Why small tasks pile up

Updated 2026-09-12

The short answer

Small tasks accumulate from both ends at once.

They are cheap to accept — nobody refuses a two-minute request, so they arrive faster than large work does. And they are cheap to defer — each one is individually trivial, so postponing any single one costs nothing visible. Arrivals outpace departures, and the pile grows.

The pile is then expensive in a way no individual item is, which is why it never triggers an alarm until it is large.

The two rules that disagree

This is not a question with one right answer, and knowing that stops the guilt.

Shortest job first minimises average waiting time. Clearing the small things gets the queue down fastest and makes the most people waiting on you stop waiting. It is provably good at that objective.

Priority scheduling says importance should beat size. Doing six ten-minute jobs while the thing that matters does not move is a bad day, however many items got ticked.

Neither dominates, because they optimise different things — there is no universally best ordering rule. So the question is not which is correct, it is which objective today has.

The failure of doing them first

A day that opens with small tasks tends to consist of small tasks. There is always another one, they all feel finishable, and the momentum is real but points sideways.

The specific cost: the substantial work needs the hours when you have the most context loaded, and those are the ones being spent on admin. By the time the pile is clear, the capacity for the hard thing is not.

The failure of never doing them

One of them is blocking someone. A two-minute reply you have not sent may be holding up a week of somebody else's work, and the size of the blocker says nothing about the size of what it blocks. That is priority inversion, and it is the most expensive thing in the pile.

They stop being small. An unanswered message becomes a chased message becomes a meeting. Small work decays into larger work at a fairly reliable rate.

The pile itself costs attention. Every read of the list pays for it, and the background awareness of forty undone trivial things is not free even when none of them is done.

What actually works

Batch them into one block. One deliberate hour, not forty interruptions. The cost of small tasks is mostly the switching, not the doing, and switching is charged per switch rather than per minute.

Put the block after the important work, not before. Late morning or late afternoon. The pile will still be there; the capacity for the hard thing will not.

Do the blocking ones immediately, regardless. Anything holding up another person jumps the queue no matter how small. Ten minutes each morning on these returns more than any other habit in this article.

Do them properly or delete them. Half-done small tasks are the worst case: still open, still counted, no longer cheap.

Watch the arrival rate, not just the pile. If small work arrives faster than one block a day can clear, no ordering fixes it — that is a capacity fact, and the lever is refusing or redirecting some of it.

The honest ratio

Most people can absorb roughly one block a day of small work without it displacing anything that matters. If yours needs more than that consistently, the problem is the inflow rather than your handling of it — and that is worth showing someone with two weeks of planned-against-actual rather than absorbing quietly.

Where to go next

• Should you do the hardest thing first? — the same tension, from the other side
• Why to-do lists stop working — the convoy effect and arrival order
• How to say no to more work — when the inflow is the problem

Using this in Decide One

Decide One asks for a duration on each line, which makes a block of small work a visible, bounded thing rather than something that quietly fills the gaps between everything else.

A line you did not finish is unfinished. It is not marked against you, and a day you did not open is not counted.

Open Decide One — free to use, with no account. Nothing you write leaves your device.