How to build a habit that actually sticks

Every January I used to pick three habits. Run in the morning, read before bed, no phone at the table. By the third week I was doing none of them and had a fresh reason to think I was the problem.

I wasn’t. The plan was. Almost everyone who fails at a new habit fails the same way, for the same handful of reasons, and none of them is a character flaw. This is the long version of what I wish someone had told me: how habits actually form, why the usual approach breaks, and a way to start that has a real chance of still being there in March.

The usual approach goes: decide to change, feel motivated, do the big version of the new thing for a few days, get tired or busy, miss once, feel bad, miss again, quietly stop. If that sounds familiar it is because nearly everyone does it, and the design of the plan guarantees the ending.

Three things are wrong with it. First, it relies on motivation, which is a feeling, and feelings are not on a schedule. Second, it starts at full size, so the cost of doing it on a bad day is high. Third, it treats a missed day as a verdict instead of as a data point.

The fix for all three is the same: make the habit so small and so well-placed that you can do it on the worst day of the week without wanting to. Then let repetition do what motivation could not.

A habit is a behavior your brain has learned to run when it sees a particular cue, without you having to decide. Brushing your teeth is the standard example. You do not weigh the pros and cons at 11 p.m. You see the sink, and your hand reaches for the brush.

Researchers who track this find that somewhere around 40 percent of what we do each day is habitual — done in the same place, at the same time, with almost no conscious thought (Wood, Quinn & Kashy, 2002). That is a huge amount of your life running on autopilot. The goal is not to remove the autopilot. It is to put something better on it.

The key ingredient is not effort. It is a stable cue. The same context, every time. When the context wobbles, the habit wobbles. When you move, change jobs, or go on vacation, old habits break and new ones fail, not because you became weaker but because the cues vanished.

You have probably heard that a habit takes 21 days. That number comes from a 1960s plastic surgeon writing about how long patients took to get used to a new face. It has nothing to do with habits, and it has been repeated for sixty years because it is short and pleasant.

The study that actually measured this followed people forming one new daily habit — eating a piece of fruit with lunch, running for fifteen minutes — and asked each day how automatic it felt. The median was 66 days. The range ran from 18 days for the easiest behaviors to more than 250 for the hardest (Lally et al., 2010).

Two useful things fall out of that. One: if it still feels like effort after three weeks, nothing is wrong. You are on schedule. Two: the simpler the behavior, the faster it locks in. Which is the argument for starting small, and it is a strong one.

Here is the rule that changed the most for me: if the habit takes more than two minutes, shrink it until it does.

Not “run three miles.” Put on running shoes and step outside. Not “read for thirty minutes.” Open the book and read one page. Not “meditate for twenty minutes.” Sit down and take three slow breaths.

This feels ridiculous the first time. It is meant to. The point in the first month is not to get fit or well-read. The point is to become the kind of person who shows up at that time, in that place, every day. Once showing up is automatic, making it longer is easy. Doing it the other way around — starting long and hoping showing up follows — is the plan that fails.

James Clear calls this the two-minute rule in Atomic Habits. BJ Fogg, who has run a behavior design lab at Stanford for years, calls the same idea Tiny Habits and goes even smaller: floss one tooth. The research and the practitioners land in the same place. Scale down until it is almost embarrassing, then do that.

A small habit also passes the bad-day test. Anyone can put on running shoes when they are tired, sad, or behind at work. Almost nobody can run three miles in that state. The habit that survives is the one you can do on your worst day, not your best.

The second rule: do not rely on remembering. Attach the new habit to an old one.

You already have dozens of rock-solid habits. Kettle on in the morning. Sitting down at your desk. Brushing your teeth at night. Locking the front door. Each of those is a cue that fires every day without fail. Borrow one.

The format is: After I [existing habit], I will [tiny new habit]. After I pour my coffee, I will write one line in a notebook. After I sit down at my desk, I will write down the one thing I need to do today. After I brush my teeth at night, I will lay out tomorrow’s clothes.

Pick the anchor carefully. It needs to happen every day, at roughly the same point, and it needs to be physically near where the new habit will happen. “After lunch” is vague. “After I put my lunch plate in the sink” is a cue. The plate in your hand is the reminder.

This works because the old habit already has the automatic quality you want. You are not building a new cue from scratch. You are stapling something small to a cue that already fires.

Small changes to your surroundings do more than any amount of resolve. Put the book on your pillow in the morning so it is in your hands at night. Leave the running shoes by the door, not in the closet. Put the water bottle on the desk, filled, before you leave the night before.

The opposite works too. If the habit you want to break is scrolling in bed, charge the phone in the kitchen. Not forever, just for a month, until the reaching-for-it reflex fades. You are not testing your discipline. You are removing the cue so there is nothing to be disciplined about.

A useful question for any habit: what would make this take one fewer step? Then do that thing today, once, so tomorrow-you does not have to.

You will miss a day. Not might. Will. Illness, travel, a bad night, a Tuesday that gets away from you. The people who keep habits are not the people who never miss. They are the people who have a rule for missing.

The rule is: never miss twice.

One missed day is nothing. The research on habit formation found that a single skipped day had no measurable effect on how automatic the behavior became. What breaks a habit is the second miss, because the second miss is the start of a new habit — the habit of not doing it.

So when you miss, do not make it up. Do not double tomorrow. Do not review why you are like this. Just do the tiny version the next day, at the usual anchor, and carry on. If you cannot manage even the tiny version, do something absurdly small in the right place: stand in the spot where you would have done it. Touch the book. Put on one shoe. It sounds silly and it keeps the cue alive.

A visible record does two things. It reminds you, and it makes the streak something you do not want to break.

The simplest version is a wall calendar and a pen. Do the habit, put an X on the day. The story gets told about Jerry Seinfeld writing jokes this way; whether or not he actually did, the method is sound. After a week the row of X’s is its own reason to continue.

You can use an app. Plenty of people do. But for most of the people I know, the paper version wins, for a reason that sounds small and is not: the calendar is on the wall in the room where the habit happens. The app is in your pocket, behind a lock screen, next to everything else that wants your attention. The tracker should be part of the cue, not a competitor to it.

Track one habit at a time. If you are tracking five, you are running a project, and the tracker becomes one more thing to feel bad about. If you want something purpose-built instead of a plain calendar, here is a short list of trackers that are not apps.

This is the part people skip. One habit at a time. Not three, not a morning routine, not a new life. One.

Each new habit costs attention in the first weeks, and attention is the thing you have least of. Three new habits means three things to remember, three anchors to hold, three streaks to feel guilty about. Something gives, and usually all of them go together.

Pick the one that would make the most difference if it were still there in three months. Give it its own anchor and its own month. When it feels boring, when doing it is no longer a decision, add the next one. A year of this gives you eight to ten real habits, which is more than most people build in a decade of January resolutions. If the habit you want is about getting work done, the productivity system is built to sit on top of this one.

If you have done all of the above and the habit still keeps dying, one of a few things is usually going on.

The habit is still too big. This is the most common one. Cut it in half again. If “read one page” is not happening, make it “open the book.”

The anchor is wrong. It does not happen every day, or it happens in a different place from the habit, or it is too vague. Change the anchor, not the habit.

You do not actually want it. Sometimes we pick habits because we think we should. If a habit has failed four or five times and you feel relief when you drop it, listen to that. Pick something you want instead.

Something bigger is in the way. If you are not sleeping, if you are running on empty, if getting through the day is already the whole job, a new habit is not the fix and it is not the priority. Look after the basics first. If it is more than tiredness, the Help page has people who are trained for exactly that.

If you want to start tonight, here is the first week. It is deliberately small. Do not upgrade it before day eight.

1. Tonight. Pick one habit. Shrink it to two minutes or less. Write it in this form: After I ___, I will ___. Put a calendar or a sheet of paper where the habit will happen.
2. Day 1. Do the tiny version at the anchor. Mark the day. That is the whole task.
3. Day 2. Same. Notice whether the anchor actually fired. If you forgot, the anchor is the problem, not you. Move it closer to where the habit happens.
4. Day 3. Same. Do one thing to make tomorrow easier: lay something out, move something, remove one step.
5. Day 4. Same. If you feel like doing more, you may, but the tiny version is what counts and it is all you have to do.
6. Day 5. Same. If you missed a day this week, tonight is the never-miss-twice day. Just the tiny version.
7. Day 6. Same.
8. Day 7. Same. Look at the marks. Seven X’s means it is working. Do not add a second habit yet. Keep going, and expect it to feel like effort for another month or two. That is what forming looks like.

That is it. No app, no system, no morning routine. One small thing, in the same place, after the same cue, every day, with a rule for the day you miss. It is less exciting than a new life. It is also the version that is still there in March.

• Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998–1009.
• Wood, W., Quinn, J. M., & Kashy, D. A. (2002). Habits in everyday life: Thought, emotion, and action. Journal of Personality and Social Psychology, 83(6), 1281–1297.
• Clear, J. (2018). Atomic Habits. Avery.
• Fogg, B. J. (2019). Tiny Habits. Houghton Mifflin Harcourt.