The Two-Minute Rule, and the Three Ways It Fails

The rule is from David Allen's Getting Things Done: if a task takes less than two minutes, do it immediately rather than tracking it. Tracking costs more than doing.

It is genuinely good advice. It also fails in three specific ways, and all three are common enough to be worth naming.

Failure one: everything becomes two minutes

The rule is a filter, and filters get gamed — by you, quietly, in your own favour. A reply that would take two minutes becomes a reply that takes twelve because you looked something up. Three two-minute tasks become forty minutes because each spawned another.

By eleven you have been busy since eight and have not touched the thing that was actually due.

The fix: apply it only when you are already in the mode. Two-minute tasks during an email session, yes. Two-minute tasks in the middle of deep work, no — they cost the re-entry, not the two minutes.

Failure two: it becomes a hiding place

Small tasks are the safest work there is. They finish, they cannot really be done badly, and each one produces a small sense of completion. The hard thing produces none of that and might go wrong.

So a rule intended to clear trivia becomes an entirely legitimate-looking way to avoid the one task that matters. You are not procrastinating; you are being efficient. That is what makes it durable.

The fix: the hard thing goes first, before any two-minute anything. If the day begins with small tasks it will not contain a large one.

Failure three: it does not work for the tasks you actually avoid

The tasks that sit for nine days are almost never two-minute tasks. They are the message with feelings attached, the form that needs a decision, the email where you have to admit something.

Those take two minutes of typing and carry a cost the rule does not measure. Applying it to them just proves it does not work, which is a bad lesson to learn about a good rule.

The fix: treat those as their own category. They need a smaller first step — open the document, write the worst possible version — rather than a promise that it will be quick. It is not quick, and telling yourself it is makes it worse.

Where it genuinely shines

Batched admin, exactly as intended. In an email session, on a backlog, when the alternative is a list of forty items you will never look at again — the rule does real work and does it fast.

Use it inside a bounded session with a timer running, and it is excellent. Use it as a general policy for the day, and it will fill the day.