Missed a day? What a broken streak really does to your habit

What breaks is the number on the screen, and that number pulls harder than it should. Researchers have watched people treat a broken record as a reason to stop entirely: dieters, people quitting smoking, university employees in a step challenge, even people playing a trivial word game. The same research also shows that small design choices, such as a bit of built-in slack or a way to repair a miss, change how people respond.

Missed a day: the habit is fine, the streak is fragile

In 2010, Phillippa Lally and colleagues asked 96 volunteers to pick one eating, drinking or exercise behaviour and repeat it daily in the same context, such as “after breakfast”, for 12 weeks (Lally et al., 2010). Each day they rated how automatic it felt. Automaticity rose quickly at first and then levelled off, and reaching 95% of each person’s plateau took anywhere from 18 to 254 days. One sentence in the abstract matters here: missing one opportunity to perform the behaviour “did not materially affect the habit formation process.”

Keep the scale in mind. Usable data came from 82 people, the curve fitted well for only 39 of them, and automaticity was self-reported. It is still the best real-world evidence on the question, and it points one way: habit strength builds up gradually and doesn’t vanish after one gap. How long the climb takes is its own topic, covered in how long it takes to form a habit.

A streak works differently. It counts consecutive days, so a single zero wipes it out no matter how much of the habit you have built. Day 40 of a habit with one miss on day 39 looks like day 1 on a strict counter. Your behaviour has barely changed, but the picture of it has, and the studies below suggest people react to the picture.

Why a broken streak makes you want to quit

Psychologists have a blunt name for this. In a 1975 study, C. Peter Herman and Deborah Mack gave dieters and non-dieters a “preload” of food of different sizes and then measured how much more they ate (Herman & Mack, 1975). Non-dieters ate less after a bigger preload, as you would expect. Dieters did the opposite and ate more. Herman and Janet Polivy later called it the “what the hell effect”, as if the dieter were thinking: my diet is already broken today, so I might as well eat and enjoy it (Polivy & Herman, 2020).

The idea soon moved beyond food. Wendy Cochran and Abraham Tesser wrote about the “what the hell effect” in terms of goals and performance in a 1996 chapter on goal proximity and goal framing (Cochran & Tesser, 1996). Addiction research has its own version. Alan Marlatt and Judith Gordon’s relapse-prevention model describes an abstinence violation effect: after a first lapse, self-blame, guilt and lost confidence are supposed to push a person from a single slip toward a full relapse.

When that model was tested closely, the picture got messier. Thomas Kirchner, Saul Shiffman and Paul Wileyto followed 203 people quitting smoking through 1,001 lapses, recorded in near real time (Kirchner et al., 2012). Reactions to the first lapse didn’t predict relapse. Across later lapses, a drop in confidence after slipping was linked to reaching the next lapse sooner, while blaming themselves, against the theory, slightly protected people.

A streak counter is not a diet or a cigarette, and the stakes aren’t comparable. The overlap is narrower: when a rule is all-or-nothing, one breach can make the whole goal feel lost, and the feeling of being lost is what people act on.

What studies of streaks actually found

The most direct evidence comes from Jackie Silverman and Alixandra Barasch, who ran seven studies on how logged streaks affect what people do next (Silverman & Barasch, 2023). They define a streak as the same behaviour done three or more times in a row.

In a field study with a university wellness program, 980 employees joined a 30-day challenge to walk at least 7,000 steps a day, using an app that showed their current streak. People were more likely to reach 7,000 steps on a day that could extend an intact streak than on a day after a broken one. That data is correlational, so the authors moved to experiments in which everyone did exactly the same thing and only the log differed. In one, 601 online participants did four real strength exercises. Some saw a log of four in a row, others saw the run broken. Offered one more exercise, 66% of the intact-streak group chose another strength exercise, compared with 58% of the broken-streak group.

The part worth underlining: the effect didn’t depend on what people had done, only on how the log displayed it. The authors’ explanation is that keeping a logged streak alive becomes a goal in itself. Two results follow from that. The drop after a break was bigger when people blamed themselves for it. And it shrank when they could repair the streak: in the final study, 93% of people with an intact streak went on with the same game, against 69% with a broken streak and 85% when they were told that continuing would restore it.

Strict vs forgiving streaks

A strict streak treats every miss the same way: back to zero. A forgiving streak keeps a small buffer, so an isolated miss doesn’t erase the count. Here are the same three weeks of a daily habit, with misses on day 9 and day 16, counted both ways.

Both rows describe 19 completed days out of 21. One ends at 5 and the other at 21. If people respond to the log more than to their actual record, as Silverman and Barasch found, that gap matters more than it looks.

Slack has also been tested directly. Marissa Sharif and Suzanne Shu gave students and staff at a US university a personal daily step goal for four weeks (Sharif & Shu, 2021). One group had to reach it 7 days a week, another 5 days a week, and others 7 days plus two “emergency skips” a week (or eight a month) that they could apply after a miss, which filled that day in on their progress chart. On the day after a miss, people with the strict 7-day goal met their goal 37% of the time. With two weekly emergency skips it was 55%. Over the four weeks, the weekly-skip group reached the goal on 4.0 days a week, against 2.8 for the strict group and 3.1 for the plain 5-day goal, even though the 5-day goal allowed exactly as many misses.

Their explanation is perceived progress. Applying a reserve made the goal still feel on track, so people stayed committed. People in the skip groups who missed but didn’t or couldn’t apply a reserve persisted about as much as everyone else, so the benefit came from actually using it.

The largest test of a related idea is a megastudy with 61,293 members of a US gym chain, randomly assigned to 54 different four-week programmes (Milkman et al., 2021). The best performer paid a tiny bonus, worth about 9 US cents, for coming back to the gym after a missed workout.

None of these studies tested a grace day inside a habit tracker. They do point the same way: designs that treat a miss as recoverable keep more people going than designs that treat it as the end.

How to restart after you break your streak

If the streak is already gone, the research suggests a few concrete moves. None of them involve a dramatic new “day 1”.

1. Do the next scheduled session, even a small one. The day after a miss is exactly where the strict-goal walkers in Sharif and Shu’s study lost ground. A shorter walk or two pages of a book still keeps the chain of repetitions going.
2. Look at what got in the way rather than passing a verdict on yourself. In the smoking study, falling confidence after a lapse predicted the next lapse sooner. A specific obstacle you can plan around is easier on your confidence than a general conclusion about your willpower.
3. Decide the comeback before you need it. A plan like “if I miss my evening walk, I walk before work the next morning” turns the miss itself into a cue. More on that in if-then plans.
4. Separate planned breaks from misses. Illness, travel or a rest day you chose aren’t failures, so mark them as what they are, ahead of time if you can.
5. Judge the habit by the longer record. 19 days out of 21 is 90%, which a calendar or completion rate shows and a reset counter hides. Whether tracking helps at all is covered in does habit tracking work.
6. Save the fresh start for a real gap. Dai, Milkman and Riis found that people take on goals more often after temporal landmarks: at one university gym, the probability of a visit was 33.4% higher at the start of a new week (Dai et al., 2014). That helps after a two-week lapse. After one missed day, waiting for Monday only adds missed days.

What the research doesn’t tell us

• No study we found tests grace days in a habit app directly. The closest evidence is emergency skips for step goals and streak repair in a game, both over short periods.
• Silverman and Barasch’s field study is correlational, and their experiments used online participants making small, low-stakes choices: one more exercise, one more round of a game.
• Sharif and Shu’s field study involved 273 students and staff at one university, who tracked steps for four weeks and logged them in a shared spreadsheet.
• Lally’s habit curves rest on self-reported automaticity, with a good model fit for 39 people.
• In the gym megastudy, only 8% of programmes produced a significant change that was still measurable after the four weeks ended.
• The abstinence violation research comes from addiction, where the stakes are very different, and even there the evidence is mixed.
• Slack can be overused. In Sharif and Shu’s design the 7-day target stayed the goal and the skips were for emergencies. Our reading is that a buffer you plan to spend every week is simply a lower goal.