like | Why days to close misses it |
| ----------------- | -------------------------------------------------------------------------------------------- | -------------------------------------------------------- |
| Before day 1 | Reconciling to date, preparing accruals, and chasing approvals in the last week of the month | The clock starts at month-end |
| During the close | Long days and weekend work to hit the deadline | The calendar shows three days, not the hours inside them |
| After sign-off | Reconciliations finished the next week, reclasses, and old reconciling items carried forward | The clock stops at sign-off |
| Waiting on others | Late invoices, payroll files, and approvals from outside finance | Waiting time looks the same as working time |

A CFO we spoke with in September 2026 runs a three-day close. When the same team reconciles the balance sheet, it still finds missed accruals and accruals that were booked but never used. It also finds supplier advances that were never matched to an invoice. The close is fast, and the clean-up lands where the close calendar doesn't show it.

An illustration of our Close Agent's balance-sheet review listing an unmatched $10,000.00 supplier advance, an unused $6,500.00 accrual, and a June hosting cost with no invoice yet. Demo data.

Why does month-end close take so long when the tasks are routine? Capacity is a large part of the answer. Gartner's February 2024 survey of 497 controllership staff found errors “closely linked to concerns about low capacity.” In the same survey, 73% of accountants said new regulations had increased their workload over the past three years. To turn that last-week scramble into a planned routine, see the [month-end close process](https://emptyspaces.ai/resources/blog/month-end-close-process).

## What is a zero-day close?

A zero-day close means your books are reconciled and ready to report on the last day of the month. It's the end point of a continuous close, where reconciliations, accruals, and reviews run daily or weekly instead of after month-end.

The idea has a high-profile backer. In an August 10, 2026 essay, OpenAI CFO Sarah Friar wrote, “We set two bold ambitions: a zero-day close and automated, continuously updated forecasting.” She described the goal as “a real-time, reconciled, and traceable view of the company's financial position.”

Her description of the end state matters more than the name. “The close does not disappear,” she wrote. “What begins to disappear is the scramble to reconstruct the business after the period ends.” People still own the result: “Finance validates the numbers, applies judgment, and owns the final sign-off.”

She didn't propose measuring it in days alone, either. Her suggested scorecard “might include cycle time, the share of transactions reconciled automatically, the number of exceptions requiring review, and the time required to explain a variance.” Three of those four measures describe the work, not the calendar.

Is a zero-day close realistic for a mid-market finance team? It depends on reliable data feeds and automated matching, so a practical first step is to run your high-volume accounts continuously. The benefit of a faster close is a current view of the business. You only get it if the work is actually done, not deferred.

## Does a continuous close reduce work or just move it?

A continuous close moves work first, and it only reduces work when steps get shorter or disappear. Measuring hours tells you which one is happening.

Moving bank matching from month-end to every week spreads the same hours across the month. The close gets shorter, but your team works the same total. Hours fall only when something removes steps: matching that runs without a person, accruals that post from a schedule, or exceptions explained before anyone investigates them.

That's why a shorter close can mislead you. If the close drops from six days to three while total hours stay flat, you've changed the calendar, not the workload. If it drops to three days and total hours fall, the process improved.

## How do you measure close effort, not just close days?

Measure close effort by logging hours by task for one full close, then tracking a few ratios every month. This gives you a baseline that shows whether changes save hours or only shift them.

| Metric | How to calculate it | What it tells you |
| ------------------------------ | ---------------------------------------------------------------------- | --------------------------------------- |
| Days to close | Business days from month-end to sign-off | How fast the books close |
| Close hours | All hours logged to close tasks, including pre-close work and overtime | The total cost of the close |
| Hours outside the window | Hours before day 1 or after sign-off ÷ close hours | How much work the calendar doesn't show |
| Work finished after sign-off | Reconciliations or entries completed after sign-off | Whether “closed” means done |
| Aged reconciling items | Items older than one month on any reconciliation | Clean-up work carried forward |
| Share reconciled automatically | Transactions matched without a person ÷ all transactions | Whether hours can actually fall |

Example

Your team of five closes in three business days. Here's where the hours go:

| Phase | Hours |
| --------------------------------------------------------------------------------------- | ----- |
| Week before month-end: two people spend half their time on close prep (2 × 4 × 5) | 40 |
| Close days 1–3: all five work 10-hour days, including 30 hours of overtime (5 × 10 × 3) | 150 |
| Week after sign-off: 12 reconciliations at 2.5 hours each, plus 10 hours of reclasses | 40 |

Your close hours are 40 + 150 + 40 = 230. Only 150 of them, about 65%, fall inside the three days. The other 80 hours, about 35%, equal two full 40-hour weeks that days to close never shows. The work actually spans 13 business days, not three.

Now you have a baseline. Say next quarter's close still takes three days but uses 180 hours, with 20 outside the window. Your process improved even though your days to close didn't change. For time benchmarks, see [how long month-end close should take](https://emptyspaces.ai/resources/blog/month-end-close-time).

## How do I speed up month-end close without losing controls?

Speed up the close by moving preparation into the month while keeping review, approval, and the period lock at month-end. Separating what can move from what has to wait protects your controls as the close gets shorter.

| Work | What can move into the month | What still happens at period-end |
| ----------------------------- | ----------------------------------------------------------- | ----------------------------------------- |
| Bank and card reconciliations | Matching and exception review, daily or weekly | The final statement days and sign-off |
| Recurring entries | Prepaids and depreciation on a schedule | Review of anything that changed |
| Accruals | Estimates from open purchase orders and recurring suppliers | True-ups and the final completeness check |
| Supplier advances | Matching each advance to its invoice as invoices arrive | Review of unmatched advances |
| Management reporting | Commentary drafted on preliminary numbers | Final numbers and CFO review |

An illustration of task history for a June bank reconciliation showing weekly matching on June 8 and June 15 and the final workpaper submitted for review on July 1. Demo data.

Three controls stay fixed however fast you close. A named approver signs off every entry before it posts, and the person who prepares the work doesn't review it. The period also locks after sign-off. For the changes that shorten the close, see [how long month-end close should take](https://emptyspaces.ai/resources/blog/month-end-close-time). For each task with its owner and reviewer, see the [month-end close checklist](https://emptyspaces.ai/resources/blog/month-end-close-checklist).

## How we approach it

Our Close Agent works your close checklist through the month on the schedules you set. Every task has a named reviewer who's separate from the preparer. If you want to see the preparation move off your team's calendar instead of just earlier in it, explore [our Close Agent](https://emptyspaces.ai/product/close).

Illustration: the cash and bank group of the March 2026 close checklist in Autonomous Close. The operating-account bank reconciliation is signed off, the payroll account is in review, the receivables tie-out is in progress and two tasks have not started. A chip shows Run all started four tasks.

Share this article

- [ Share on X ](https://x.com/intent/post?url=https%3A%2F%2Femptyspaces.ai%2Fresources%2Fblog%2Ffaster-month-end-close&text=Zero-day%20close%3A%20Why%20a%20short%20close%20can%20still%20hide%20a%20week%20of%20manual%20work)
- [ Share on LinkedIn ](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Femptyspaces.ai%2Fresources%2Fblog%2Ffaster-month-end-close)

## See our agents work through a finance process that matters to your team.

Tell us where the work gets stuck. We’ll walk through how the agents could help.

[Book a working session](https://emptyspaces.ai/contact-us)