What to Do When You Break a Streak and Want to Quit
When you break a streak, do the habit again today, at whatever size today allows, and then deal with the number. One missed day doesn't undo weeks of practice. What it breaks is the count, and the count has a pull of its own. In experiments on streaks, people who had done exactly the same thing chose to do it again less often if their log showed a broken streak than if it showed an intact one. What moved was the next choice, which is a smaller thing than quitting, and it's the choice in front of you today. So if your app offers a repair, take it now, and keep a second count that one bad day can't knock to zero.
Say it was Portuguese. You had 143 days in a row in a language app, and you lost the streak on the night of a friend's wedding, with your phone dead in a coat pocket. The next morning the counter says 0. You know that number is wrong about you. If you're honest, you also feel a small pull to try Italian instead, or to let the whole thing lapse for a while.
Why does a broken streak make you want to quit?
Jackie Silverman and Alixandra Barasch, who study consumer behavior, ran seven studies on streaks, published in the Journal of Consumer Research in 2023. People whose log showed an intact streak went on to choose the logged activity more often than people whose log showed a broken one. What drove it is the striking part. In their words, the effect was "independent of actual past behavior and depends solely on how that behavior is represented within the log."
Their own summary of one experiment gives a sense of the size. Everyone did the same four strength exercises in a row, and some were then told the fourth hadn't been logged. Offered another strength exercise or a different kind, about 66 percent of the people with an intact streak chose strength again. Among those whose log showed a break, about 58 percent did. That's eight points on a single choice, a tilt in what people picked next and not a count of people who gave up. But every one of them knew they'd done four in a row. Only the log differed.
The pull toward Italian turned up in their work too. Part of the research was done with a language-learning app, and Barasch has said that when a streak broke, people would switch to a new language as opposed to building a new streak in the same one. Her explanation is that people turn to something new "because their record no longer represents them." Hold that loosely for your own case. In one language experiment the authors have described, participants were learning Portuguese, Hawaiian or both in an app still under development, which is a long way from 143 days into a language you chose yourself.
So the urge to quit is real, but if these experiments are a guide, it's aimed at the wrong thing. It answers the display. Your Portuguese is where it was on day 143; only the number moved. And if the display is what tilts the choice, the display is also the thing to work on. If the same urge ended earlier habits too, it may be the answer to why you keep quitting things you start.
Does missing one day undo the habit?
Probably not. The study people quote most on this, by Phillippa Lally and colleagues at University College London, was published in the European Journal of Social Psychology in 2010. It recruited 96 volunteers to repeat one eating, drinking or activity behavior of their choice every day for 12 weeks, and 82 of them gave enough data to analyze. Missing a single day didn't materially affect how their habits formed, and a single day is what you missed. You can still read the menu.
So practice today. If today can only hold a smaller version, do the smaller version; picking a size that survives your bad days is worth doing once the dust settles. Don't pay the missed day back with a double session tomorrow. Nothing is owed, and a doubled lesson only sets a bar you'll have to clear again the day after.
Does it matter whose fault the break was?
More than you'd expect. Silverman and Barasch found that a broken streak's effect on what people chose next was stronger when they put the break down to themselves than when they put it down to something outside them.
That isn't a reason to go looking for an excuse. The wedding was an outside cause, and saying so is accurate. But take a different break. You've done a drawing a day for 61 days, and last night you scrolled your phone until midnight with the sketchbook zipped inside a bag by the door. That one was yours.
When it's yours, keep the blame the size of the cause. "I have no discipline" is a verdict on your whole character drawn from one evening. "The sketchbook was in the bag, and I didn't think of it once I sat down" is still your fault, but it's a cause the size of a bag, and you can fix it tonight by leaving the sketchbook open on the table. There's no promise the smaller blame stings less. It's just the only kind that gives you something to do.
Is a streak freeze or a streak repair cheating?
No. In Silverman and Barasch's studies, a broken streak's effect was weaker when people had a chance to repair the streak. A repair taken the morning after is a chance of that kind, and by the same reasoning a freeze set aside before a busy weekend does the job a day early: one bad night never shows up as a break. If your app offers either and it keeps you practicing, use it without guilt. Its rules were set before you broke anything, and you're playing by them.
And neither has to hide the miss. Oathcraft, the daily habit tracker behind these guides, is one example. You end each day there by marking whether you did the thing you swore to do, and the app keeps two day counts, which it calls Run and Streak. A repair can hold them through a missed day, and it fits a night like the one with the sketchbook, when the phone was in your hand. Close that day with the drawing undone and it's saved as a Break. Repair the Break by the end of the following day (a day there ends at 3:00 AM) and both counts hold where they were without gaining a day, provided it's your first Break since a day you kept. Nothing is deleted. The Break remains in History and the repair is entered alongside it, so the number that carries on has the miss on record under it, and you can go on believing it.
The app's nearest thing to a freeze is a Rest day, and it fits a night you can see coming, like the wedding. You get one in every seven days. You mark the day as Rest, with no lesson done; both counts stay where they are, neither goes up, and History lists the day as Rest, not as kept. But a Rest has to be declared on the day itself, before that day ends. For the wedding, that meant before the phone died. The morning after is too late. A finished day can't be made a Rest, and a day nobody closed has already become a Break by default, which puts both counts at zero. A repair only holds a Break you closed yourself.
A Break repaired late, or not at all, sets both counts to zero. So does a second Break that arrives before you've kept another day, even if you repair it, and a Rest in between doesn't count as that kept day. Beyond the repair and the Rest there's nothing: no stock of freezes or tokens to build up, and no paying to get a number back.
Should you start counting from zero again?
Not if you can get the old number back. If you took a repair, or a freeze covered the night, the count carries on and you haven't started over. If your app offers neither, then yes, the counter starts again today, and there's no use arguing with it. Just don't make it the only number you believe.
The number worth believing is the one you could say to a friend and mean. For the Portuguese, that's 143 of the last 144 days. That sentence is true in a way the 1 on the screen isn't, and it holds after a longer gap too: if the wedding turned into a week off, 143 of the last 150 is still yours. The same test works on a count that a freeze kept going. If you froze the wedding night and you'd have done the lesson with a working phone, the unbroken number still describes you. If freezes have covered six days this month, it describes someone who practices more than you do, and sooner or later you may stop believing it.
Keep the true count somewhere you'll see it, not just in your head. A note on your phone works. So does a paper calendar on the fridge, where one blank square sits among the marks instead of wiping them out.
While the break is fresh, use it to settle what counts as a day next time. Silverman has said that "what counts as a streak is malleable." She was describing an experiment with games: one group could keep a streak going by playing either of two kinds, and another had to stick to a single game for the streak to count. With your own habit, you're the one who writes that rule. A drawing finished at 12:40 at night can count for the day before. A day in bed with flu can be a rest day. Write those rules down this week, for the next break, not for this one. A rule invented the morning after to rescue a number is one you'll never quite believe.
How do you know when a streak is really over?
Sometimes a break is telling you something true. The sketchbook may be full, and filling it was the point. The Portuguese may have been for a trip that's now behind you. A habit can also settle in so deeply that counting it adds nothing. Ending a streak on purpose is fine.
If breaks keep landing on the same kind of day, every late Friday or every trip, that's a pattern rather than bad luck, and the habit probably needs new terms. How to Keep Promises to Yourself walks through changing a promise in the open instead of letting it erode.
The night of the break is the worst time to decide any of this, though, because that's when the number looks worst and the pull toward something new is strongest. Pick the practice up today and ask the bigger question in a week, when you have a few days of return to look at beside the miss. If it still looks finished then, end it deliberately.
Sources
- doi.org/10.1093/jcr/ucac029
- knowledge.insead.edu/marketing/consumer-streaks-are-motivating-key-keeping-them-alive
- www.colorado.edu/business/news/2023/04/20/research-streaks-marketing-tech-barasch
- lerner.udel.edu/seeing-opportunity/lerner-professor-researches-how-streaks-motivate-us
- doi.org/10.1002/ejsp.674