The agenda that finished in twelve minutes
There is a particular kind of meeting everyone has sat through. The real business — the decision, the update, the one question that needed an answer — is done inside the first fifteen minutes. Then something strange happens. Nobody leaves.
Instead, the conversation softens and spreads. Someone circles back to a point already settled. Someone else raises a tangent that doesn't need this room to resolve. A third person, sensing the lull, offers an update nobody requested. The clock, which started this whole thing, quietly finishes its job: the meeting ends at the top of the hour, because the meeting was booked for an hour.
We tend to explain this away with the people in the room. Someone talks too much. Someone is unprepared. But the more honest explanation has nothing to do with personalities and everything to do with the container. The work didn't take an hour. The hour took an hour.
What Parkinson actually observed
In 1955, the British historian C. Northcote Parkinson opened an essay in The Economist with a sentence that has outlived almost everything else he wrote: "Work expands so as to fill the time available for its completion."
He was writing satirically, about bureaucracies that grew their headcount regardless of how much work there was to do. His famous illustration was an elderly woman with nothing to occupy her day except mailing a single postcard. The task — finding the card, locating her glasses, writing the address, deciding whether to take an umbrella — consumes the entire day. A busy person would have done it in three minutes between other things. The work was identical. The time available was not.
What began as a joke about civil servants turned out to describe something durable about human effort. Given a generous deadline, we don't finish early and bank the surplus. We unconsciously inflate the task to match the runway. The deliberation grows more elaborate, the standards creep upward, the small decisions acquire a weight they wouldn't otherwise carry. This is Parkinson's Law, and a meeting is one of its purest demonstrations, because a meeting comes with its boundary printed right on the calendar.
Why the hour is the enemy
The trouble starts before anyone speaks, at the moment the meeting is created. Calendar software offers you defaults — thirty minutes, sixty minutes — and almost nobody overrides them. We don't estimate how long the conversation needs and then book that. We book a round number and then let the conversation discover that it has that much room.
There's a well-documented cognitive shortcut at work here that psychologists call unit bias: our tendency to treat one standard unit of something as the correct amount. People given a larger serving spoon take more food; people given a bigger plate finish it. The portion feels right simply because it's the portion offered. An hour-long slot is a serving size, and we eat the whole thing without noticing we're full.
A meeting also lacks the one thing that makes most tasks self-limiting: a natural finish line. When you cook dinner, the food is either done or it isn't. When you fix a bug, the test passes or fails. But a discussion has no built-in moment where it is objectively complete. There is almost always one more angle to consider, one more person who could weigh in, one more "while we're all here." Without an external stop, the conversation keeps finding reasons to continue — and the only external stop available is the calendar boundary you set at the start. So the meeting expands politely, agreeably, right up to the edge of the hour, and then mistakes that edge for the moment the work was done.
The cost isn't just the extra minutes
It's tempting to think the waste here is small — what's twenty surplus minutes? But the cost compounds in ways the calendar hides.
The obvious one is arithmetic. Twenty padded minutes in a meeting of eight people is not twenty minutes; it is more than two and a half hours of collective human attention, spent generating the appearance of thoroughness. Multiply that across a week of recurring meetings and the number stops being trivial.
The subtler cost is to the quality of the thinking. When a decision has fifteen minutes, people get to the point. When it has an hour, the extra time doesn't buy better decisions — it buys over-deliberation. Settled questions get reopened. Confidence gets talked back down into doubt. The group polishes a conclusion that was already good enough, or worse, second-guesses it into something murkier. Parkinson's Law doesn't just waste time; it actively degrades work by giving it room to wander.
And there's a credibility cost. When meetings reliably run their full length regardless of content, people stop trusting the schedule. They arrive late because they assume the first ten minutes are throat-clearing. They multitask because they've learned the dense part might not come until minute forty. The padding becomes self-fulfilling: everyone behaves as though the hour is loose, so the hour stays loose.
Shrinking the container
The useful thing about Parkinson's Law is that it cuts both ways. If work expands to fill the time available, then reducing the time available compresses the work — and most of what gets squeezed out is the padding, not the substance.
Book the meeting you need, not the meeting the calendar offers. The single highest-leverage move is to override the default. Try twenty-five minutes instead of thirty, fifty instead of sixty. The odd numbers do real work: they signal that the length was chosen deliberately, and they hand people back the buffer between meetings that the round numbers quietly steal.
Put the decision first, not last. Agendas tend to build toward the important item, which guarantees the trivial ones get full attention while everyone is fresh and the real question gets a tired, rushed ten minutes. Lead with what actually needs the room. If the meeting can end once that's resolved, you've created a natural finish line where there wasn't one.
Make ending early not just allowed but admired. A culture where giving people back twenty minutes is a small triumph — rather than a sign the meeting was unnecessary — breaks the expansion at its root. The goal of a meeting is to finish, not to fill.
Default to a smaller container and only grow it on evidence. It is far easier to add five minutes to a conversation that genuinely needs them than to reclaim twenty from one that has already spread to fill them. Start tight. Let the rare meeting that truly needs more time prove it.
None of this requires anyone to talk faster or care less. It requires noticing that the hour was never a measurement of the work. It was a guess, made in advance, by someone clicking the default — and then quietly enforced by everyone in the room.
The meeting you can actually feel
The reason Parkinson's Law is so hard to see in the moment is that an expanding meeting doesn't feel wasteful from the inside. It feels collaborative. It feels thorough. Everyone is engaged, nobody is being rude, and the conversation flows right up to the polite stopping point. The waste is invisible precisely because it looks like diligence.
That's where a record helps more than a resolution. It's one thing to suspect your meetings run long; it's another to see, week over week, that the recurring sync booked for sixty minutes reaches its real decision at minute fourteen and then drifts. MeetingMortem exists to make that drift visible — a quiet post-mortem on where the time actually went, so you can size the next container to the work instead of to the default. Once you can see the padding, you can stop booking it. If your meetings have a habit of finding exactly enough to say to reach the top of the hour, you can start watching for it at meetingmortem.lumenlabs.works.