Work Expands. Time Vanishes. Here's Why.

Planning Laws: Parkinson & Hofstadter

Dec 11, 2025

It’s probably one of the most memorable origins of laws I’ve heard of, because it reminds me of my grandmother:

Cyril Northcote Parkinson, in 1955, published an article in The Economist stating: “work expands so as to fill the time available for its completion”. He illustrated this by describing an elderly lady spending an entire day mailing a postcard. First, searching for the card, her glasses, the right address, composing the message, spending twenty minutes deciding whether to take an umbrella, and eventually going to the mailbox. Something that would take a busy person three minutes can stretch to fill an entire day.

That’s Parkinson’s Law: work expands to fill the available time.

I can trace my own experience of this law back to school: doing homework at the last possible moment was the norm. It didn’t get better at university, where friends and I studied for exams just days before. The pattern followed me into work, and even into this newsletter.

Leaked Decision-Log: Deciding on Decision-Making Masterclass

Michał Poczwardowski

Jan 15

Read full story

Why does work expand? For me, it’s because I feel there’s no need to rush, which is often an illusion. The lack of urgency creates space for perfectionism, endless small improvements, and prioritisation. I fill time because it’s there to be filled.

But here’s the catch: even as we procrastinate, we’re also too optimistic about our estimates:

Douglas Hofstadter, in 1979’s “Gödel, Escher, Bach,“ wrote about early predictions in computer chess. Experts said it would take “about ten years” for a program to become a world champion. Ten years later, it was still “about ten years away.” These days, we hear similar claims: fully self-driving cars being just a few years away, for instance.

This became Hofstadter’s Law: it always takes longer than you expect, even when you take Hofstadter’s Law into account.

Two Laws in Practice

I’ve witnessed teams hiding real deadlines from each other. Afraid that if developers knew the real date, work would expand to fill it (Parkinson), while still taking longer than expected (Hofstadter). Some teams added multipliers to estimates, turning a 3-day estimate into 6 days automatically. This blurred what was being estimated in the first place.

For me, pushing deadlines in team settings is fine when it’s transparent and comes with full context. Hiding real deadlines entirely is an antipattern indicating mistrust or misalignment. When working individually, I like setting earlier deadlines. It forces me to build in safety buffers.

Tactics

Some tools can help us address the negative effects of these laws without resorting to fake deadlines.

Timeboxing

Timeboxing fights Parkinson’s Law by creating artificial scarcity. It also combats Hofstadter’s Law by forcing you to ship something imperfect rather than optimise endlessly.

I moved my publishing schedule several times this year, finally settling on Thursday morning and don’t want to change it. Now, having a fixed schedule helps me avoid over-editing articles.

Clear Definition of Done

Listing what “done” means helps with finishing. While writing articles, optimisations are infinite. I can always add something new, reframe a sentence, and find more examples. An article can be written in one hour or one week. Without clear criteria, I’d spend much more time writing.

My definition of done: at least 400 words, practical advice included, something new learned, no grammar errors.

Social Pressure

Public commitment builds enough pressure to stop procrastinating and actually start earlier. At the beginning of this year, we sent an email course for engineering managers. The commitment was to deliver high-quality lessons every week. It was demanding, but those emails were ready a week in advance to be sure there were no delays.

Balancing Uncertainties

Understanding uncertainties is important. I’ve written about it while explaining the Margin of Safety mental model. The short: list explicit risks, add buffers only where impact is high, and run pre-mortems.

The Good Side

These laws aren’t meant to be negative or positive. They describe how things work.

Parkinson’s Law can be beneficial: extra time lets ideas settle, creates space for exploration, and often leads to quality improvements.

Similarly, Hofstadter’s Law protects us from over-commitment and forces necessary scope cuts. The key is knowing when and how to work with these laws.

Discipline & Trust

The discipline is an antidote. You can timebox your work, craft a good definition of done, and broadcast your commitment. But the truth is, when my motivation is high, none of this really matters, work just happens anyway, it takes priority. Discipline matters most when motivation fades. And eventually, it always does.

Or maybe these laws reveal something simpler: a deficit of trust. We don’t trust ourselves to start early, so we procrastinate. Organisations don’t trust teams with real deadlines, so they hide them. The tactics help, but perhaps the more important work is to rebuild trust. Trust in our own capabilities, and in our teams.

Where did you experience these laws in action?

Thanks for reading,
— Michał

Share Perspectiveship

P.S. Excellent thoughts on this topic, through the lens of engineering teams, by James Stanier and Anton Zaides: Using fake deadlines without driving your engineers crazy