Four O'Clock, Full Stop: A Two-Week Trial of Not Starting Anything New

There is a particular kind of optimism that lives in a small firm at ten to four. Someone says, “I’ll just get this proposal started,” and by the time the laptop closes it is half six, the office is dark, and tomorrow’s first hour has already been spent on today’s loose ends. The 4pm cut-off is an attempt to interrupt that pattern before it becomes the day.

The idea is simple enough: after four o’clock, you do not begin new work. You finish, you tidy, you reply to what is already in play, and then you go. For two weeks, a small UK firm treats 4pm as a genuine boundary rather than a vague aspiration. The point is not to become rigid. It is to find out what actually happens when the last hour of the day stops being an overflow channel for everything that arrived after lunch.

What counts as new work?

This is where most trials fall apart, so it is worth writing the rule down in plain English before day one. New work means starting a task that did not exist on your list before four o’clock. That could be a fresh client enquiry, a report you have only just opened, a quotation for a job that came in at 3:55, or the “quick” bit of admin that nobody planned for. It is the beginning of something, not the finishing of something.

Finishing is allowed. So is replying to a message that genuinely needs a same-day answer, attending a meeting that was already in the diary, and dealing with anything that falls under your exception list. The distinction matters because the cut-off is not a ban on working after four. It is a ban on pretending that a new task will take ten minutes when everyone in the room knows it will take forty.

Set the rule so it sticks

Announce the trial properly. A note in the team chat is fine, but a short conversation is better. Explain that for ten working days, nobody starts anything new after four unless it meets one of the agreed exceptions. Put the exceptions in writing. Typically they are: a safety issue, a legal or regulatory deadline, a client system that has gone down, or a payroll or invoicing deadline that cannot move. Everything else waits until the morning.

Give the day a physical signal. One firm I heard about sets a timer for 3:50pm and uses those ten minutes for a “landing sweep”: checking what is already open, noting what can wait, and deciding what genuinely needs finishing before five. Another keeps a shared note titled “After Four” where anyone can add a task that has been deliberately parked. The note is reviewed at 9am the next day. It is a holding pen, not a graveyard.

The first week will feel odd. There is a kind of withdrawal that comes from not responding to every late-afternoon ping. Expect the team to test the edges. Expect at least one person to say, “But this is different.” Sometimes it will be. Most of the time it will not.

Track what spills into the evening

You cannot judge the trial on vibes alone, so keep a light log. A simple spreadsheet is enough: date, time, task started after four, reason it could not wait, who started it, and what happened the next morning. Do not log every task you finished after four. Only log the new starts. That is the behaviour you are trying to understand.

After two weeks, read the log with the team. You are looking for patterns. Are the exceptions genuine, or are they mostly “the client expects it”? Is one person absorbing most of the late starts? Are the same types of task appearing again and again? If the same client sends a new brief at 4:15 every Tuesday, that is not an emergency. That is a scheduling problem wearing a disguise.

Measure the morning after

A cut-off is only useful if it improves something. The most practical thing to measure is next-morning focus. Ask everyone to rate their first hour on a simple five-point scale: one for scattered and reactive, five for clear and deliberate. Add a second question: how long did it take before you started something that genuinely mattered? Compare the two trial weeks with the week before.

You may also notice a change in decision quality. Late-afternoon decisions tend to be either rushed or overthought, and the difference shows up the next day in rework, clarification emails, and the quiet frustration of having to reopen something you thought was finished. A short note in the log about “what needed revisiting” is often more revealing than any productivity metric.

The exceptions are the point

A 4pm cut-off is not an argument for leaving a client stranded. It is an argument for knowing the difference between a real exception and a habit. If your team logs five exceptions a day, the rule is not working, but the answer is not to abandon it. The answer is to look upstream. Perhaps the day is over-committed. Perhaps the morning has too many meetings. Perhaps one client needs a clearer service agreement about response times.

The trial works best when it is treated as an experiment, not a policy. Two weeks is long enough to see the shape of the problem and short enough that nobody feels trapped. At the end, ask three questions. Did the mornings improve? Did the evening spill reduce? And did anyone finish earlier without the work suffering?

If the answer to all three is yes, keep it. If the answer is mixed, adjust the cut-off to half past four or three thirty and run it again. The number matters less than the principle: the last hour of the working day should belong to finishing, not to beginning. A week has 168 hours. The point of a cut-off is to stop one of them quietly eating the rest.