How to build a habit that sticks: attach it to something you already do

Ralph Varcoe
5 min read
To build a habit that sticks, keep it small and attach it to something you already do every day, so that the old routine reminds you to do the new one: "After I pour my coffee, I will write my three priorities." That idea, often called habit stacking, comes from BJ Fogg's Tiny Habits work at Stanford and reached a much wider audience through James Clear's Atomic Habits. It also helps to be patient with yourself, because research by Phillippa Lally and her colleagues found that new habits took a median of 66 days to become automatic.
Why do new habits fall apart after a week or two?
Picture a familiar January. You decide that this is the year you'll meditate for twenty minutes every morning, and on the first day you do it, sitting cross-legged on the bedroom floor feeling rather pleased with yourself. On the third day you do it again. By the eighth day you've overslept, the kids can't find their shoes and the meditation gets skipped, and by the ninth day you've forgotten about it altogether.
By February it has gone the way of the gym membership and the language app, and it's very easy to decide that you simply aren't a disciplined person.
In most cases, though, two much more forgivable things went wrong. The habit was too big to fit into a real morning, and the cue was too vague to notice. "In the morning" covers about four hours and dozens of small decisions, so there's no single moment that reminds you, and the habit ends up depending on willpower and memory, which are both in short supply at seven o'clock on a school day.
What is habit stacking, and where does it come from?
BJ Fogg, a behaviour scientist who founded the Behavior Design Lab at Stanford, developed the idea of anchoring a tiny new behaviour to one you already do without thinking. His format is "After I [existing habit], I will [new tiny behaviour]", and he set it out in full in his 2019 book Tiny Habits. James Clear gave it the name habit stacking and brought it to millions of readers in Atomic Habits in 2018. The appeal is that you already pour your coffee every morning, so the coffee does the remembering for you.
The question everyone asks next is how long it takes before the new habit runs on its own. Phillippa Lally and her colleagues at University College London followed people who were trying to build a new daily habit and published what they found in 2010. The median time for the behaviour to become automatic was 66 days, and the range ran all the way from 18 days to 254, depending on the person and the habit.
They also found something that should take a lot of pressure off. Missing a single opportunity didn't materially affect how the habit formed, so the familiar idea that one slip ruins everything turns out to be wrong, and so does the popular "21 days" figure. For most of us the honest answer is somewhere around two months of gentle repetition, and that's perfectly achievable.
How do you use Building Habits in PiViva?
Building Habits helps you design one habit properly. You name the cue, which is the thing you already do, then the routine, which is the new behaviour, and then the reward, which is whatever makes it feel good afterwards. Finally you rate how confident you are that it'll stick.
It helps a great deal to make the new behaviour almost embarrassingly small to begin with. "After I pour my coffee, I'll do forty minutes of admin" sounds impressive, but it's likely to collapse within a week, whereas "after I pour my coffee, I'll open my task list" is so easy that you'll do it even on a bad day. Once it's automatic you can build it up.
The cue deserves the same care. "After I put my keys on the table" is a moment you'll notice every single day, while "when I get home" is a fuzzy stretch of time that slips by without any clear signal.
Come back to the confidence rating every week or so. If it's drifting down, the habit is probably too big or the cue isn't quite right, and you can adjust either in a couple of minutes. Every habit you build here appears in your Habit Tracker, ready for a daily tick.
What if the habit still won't stick?
That happens to everyone at some point, and it usually means something small needs adjusting.
A Habit Audit is a good next step. It checks your habit against James Clear's four laws of behaviour change and shows you which one is weakest, so you can fix the part that's actually causing the trouble. If the habit keeps losing to one particular moment, such as tiredness in the evening, the sofa or the pull of your phone, an Implementation Intention lets you plan exactly what you'll do when that moment arrives.
It's also worth working on one habit at a time and giving each one a month or so to settle before you add the next. Habits built that way tend to stay, and before long you may find you've become the sort of person who simply does these things, without having to think about it at all.
Questions people ask
How long does it take to build a habit?
In Phillippa Lally's 2010 study at University College London, the median time for a new habit to become automatic was 66 days, with a range of 18 to 254 days depending on the person and the habit.
What is habit stacking?
Habit stacking means attaching a new habit to one you already have, using the format "After I [current habit], I will [new habit]." It comes from BJ Fogg's Tiny Habits method and was popularised by James Clear.
Does missing a day ruin a habit?
No, it doesn't. Lally and her colleagues found that missing a single opportunity did not materially affect how the habit formed, so the best thing to do is simply pick it up again the next day.
How small should a new habit be?
Small enough that you'd still do it on your worst day. Once it has become automatic you can gradually make it bigger.
Sources
- Fogg, B. J. (2019). Tiny Habits: The Small Changes That Change Everything. Houghton Mifflin Harcourt (opens in a new tab)
- James Clear: habit stacking (opens in a new tab)
- Lally, P. et al. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40(6), 998-1009 (opens in a new tab)