Most weekly plans are lists of everything the writer wishes were true, produced on Sunday evening in an optimistic mood, and abandoned by Tuesday lunchtime. The failure is predictable and it has three causes: the plan contained more than a week holds, it was a list rather than a schedule, and nothing was consciously dropped.
A plan that survives contact with the week has to do the opposite of what feels productive on Sunday. It has to be small.
Look back before you look forward
Start with last week, not next week. Ten minutes.
- What actually happened? Skim your calendar and notes. This is not nostalgia — it is the only reliable data you have about how much you actually get done, and your estimate without it will be wrong in the same direction it is always wrong.
- What did you say you would do and didn't? Two categories: things still worth doing, and things you should now delete. Be willing to delete. A task carried for six weeks is telling you something.
- What went well? Worth thirty seconds, because you will otherwise only ever review failures.
This is the review half of the process, and skipping it is why plans repeat the same overcommitment indefinitely. See the weekly review for the fuller version.
Empty your head
Before choosing anything, get everything out. Every open loop, commitment, half-thought and nagging obligation, written down without filtering.
This works for a reason beyond tidiness: unfinished tasks occupy attention continuously while they remain only in your head, an effect named after Bluma Zeigarnik. Writing them down releases most of that, provided you trust the list. See the Zeigarnik effect and the note inbox.
If you already capture consistently through the week, this step takes two minutes because the list mostly exists.
Pick three
Not ten. Three things that, if they were the only things you achieved, would make the week a success.
Three is not arbitrary. It is roughly what a week with meetings, admin and the unexpected in it will actually absorb. A list of ten produces a week where everything is half-done and nothing is finished, and that is a worse outcome than three done and seven honestly deferred.
Choose them with the urgent/important distinction in mind, because the important-but-not-urgent things are the ones that will lose by default if you do not protect them deliberately — see the Eisenhower matrix.
Put them in the calendar, specifically
A task on a list has no time attached and will lose to anything that does. A task in the calendar has already claimed its hour.
- Block the actual time, with the actual first action named. "Tuesday 9–11, draft the pricing section" — not "work on proposal".
- Put demanding work in your best hours and admin in your trough. See time blocking.
- Decide when, not just what. This is Gollwitzer's implementation-intention finding: deciding in advance when and where you will do something substantially raises follow-through compared with intending to do it.
Leave half the week empty
The single most common planning error is scheduling to capacity. A week that is fully booked has no room for the thing that always happens, so the first interruption destroys the plan and the plan gets abandoned entirely.
Aim to schedule roughly half your available hours. The rest will fill itself — with the unplanned, the overrunning, and the genuinely urgent. If you end a week with unscheduled time left, that is not waste. That is the buffer working.
Decide what you are not doing
Write it down explicitly. "Not this week: the website redesign, the conference proposal, replying to the newsletter backlog."
This feels like an odd step and it is the one that makes the rest hold. An unwritten deferral is not a decision — it stays live in your head, generating guilt and competing for attention. Written down, it is handled, and you stop relitigating it on Wednesday.
Mid-week, look once
Ten minutes on Wednesday. Are the three still the right three? What has arrived that changes them? Adjust deliberately rather than drifting.
A plan you never look at again was journaling, not planning.
The whole thing, in order
- Ten minutes on last week
- Empty your head onto a list
- Pick three outcomes
- Block them in the calendar with named first actions
- Leave half the week free
- Write down what you are not doing
- Check once mid-week
Thirty minutes, once a week. If it is taking two hours, the system has become the work — a failure mode worth watching for, and the reason PKM mistakes exists as a page.
Why capture is the precondition
Weekly planning only works if the raw material is already there. If you spend the first twenty minutes trying to remember what you committed to, the plan is built on whatever you happened to recall, which systematically excludes the things you forgot — and those are frequently the ones that matter.
A week's worth of captured thoughts, commitments and half-ideas turns planning from an act of recall into an act of selection, which is both faster and much better. That is what a one-gesture capture tool like Clair Mind is for: not to plan for you, but to make sure the plan is built from everything, not just what surfaces on a Sunday evening.
More in how to review your notes, getting things done, and how to take notes.