How to set a closing time for your workday (and keep it)

23 August 2026 · 6 min read

Pick a time, put it where other people can see it, and attach a visible event to it so the day ends with something happening rather than with you deciding to stop. Then measure whether you kept it, because a closing time you never check is a wish. The steps below take about ten minutes to set up and two minutes a day to keep.

Why closing time matters more than start time

Nobody struggles to start work. The calendar starts it for you, the first message starts it, the coffee starts it. The end has no such help. Work from home and the desk is never more than a room away; work in an office and the laptop comes home anyway. Without a closing time, the day ends when you run out of energy, which is a bad way to choose and gets later every week.

A closing time does three jobs. It gives you a point to plan toward, so the afternoon has a shape. It gives the people around you something to rely on. And it gives you a number to check, which is the part most advice skips. why-remote-work-needs-a-closing-time-more-than-an-opening-time.html makes the longer case.

Step 1: choose a time you will actually keep

1. Look at the last two weeks honestly. When did you really stop? If the answer is seven, a closing time of five is a resolution, not a plan. Start with something between the two.
2. Pick a clock time, not a condition. “When the inbox is empty” is not a closing time; it is a trap. Six o’clock is a closing time.
3. Decide what counts. Reading messages on the phone at eight is working. Be honest about whether your closing time closes the laptop or closes the work.
4. Tell someone. A partner, a housemate, the team channel. A closing time said out loud is harder to slide.

Step 2: make the Mac observe it

The time only holds if something happens at that time without you doing it.

• Calendar. Add a repeating event at closing time called “Close”. Give it an alert. It is the simplest version and it works better than it should.
• Focus. Open System Settings, choose Focus, and add a Focus with a schedule that starts at your closing time. Notifications from work apps stop arriving, which removes the main reason the evening leaks.
• Screen Time Downtime. In System Settings under Screen Time, Downtime can block chosen apps on a schedule. It is blunt, and you can override it, but the friction is real.
• A receipt. Punchcard sits in the menu bar and, at the closing time you set, prints a paper receipt of the day: itemized by app with time on each line, a day total, a stamp. The paper feeds out over a couple of seconds with a dot-matrix whirr. It is hard to keep working through a printed receipt of the day you just had; the day has visibly been totaled.

Punchcard has one setting that matters for this, the closing time, and no others to fiddle with: no timers, no projects, no account. It records app names only and asks for no macOS permissions. What it does not do is enforce anything. The receipt is a signal, not a lock, and whether you stop is still yours.

Step 3: a two-minute ending

A closing time works better when it closes something. The ending does not need to be elaborate.

1. Read the receipt or the day’s calendar. One minute. Notice the total and the top line.
2. Write tomorrow’s first task, one line, where you will see it in the morning.
3. Quit the work apps. Not minimize. Quit.
4. Tear off the receipt (drag down in Punchcard; it rips like paper) and either keep it or let it go.

a-shutdown-ritual-that-takes-two-minutes.html lays this out with a bit more room. The point is that the ritual is the same every day, so that by the second week it happens without deciding.

Step 4: when you blow through it

You will. The question is what happens the next day.

• If you ran late because of a genuine one-off, nothing. Tomorrow is six again.
• If you are late three days running, the closing time is wrong, not you. Move it thirty minutes later and keep the new one. A kept six-thirty beats a broken six.
• If the late nights are the pattern described in the-one-more-thing-trap-at-the-end-of-the-day.html, the fix is in the ending ritual: write the one more thing down as tomorrow’s first task instead of doing it.

Do not skip the receipt on a late day. The day total is the most useful number precisely when it is larger than you wanted.

Step 5: measure whether it held

This is the step that turns the closing time from intention to fact. You need the day totals for a few weeks in one place.

Punchcard prints a week roll on Sunday with the week’s days on it, and a month roll at month end. Lay two or three week rolls side by side and the pattern is obvious: which days held, which did not, and whether the totals are drifting. If you prefer a spreadsheet, Export CSV gives you the same numbers.

Look for two things. Are the day totals clustering near the hours your closing time implies? And are the days that broke the time the same weekday, the same kind of day, the same app at the top of the list? The second question is where the fix usually is. how-to-use-a-daily-total-to-set-a-weekly-limit.html takes the totals one step further.

Questions

What if my work genuinely has no fixed end, like support or on-call?

Then set the closing time for the part of the day that is yours to control, and treat after-hours work as the exception it is supposed to be. The receipt still shows the total, which is what lets you notice when the exception has become the rule.

Should the closing time be the same every day?

Easier if it is. A single time becomes a habit; a schedule becomes a decision each day. If you need two different times (say, one day with evening commitments), keep it to two.

Does a receipt at closing time stop me from working later?

No. Nothing stops you except you. What the receipt changes is that the day has been visibly counted, which makes the choice to keep going a choice rather than a drift. Most people find that is enough most days.