Is a long to-do list bad?
The short answer
No — as long as it is a capture list and not a plan. Those are two different objects, and almost every complaint about long lists is really about one object being asked to do both jobs.
A hundred captured things cost you nothing in throughput. What costs you is how many things are simultaneously open — started and unfinished — and that is a completely different number, usually much smaller and almost never counted.
The guilt is misplaced. The confusion is the actual problem.
The distinction that resolves it
Capture is everything you might do. Its job is to be complete, so nothing lives in your head. Length is a feature: you cannot choose between options you have not written down.
A plan is what you have committed to for a bounded period. Its job is to be small enough to be true, and a plan you cannot finish is not a plan.
When these are the same document, the plan inherits the capture list's length and stops being a decision. That is the moment a list stops working — not when it got long, but when nobody separated the two. More on how lists degrade.
Why length genuinely does not cost you
By Little's Law, average completion time is proportional to work in progress — items in the system, meaning started and not finished. An unstarted captured item is not in the system. A hundred of them do not make anything slower.
Three half-written documents, on the other hand, are three items in progress, and they make each other slower whether or not any of them appears on today's list. The arithmetic in full, including what it does not claim.
So the number to watch is not the list's length. It is: how many things have I started and not finished? That one is usually higher than people expect, because things stop feeling active without ever being closed.
When a long list does start costing something
Three specific cases, and all of them are fixable without shortening it:
When reading it produces dread rather than options. If opening the list is unpleasant, you will stop opening it, and then it stops working as capture too. That is the real signal to delete — not the count.
When it is stale enough to be untrustworthy. A list with a hundred items where forty are done, obsolete or duplicated is not a hundred-item list; it is a sixty-item list with forty pieces of noise, and every read pays for the noise.
When choosing from it takes real time. If picking today's three takes fifteen minutes, the list needs a short top rather than a full sort. How to keep a backlog usable.
What to actually do
- Keep capturing everything. This is the half of the habit that is not broken, and it works.
- Never plan from the whole list. Take a small number for today — three is practical — and close the rest.
- Delete without ceremony, regularly. Anything untouched for three months is not going to happen. Removing it is not failure; it is removing a cost that was returning nothing.
- Count what is open, not what is listed. Once. It is the number that actually predicts how your week goes.
The thing nobody says
A long list is often evidence that you can see your work clearly, which is a good position to be in. The alternative is not a short list — it is the same amount of work, unwritten, running in your head where it takes up room and is never fully visible.
Given the choice between a hundred things written down and a hundred things half-remembered, the written version is better in every respect. It just must not be mistaken for a plan.
Where to go next
- Little's Law, applied to a working day — why open items are the number that matters
- Why to-do lists stop working — what happens when capture and plan are one surface
- How to prioritise a long backlog — keeping a big list usable
Using this in Decide One
Decide One keeps the two surfaces apart on purpose. Everything competing for your attention stays visible and unbounded; what you chose for today is small and the bound is enforced rather than suggested.
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.
Decide One is a priority instrument. Bring the day into view, choose a method, give the first thing real time.