A to-do list is a place to put things so you can stop holding them. That is a real service. It is not a plan.
The trouble starts at about item eight, when the list gets long enough to negotiate with. You pick the two that are pleasant. You tell yourself the other ten were aspirational. The one that actually mattered moves to tomorrow for the fifth time, and nothing in the system objects, because the system's only job was to remember.
Here is what a plan does instead, how to size a step so it survives a normal week, and the three reasons a step keeps moving.
Four or five steps is where the negotiating stops
A short ordered plan is a different object from a long list, and the difference is that it has no slack in it.
Four or five milestones fit in your head. They have an order, so there is a next one rather than a choice of eleven. And there is no pleasant item to retreat to, because at four items every one of them is load-bearing.
You have either cleared the next one or you have not. Both answers tell you something, which is more than a half-crossed list of twelve tells you.
Nobody climbs a wall. They climb stairs. Same height.
Worth knowing: four or five is also roughly the point where a plan can be read in one glance, which matters more than it sounds. A plan you have to study is a plan you check on good days only.
Size the step by the time you actually have
Under forty-five minutes, finishable by someone who also has a job. That is the whole specification.
The count matters less than the size. A milestone that needs a clear afternoon is not a step, it is a project wearing a step's clothing, and it will sit at the top of the plan for a month looking reasonable.
Three tests that catch an oversized step before it costs you the month:
- Can you name the first physical action? "Research suppliers" fails. "Email three suppliers from the list I made" passes.
- Could you finish it on a Tuesday after work? If the honest answer needs a Saturday, it is two steps.
- Would you bet money on clearing it this week? Under about eight in ten, cut it in half and ask again. That is the GROW framework's most useful question and it works on any single step.
Keep a fifteen-minute version of each one
The small version has to exist before the bad day, because nobody invents one on a bad day.
Every step gets two sizes: the real one, and the fifteen-minute version that still counts. One paragraph instead of the section. Twenty minutes walking instead of the run. The three slides on pricing instead of the deck.
This is not lowering the bar for its own sake. It is the difference between a week that produced something small and a week that produced a zero, and over a quarter those two compound in opposite directions.
If you only ever hit the small version, the plan is telling you the steps are the wrong size. That is useful information rather than a verdict, and week three is usually when it arrives.
The step that keeps moving is telling you something
A repeated miss is data about the step, not about you, and there are only three things it usually means.
- It is too big. It looked like one action and it is four. This is the most common by a wide margin, and the fix is a knife rather than more resolve.
- It is blocked on someone else. Those need to start earlier than feels natural and they need to be labeled as blocked, because waiting and avoiding feel identical from the inside.
- It does not actually matter. More common than people admit. Spotting it is a result, not a failure. Dropping it is fine, and carrying it around for another year is worse.
A list that silently re-adds the same item is not neutral. It is quietly teaching you that the items are optional.
Go further: this is most of what a weekly review is for. One of the three questions is what got in the way, and answering it specifically is how a step that is secretly too big stops being carried forward.
Write the plan with the order in it, not just the contents
Two steps that could happen in either order are a list. A sequence is a plan.
Ordering does one thing a list cannot: it removes the daily decision about what today is for. Most days are lost in that negotiation rather than in the work, and arriving to find the next thing already chosen is what closes it.
Where the order is genuinely open, pick one anyway. An arbitrary order you follow beats a correct order you keep re-deriving.
The exception is the blocked step. Those go early regardless of where they belong, because the waiting happens in parallel with everything else and you only get that for free if you started it.
Two goals at a time, not eight
The same logic applies one level up, and it is the level people skip.
A plan of four steps is useless if you are running six of them at once. Six plans is twenty-four steps, which is a long list again with extra ceremony on top.
Two at a time is a reasonable ceiling for someone with a job, and the reason to hold it is not modesty. It is that two goals can both have a defended hour in the week and six cannot. Finding one hour that survives a bad month is hard enough once.
The plan, in five lines
- Four or five steps, in order, each one under forty-five minutes
- A fifteen-minute version of each, written down before you need it
- Blocked steps started first, and labeled as blocked
- Two misses on the same step means resize it, not repeat it
- Two goals running at a time, maximum
Common questions
How is a milestone different from a task?
A milestone is a state you can check. A task is an action you can perform. "Chapter two drafted" is a milestone; "write for an hour" is a task. Plans work better in milestones because a milestone cannot be satisfied by activity that produced nothing, which is the failure a task list is blind to.
What if the goal genuinely needs more than five steps?
Then five is the first five, and step five is "work out the next five". Long plans are guesses about conditions you have not met yet, so the back half is fiction you will rewrite anyway. Plan to the horizon you can see.
Should the steps have dates?
The next one should. The rest should have an order and nothing else. Dating step four in advance produces a plan that is visibly behind by week two, and a plan that is visibly behind gets avoided rather than corrected.
Is a checklist app enough for this?
For holding the steps, yes, and most of them do it well. What a checklist cannot do is notice that the same item has rolled over three times, because rolling an item over is a supported feature rather than an event. If you use one, add the manual step of reading last week's list before you write this week's.
The takeaway
The difference between a list and a plan is what happens when you miss something. A list absorbs the miss and re-presents the item tomorrow, unchanged, forever. A plan of four ordered steps makes the miss visible on the second occurrence, which is early enough to fix the step instead of abandoning the goal.
That is the whole argument for keeping it short. Not tidiness. Feedback.
One thing to do today. Open whatever holds your goal, count the open items, and if there are more than five, move everything below the fifth into a separate list called later. The plan is what is left.
