Parkinson's law: why a 10-minute job takes all afternoon

You sit down after lunch to write a short party invite. It should take ten minutes. But there's no rush, so you look for a nicer font, rewrite the opening, check the date again, try a different picture. At five o'clock you're on version seven, and it isn't really any better than version one.

That's Parkinson's law: "work expands so as to fill the time available for its completion." Give a job all afternoon and it will happily take all afternoon. Below is where the law comes from (it began as satire, which matters), how to put it to practical use with short, realistic deadlines, the catch that keeps that from turning into rushing, and how to set it up in dypt.

Parkinson's law at a glance

• What it says: work expands to fill the time available for its completion.
• Where it comes from: a satirical 1955 essay by C. Northcote Parkinson about the growth of government departments.
• How people use it: give a task a short, realistic deadline instead of an open-ended one.
• The catch: decide what "done" looks like before you start, or a short deadline just means rushing.
• Who it helps: anyone whose small jobs keep swallowing whole afternoons.

Where it comes from

The law is the opening line of an essay by the historian C. Northcote Parkinson, published in The Economist in November 1955 (reprinted here). His first example is about one person: an elderly lady of leisure who spends an entire day writing and posting a single postcard. She spends an hour finding the postcard, another hunting for her glasses, half an hour searching for the address, and so on, for a job that would take a busy person three minutes.

But the essay isn't really about personal productivity. It's a satire of bureaucracy. Parkinson's real targets were government departments such as the Admiralty and the Colonial Office, whose staff kept growing whatever happened to the amount of work they had to do. As the author Cal Newport points out, the famous opening sentence is now quoted far more than the argument behind it.

So it's worth being clear about what follows. Parkinson didn't run an experiment showing that tight deadlines make people more productive. Using short deadlines on your own work is an application of his observation, and a useful one, not something he proved.

Why work expands

You'll recognise the pattern even without the civil service. A few things tend to happen when a job has no edges:

• There's no point at which it's finished. Without a limit, you can always do a bit more: another check, another tweak, another version.
• Polishing feels like progress. Version seven feels more finished than version one, even when nobody would notice the difference.
• Small decisions get room to grow. With all afternoon, choosing a font becomes a twenty-minute research project.
• Starting gets put off. If you have until five, there's no hurry to start at two.

A limit changes the question. Instead of "how good can I make this?", you're asking "what matters most in the time I've got?"

How to use Parkinson's law

1. Pick one job. Start with something that usually drags: an email you keep redrafting, a short report, a bit of tidying.
2. Decide what done looks like. Before you start, write down what the finished job needs. For a party invite: a guest list, the date and place, and sent. That's it.
3. Give it a short, realistic deadline. Short enough to stop it spreading, long enough to do it properly. Ten minutes for the invite, not all afternoon. Not two minutes either.
4. Work to the deadline. When you notice yourself polishing, check your definition of done. If it's met, stop.
5. When the time's up, finish or decide. If it's done, send it. If it's genuinely not, decide how much more time it needs, and set a new deadline rather than drifting.
6. Learn from it. Note how long it really took. Next time, your deadline can be more realistic.

The word that matters most is realistic. The aim is to stop work spreading, not to rush everything. A deadline you can't meet just moves the stress around.

The catch: decide what done looks like

A short deadline without a clear finish line is just pressure, and it can encourage rushing. Deciding what "done" means before you start is what makes the deadline useful, because it tells you what to leave out.

This isn't from Parkinson's essay; it's common advice for stopping scope creep. But the two work together. The deadline stops the job spreading in time; the definition of done stops it spreading in size.

It also helps to know which jobs deserve more time. Some work gets genuinely better with care: important writing, a tricky decision, anything where mistakes are costly. Parkinson's law is best used on jobs where extra time mostly turns into extra polishing.

Strengths and limitations

Strength	Limitation	What to watch for
Gives drifting jobs a finish line	Based on a satire, not a study of productivity	Use it as a practical habit, not a law of nature
Makes you decide what matters in the job	Tight deadlines can push you to rush	Keep deadlines realistic and define done first
Gives you a clear point to start	Some work really does need more time	Save it for jobs that tend to drift
A chance to compare estimates with actual time	Self-set deadlines are easy to ignore	Make them visible and give yourself a reminder

Parkinson's law, timeboxing and Pomodoro

Giving a job a fixed amount of time is close to what people call timeboxing, and our Pomodoro vs time blocking vs timeboxing comparison covers how those methods differ. If the problem isn't the job spreading but your focus wandering while you do it, the Pomodoro technique protects 25 minutes at a time. And when you're processing incoming work, the two-minute rule says to do any necessary action you can finish in under two minutes straight away. During focused work, leave new arrivals for your next processing session.

Using Parkinson's law in dypt

dypt has the two things this needs: a deadline and a way to write down what done means.

• Give the job a deadline with a time. Deadlines in dypt have a time as well as a date, so "Write the party invite" can be due at 2:10pm, not just "today". If you want a nudge, enable notifications on your device and add a reminder shortly before it.
• Write the definition of done as subtasks. Under "Write the party invite", add Guest list, Date and place and Send it. When those three are ticked, the job is done, however tempting another draft might be.
• Estimate first, then check. Add a time estimate before you start. When you finish, mark the task complete and enter the actual time. Over a few weeks, comparing the two shows which jobs you tend to underestimate and which ones you let spread.
• Spot what keeps slipping. The deadline filter shows what's overdue. If a job keeps becoming overdue, check its size, what it's waiting on, its finish line and the time you actually have.

dypt doesn't have a countdown timer, so if you want to see the minutes ticking down, use your phone alongside it.

Give one job a deadline today

Pick one job you'd normally leave open-ended this afternoon. Write down what done looks like, in two or three bullet points. Then give it a short, realistic deadline and start. When you hit your definition of done, stop, even if you could keep polishing.

If you'd like a place to set that deadline and keep your definition of done next to it, try it in dypt, free to start.

Further reading

• C. Northcote Parkinson: Parkinson's Law (1955), reprinted
  The original essay, including the postcard and the Admiralty figures.

• The Economist archive: Parkinson's Law
  The essay's original home, 19 November 1955 (may need a subscription).

• Cal Newport: Debunking Parkinson's Law
  A useful caution on how the law's opening line gets used out of context.