Blog · · 9 min read
Do most New Year's resolutions fail? Less often than you've heard
Fewer than you have probably heard. The widely quoted claim that 92% of New Year’s resolutions fail does not come from any study anyone can find. When researchers followed people who made resolutions, roughly four in ten were still keeping them at six months, far more than among people who wanted the same change without making one. In a year-long Swedish trial, more than half were still on track.
- The “92% fail” figure is a misquote: the university study it is usually credited to found 46% still succeeding at six months.
- Making a resolution beats waiting: 46% of resolvers were still on track at six months, against 4% of people who wanted the same change but did not make one.
- Starting beats stopping: in a trial of 1,066 people, 58.9% kept a resolution to start or do more of something at one year, against 47.1% for resolutions to quit or avoid something.
- Almost half of resolvers stop by the end of January, but a slip is not the same as stopping: more than half of the people who succeeded at two years had slipped at least once along the way.
Where “92% fail” and “Quitter’s Day” come from
Every January, articles warn that 92% of New Year’s resolutions fail, or that only 8% succeed. The figure is almost always credited to a study from the University of Scranton published in the Journal of Clinical Psychology.
There is a University of Scranton study of New Year’s resolutions in that journal, led by the psychologist John Norcross in 2002.1 It does not say 8% succeed. It says 46% of participants were still keeping their resolution at six months, and its authors wrote the paper specifically to challenge the belief that resolutions never work.
How did 46% become 8%? Around 2012, a statistics-aggregation website combined two unrelated sources on a single page: the University of Scranton’s six-month numbers, and an informal online survey by a management consultant who had written that 8% of respondents were always successful with their resolutions every year. The word “always” was dropped in the table, the University of Scranton was listed as the source at the foot of the page, and a 2013 magazine article put the two together. Within a few years, “92% fail, according to the University of Scranton” had been repeated thousands of times.
“Quitter’s Day”, usually placed on the second Friday or third week of January, is newer. The fitness app Strava coined the term in a 2019 press campaign after looking at more than 800 million logged workouts and noting the date in January when activity uploads began to dip. It is a real pattern in app uploads, not a study of how many people keep their resolutions.
What researchers actually measured
Surprisingly few studies have followed people after January 1 to see what happens. The best-known come from Norcross and his colleagues at the University of Scranton.
In the longest-running study, published in 1988, the researchers recruited 200 adults who had made a New Year’s resolution and called them over the next two years to ask how they were doing.2 Losing weight, quitting smoking and improving a relationship were the most frequent goals. One week into the year, 77% were still keeping their pledge. That share fell to 55% at one month, 43% at three months, 40% at six months, and 19% after two years.
A fair question is what would have happened without a resolution. The 2002 study tested that by following two groups for six months: 159 people who made a New Year’s resolution, and 123 people of similar age and background who wanted the same kind of change (losing weight, exercising more or quitting smoking) but had not made a resolution for January 1.1
Among the resolvers, 71% were continuously successful at one week, 64% at two weeks, 58% at one month, and 46% at six months. Among the people who wanted the same change without making a resolution, the share at six months was 4%.
That is not a randomised trial. People chose for themselves whether to make a resolution, so the resolvers were already readier to act, and their readiness before January 1 predicted who succeeded. Even so, wanting to change “sometime soon” rarely turned into action on its own.
Starting something beats stopping something
The largest experiment on New Year’s resolutions was published in 2020 by Martin Oscarsson and colleagues at Stockholm University and Linköping University.3 They recruited 1,066 people across Sweden in the last days of December and followed them for a full year. The most common resolutions were about physical health (33%), weight loss (20%) and eating habits (13%).
At the end of the year, participants rated how well they were sticking to their resolution from 0% to 100%, with 70% or above (“by and large sticking to my resolution”) counted as successful. Overall, 55% of responders counted themselves successful at one year.
Two findings stood out. First, how the resolution was phrased mattered. Psychologists distinguish approach goals from avoidance goals. An approach goal is about starting something or doing more of it, such as walking after lunch or eating a piece of fruit each afternoon; an avoidance goal is about stopping or cutting something out. People with approach goals succeeded significantly more often at one year: 58.9% against 47.1%.3
Second, the researchers randomised all 1,066 people into three levels of support:
- No extra support: wrote down their resolution and answered three check-ins across the year (55.9% average success rate).
- Some support: named one friend or family member to support them, answered monthly check-ins, and received one email on handling setbacks (62.3%, the highest of the three).
- Extended support: all of that, plus instructions to set strict, measurable deadlines, six interim goals across the year, and four coaching emails (52.5%).
Why did the group with the most instructions do worst? The authors noted a straightforward possibility: a strict deadline makes anything short of 100% look like failure, and six interim deadlines give you six extra dates to miss.3
Why January works at all
There is nothing magical about January 1, yet people use it every year. Hengchen Dai, Katy Milkman and Jason Riis studied why in 2014, calling it the “fresh-start effect”.4
Across five years of US Google searches, university gym attendance records, and commitments on a goal-tracking website, people were measurably more likely to act right after a calendar landmark. Google searches for “diet” rose by 82% in the first week of the year compared with an average week, by 14% at the start of a new week, and by 4% at the start of a new month. Gym visits jumped by 33% on Mondays, by 47% at the start of a new semester, and after federal holidays and birthdays.4
A landmark in the calendar draws a line between your past self (“I let things slide in December”) and your present self. So January is a good moment to start, and if it goes badly, the first Monday of February or the first of the next month offers the same kind of fresh start on a smaller scale.
When resolutions slip
Look back at the two-year chart above and the steepest drop is right at the start: nearly a quarter of resolvers had stopped within the first week, and almost half by the end of January.2 That is the part of the curve where a new behaviour is not yet automatic, and where treating a missed day as the end of the attempt does the most harm; both are covered in How long does it take to build a habit?
What the two-year study adds is what happened to the people who succeeded. Among those who were still keeping their resolution after two years, 53% had slipped at least once, and across the two years the average was 14 slips.2 Most people who kept their resolution did not keep an unbroken streak for two years. They slipped, most often during stress or a bad week, and started again.
What helps
- Phrase it as something to start where you can. “Drink sparkling water with dinner” or “walk for 20 minutes at lunch” gives you something concrete to do today, and approach goals held up better over a year than goals framed only as quitting.3
- Tie it to a daily cue, and write the bad-day version now. Decide when and where the action happens, and what counts on a tired Tuesday; there are examples of both in the habit post.
- Plan the day after a miss. Since most long-term keepers slip along the way, the step that matters is returning the next day. In the gym megastudy covered in Do rewards build habits?, rewarding the return after a missed workout beat 52 other ideas.
- Name one person, or do it alongside others, without piling on deadlines. In the Swedish trial, naming a support person and checking in once a month beat both doing it alone and setting six strict interim deadlines;3 in the two-year study, social support predicted who was still on track after six months.2
If you want a ready-made routine to start on Monday 4 January alongside other people, Allez! is running four January challenges from that day: Alcohol-Free January, Road to 5K, January without added sugar and Early mornings in January.
What this research cannot tell you
- Success is self-reported. In all of these studies, participants graded their own adherence. When the 2002 study also looked at weight and cigarette counts, the average changes were modest (about 3.6 kg lost, or cigarettes cut from 23 to 15 a day), suggesting people judge themselves generously.1
- Volunteers are not everyone. People who sign up for a university study on resolutions, knowing a researcher will phone or email to ask how they did, are probably more committed than the average person making a promise at midnight. The check-ins themselves also act as a reminder.1, 3
- There are very few studies. Despite millions of people making resolutions every year, almost everything we know comes from a handful of American and Swedish cohorts.
Common questions
Is January a bad time to start?
No. Dark, cold weather is real in the northern hemisphere, and there is more on handling that in Five daily habits for the winter blues. But in the 2002 study, people who used January 1 were more than ten times as likely to be on track six months later as people who wanted the same change and waited for a better time.1
How many resolutions should I make?
No experiment has directly randomised people to one resolution versus three. In the Swedish study, people were allowed to list several, and their first was tracked; the group given six interim goals across the year rated themselves less successful than the group with a simpler setup.3 Starting with one routine where the daily actions fit together is the safer bet.
Is it too late if I have already slipped in January?
Not at all. More than half of the people who kept their resolution for two years slipped along the way, 14 times on average.2 And the fresh-start effect does not belong only to January 1: Mondays and the first of any month produce the same kind of reset on a smaller scale.4
The studies behind this
- Norcross, Mrykalo & Blagys (2002). Auld lang syne: Success predictors, change processes, and self-reported outcomes of New Year’s resolvers and nonresolvers. Journal of Clinical Psychology.
- Norcross & Vangarelli (1988). The resolution solution: Longitudinal examination of New Year’s change attempts. Journal of Substance Abuse.
- Oscarsson, Carlbring, Andersson & Rozental (2020). A large-scale experiment on New Year’s resolutions: Approach-oriented goals are more successful than avoidance-oriented goals. PLOS ONE.
- Dai, Milkman & Riis (2014). The fresh start effect: Temporal landmarks motivate aspirational behavior. Management Science.