Why annual goals collapse by February — and what to do instead of restarting
The failure is built into the format, not into you. What a year-long goal does to feedback, why the restart instinct makes it worse, and what to do from the middle of a broken plan.
It follows a shape most people recognise. Late December, you decide what next year will be about. The first week is genuinely good. The second is fine. Somewhere in mid-January a work crisis or a cold takes out four days, you tell yourself you will pick it back up on Monday, and by early February the goal has quietly stopped existing. Not dramatically — you did not decide to abandon it. It just stopped being something you were doing.
The usual explanation is that you lacked discipline. That explanation is popular because it is simple, and it is wrong often enough to be worth replacing. Annual goals have specific structural properties that make them fail, and you can fix a structure.
Four reasons the format itself breaks
A year is too long to give you any feedback
A goal you can only evaluate at the end gives you no information for eleven months. Every activity that people sustain has a short feedback loop: you can see the room getting cleaner, the code compiling, the weight on the bar. A twelve-month goal has none, so between January and December you are running on belief alone, and belief is not a renewable resource.
The plan was written by someone who had not done the work
The December version of you had never done the thing. That person estimated how long it takes, how much energy it costs, and what would get in the way — with no data. Those estimates are systematically optimistic, and not because you are foolish: predicting your own future effort is known to be one of the things humans are consistently bad at.
So by mid-January you are not failing the work. You are failing a schedule invented by someone with no experience of it, and you are treating their guess as a fact about you.
It is framed as all-or-nothing
"Run every day this year", "read fifty books", "no sugar in 2026" — the format has exactly two states, and one missed week moves you permanently into the failed one. There is no version of "run every day" that survives a week of flu. So the flu does not cost you a week; it costs you the whole goal, because the goal was defined such that it can no longer be achieved.
This is why the fourth reason follows so reliably.
January is chosen by the calendar, not by your life
The first of January is not connected to whether your circumstances currently support a new habit. It is a strong cue to start — which genuinely helps at the beginning — but a cue is not a foundation. Goals that begin because the date changed tend to end when the novelty of the date does, some time in the fifth or sixth week. That timing is not a coincidence; it is the point at which the change stops being interesting and has to be carried by structure instead.
The restart instinct, and why it is the actual trap
What happens after the collapse matters more than the collapse. The standard response is to wait for the next clean boundary — the first of the month, after the holiday, next January — and start again from zero with a better plan.
Three things are wrong with that.
It throws away what you did. Eleven runs in January are eleven real runs. Treating them as void because you missed the twelfth is an accounting choice, not a fact, and it is the one that makes the work feel wasted.
It teaches you that goals are things you begin rather than things you do. Restarting is emotionally pleasant — a fresh plan is exciting in a way that resuming a broken one is not — and if you do it enough, the beginning becomes the whole experience. Some people have started the same goal five Januaries running and have never once been in the middle of it.
And it wastes the one genuinely valuable thing the failure produced: information about why the plan did not fit. A restart with a fresh plan usually keeps the flawed assumption intact, so it breaks again in the same place at roughly the same speed.
What to do instead, from wherever you are
- Resume, do not restart. Do the smallest version of the thing today. Not a new plan, not a clean Monday — today, at whatever size fits today. Nothing has been reset except by your own arithmetic.
- Find out what actually broke. There is usually a specific cause: it took twice as long as you thought, it collided with a fixed commitment, it depended on a mood you cannot summon, it needed equipment you did not have. Fix that one thing.
- Shrink the commitment until it fits the worst week, not the best. If thirty minutes daily worked in January and collapsed in February, ten minutes daily is not a lesser goal — it is the goal that survives, and a thing you keep doing at ten minutes beats a thing you stopped doing at thirty.
- Replace the year with a horizon you can see the end of. Six weeks is long enough to change something and short enough to stay real. Then look at what happened and choose the next six.
- Change the measure from a streak to a rate. "Most days" is achievable after a bad week; "every day" is not. A rate absorbs illness and travel without turning them into failure.
- Decide what the goal is for. Some goals do not survive this question, and finding that out in February is a good outcome, not a loss.
Why six weeks rather than a year
A six-week horizon fixes the feedback problem directly. You can hold the whole thing in your head, you find out whether it is working while there is still time to change it, and you get an actual ending — a moment where you look at what happened and decide the next one. A year has one ending and it is too far away to steer by.
It also makes the plan honest. You are planning for a period you can roughly foresee, rather than for a December you know nothing about. When the six weeks are up, the next plan is written by someone who has done the work, which is the whole point.
This is the structure Northstar is built around: a goal holds phases, and a phase is a stretch of weeks with an end you can recognise. You detail only the phase you are in, and the rest sit as titles until you get there.
Streaks: useful, and dangerous in a specific way
Streaks work. Not wanting to break a run of days is real motivation and worth having. The danger is narrow and worth naming: a long streak makes the cost of one missed day enormous, so the day you inevitably miss takes the whole practice with it. The mechanism that carried you for eighty days is the same one that ends it on day eighty-one.
The mitigation is to keep the streak as an observation and not as the goal. Northstar counts one and shows it, and when you break it the count resets and nothing else does — no steps are deleted, no progress rolls back, nothing locks. A gap in the log is a gap in the log. If you find yourself protecting a number rather than doing the thing, the number has taken over and you should stop looking at it.
The review that actually tells you something
At the end of a six-week block, the question is not "did I hit the target". It is "did my days go where my goals were". Those are different questions and the second one is more useful, because it explains the first.
Concretely: how many days did you do anything at all, what share of what you planned got done, and how much of your recorded effort was attached to the goal rather than to everything else that happened. That last figure is uncomfortable and informative — a month where almost nothing you did was connected to your stated goal is not a discipline problem, it is a mismatch between what you said you wanted and what your life is currently arranged to produce.
Northstar reports these over seven and thirty days, computed from the entries you made. When there is nothing to compute from, it shows an empty state instead of a number, because an app that displays a plausible-looking figure on the day you install it has taught you to ignore its figures.
What this does not solve
None of this helps if the goal is not actually yours. Plenty of annual goals are inherited from other people's expectations or from an older version of yourself, and no structure makes you pursue something you do not want. February is a good time to find that out and drop it.
Nor does it help with circumstances. If you are in a period where the goal genuinely does not fit — a newborn, a crisis, a punishing quarter at work — the right move is to shrink it to almost nothing and keep it alive, or to shelve it deliberately with a date to revisit. Both are better than failing a plan built for a life you are not currently living.
Frequently asked questions
Why do most New Year resolutions fail in February?
Because the format gives no feedback for months, the plan was written by someone who had never done the work, and one missed week moves an all-or-nothing goal permanently into the failed state. The timing lines up with the point where novelty stops carrying it and structure has to.
Should I wait until next month to start again?
No. Waiting for a clean boundary is the habit that turns goals into things you begin rather than things you do. Do the smallest version today, at whatever size today allows.
Are streaks good or bad for goals?
Both. Not wanting to break a run is real motivation, but a long streak makes one missed day catastrophic, so the day you miss can end the practice entirely. Keep the streak as an observation rather than the goal, and prefer a rate — "most days" — as the measure.
What is a better length than a year?
Around six weeks. Long enough to change something, short enough that you can see the end, and it produces a real review point where the next plan is written by someone who has actually done the work.
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.