How to break a big goal into steps you can actually do this week
The gap between "get fluent in Spanish" and anything you can start on a Tuesday. A three-layer method, the test for whether a step is small enough, and the four ways decomposition usually goes wrong.
Here is the situation this page is about. You have written down something true and important — get fluent in Spanish, change careers, finish the thesis, get properly fit — and then you sat there. Not because you are lazy, and not because you need more motivation. Because the sentence you wrote does not tell you what to do in the next hour, and no amount of wanting it converts it into an action.
That gap is not a character flaw. It is a structural problem, and it has a structural fix.
Why "make it smaller" is not enough advice
Everyone has heard "break it into smaller pieces". The advice is correct and almost useless, because it leaves out the two things that are actually hard: how far down to cut, and how much to cut at once.
Cut too little and you get "study Spanish", which is the original goal wearing a smaller hat. Cut too much and you spend a satisfying Sunday afternoon producing a hundred-item plan covering the next eighteen months, which is procrastination that produces a document. The document then goes stale within a fortnight, because you learn things by starting that you could not have known while planning.
What works is a fixed number of layers and a rule about which layer you are allowed to detail.
Three layers, and only three
Layer one: the goal
One sentence about where you are trying to get to, plus a date and a way of knowing. "Get fluent in Spanish" fails the second half — there is no moment at which it becomes true. "Hold a thirty-minute conversation with my colleague in Spanish without switching to English, by the end of June" can be checked.
People resist this because a checkable goal can be failed, and a vague one cannot. That is precisely the reason to do it. You cannot be behind on a vague goal, which sounds comfortable until you notice that you also cannot catch up on one. Northstar asks for a deadline and a measure for this reason, and computes a goal-health reading from the days remaining against how far along you are — not to judge you, but so that a goal you have not opened in three weeks says something before the deadline does.
Layer two: phases
Two to four stretches of weeks, each with an end you could recognise. For the Spanish example: get a daily habit running at all; reach the point of understanding slow speech; start speaking with a person rather than an app.
Phases are the layer most people skip, and skipping them is why long goals feel like an undifferentiated plain. Without phases, week six is indistinguishable from week two — same activity, no arrival, no evidence of movement. With them, you finish something every few weeks, which is where the sense of progress actually comes from.
Keep phases sequential and keep them few. If you have written eight, you have written a project plan, and you will not follow it.
Layer three: steps
Actions. Things that appear on a specific day and are either done or not done. "Do one Anki session." "Book the first lesson." "Watch one episode with Spanish subtitles."
This is the only layer you ever tick, and it is the layer that connects your calendar to your ambition.
The rule that keeps this from becoming a planning hobby
Only detail the phase you are in.
Name all your phases up front — that costs a minute and gives the goal a shape. Then write steps for the current phase only, and leave the rest as titles. When you finish a phase, write the next one's steps with everything you have learned in the meantime.
This is not a productivity nicety. Plans for the distant parts of a goal are written by the version of you who has done none of it, and they are usually wrong in ways you cannot see yet. A plan for phase three written before phase one is guessing dressed as diligence.
The Tuesday-morning test
For every step you write, ask: if this appeared on an ordinary Tuesday morning — not a fresh, motivated Monday, an ordinary tired Tuesday — would I know exactly what to do first, and could I do it?
If the answer is no, one of these is true:
- It is still a phase in disguise. "Set up the website" is not a step; "buy the domain" is.
- It has a hidden prerequisite. "Write chapter two" is blocked on an outline that does not exist yet. Make the outline the step.
- It is too big for the energy you will actually have. Halve it. A step you do at eighty percent is worth more than a step you skip at one hundred.
- It is vague about the finish. "Work on the thesis" has no end, so it never gets started. "Write six hundred words" does.
A step should be small enough that starting it requires no decision. Most of the friction in doing things is not the doing — it is the moment of working out what doing it means. Decomposition is an attempt to spend that cost once, calmly, in advance, rather than every morning at the point of lowest willpower.
Which steps repeat, and which happen once
Most working goals turn out to be two or three recurring steps plus a short queue of one-offs.
The recurring ones are the engine: one Anki session daily, one lesson weekly, three runs a week. They should be genuinely small, because they have to survive a bad week — the size that works is the size you would still do on the worst day of the month, not the size you can manage on the best.
The one-offs are the unblockers: book the lesson, buy the shoes, ask your colleague if they would speak Spanish with you on Fridays. They are individually trivial and collectively decisive. A goal stalled for two months is very often stalled on one ten-minute one-off that never got written down as a step.
Northstar makes this distinction explicit: a step is marked once, daily or weekly, and the recurring ones keep coming back without you re-typing them. You can also give a step a rough time estimate — which is worth doing mainly because a week later you will notice that the thing you called twenty minutes has never once been twenty minutes.
What "this week" should actually look like
- Pick one goal to be active. One. Two goals with daily steps is usually the practical ceiling for someone with a job.
- Confirm which phase you are in, and look only at that phase.
- Choose the two or three recurring steps that are the engine of that phase.
- Add at most two one-offs for the week. Prefer the ones that unblock something else.
- Each morning, do the one step the day hands you first. If you get further, good, but that one is the commitment.
- At the end of the week, look at what was actually completed rather than what you intended. If a step was skipped four days out of five, the step is wrong — not you. Rewrite it smaller.
That last instruction is the one that matters most and gets followed least. A step you keep failing to do is data about the step. The usual response is to feel bad and re-commit at the same size, which produces the same result. Halving it feels like giving up and is in fact the only move that changes the outcome.
Four ways this goes wrong
Planning the whole thing before starting any of it
The most enjoyable failure mode. Building the plan feels like progress, gives an immediate sense of control, and requires none of the discomfort of the actual work. If you have a beautiful eighteen-month plan and no completed steps, you have been doing a different activity.
Steps that are really outcomes
"Lose two kilos" is not a step, because you cannot do it on Tuesday — you can only do things that might lead to it. Anything you cannot directly perform belongs at the goal or phase layer. Steps are always verbs you control.
Too many active goals
Five goals with daily steps is not five times the progress; it is a schedule nobody can keep, followed by the conclusion that you are undisciplined. The honest move is to have one or two active goals and to let the others sit as written-down intentions with no steps attached. Writing them down is enough to stop them nagging.
Confusing logging with doing
Tracking systems can quietly become the goal. If you spend more time arranging the structure than executing steps, simplify the structure. The tracker exists to answer one question at the end of the month — did my days go where my goals were — and if it is not answering that, it is furniture.
What decomposition cannot do
It will not create motivation. If you genuinely do not want the goal, a well-cut plan just gives you a tidier view of something you are not going to do, and the useful discovery is that you can drop it. Some goals are inherited from a version of you that has moved on, and deleting them is a legitimate outcome.
It also will not survive a life event. Illness, a new job, a family crisis — these break plans and there is nothing to fix in the method. The correct response is to shrink the goal until it fits the life you currently have, not to abandon it and not to keep failing a plan built for different circumstances.
What decomposition does reliably is remove the excuse of not knowing what to do next. That is a smaller claim than most goal advice makes, and it is the one that holds up.
Frequently asked questions
How small should a step be?
Small enough that starting it requires no thinking. A useful test is whether you could do it on an ordinary tired Tuesday rather than on a motivated Monday. If you would need to work out what it means before you could begin, it is still too big.
How many goals can I run at once?
Practically, one or two with daily steps if you have a job. Others can sit written down with no steps attached — recorded so they stop occupying your attention, but not competing for your mornings.
Should I plan the whole goal before I start?
Name all the phases, then write steps only for the phase you are in. Plans for later phases are written by the version of you who has done none of the work, and they are usually wrong in ways you cannot see yet.
What do I do with a step I keep skipping?
Rewrite it smaller, or delete it. A step you skip four days in five is information about the step, not about your character. Re-committing at the same size is the one response that reliably produces the same result.
Try it on a real goal
Northstar cuts a goal into phases and phases into steps you write yourself, then names one next action a day. Its statistics come from your own entries, and it shows an empty state rather than a placeholder number. Everything stays on the device — no account, no sync, no analytics, no notifications. Free, with no in-app purchases.