By the second week of February the queue for the treadmills has shortened, and someone at work repeats what everyone has heard: 80 percent of New Year’s resolutions have failed by now. The figure returns every winter, yet it seems to rest on no study at all. Its best-known source is a 2015 column on the U.S. News & World Report website, which stated the figure without naming any research behind it.¹

New Year’s resolutions mostly fade gradually rather than collapse in February, and they work better than their reputation suggests. In the few studies that followed people for months, resolvers did better than people who meant to change later, resolutions to start something outlasted resolutions to stop, and slips were common even among people who succeeded.

The practical version fits in four moves: choose one resolution you expect to enjoy, word it as something to do instead of something to stop, name one person who will check in with you each month, and settle ahead of time how you will restart after the first missed day. The evidence for each follows. For the general method behind any goal, from if-then plans to tracking progress, see the five steps for setting goals you’ll actually achieve.

Where the “80 percent fail by February” figure comes from

The 80 percent figure appears in a December 2015 U.S. News & World Report column by Joseph Luciani, which said that by the second week of February some 80 percent of people who made resolutions would have given up. The column named no study, survey or dataset for the number.¹ We found no earlier research source for it.

A round number that matches what people see in February is easy to repeat, and each repetition makes it look better sourced. The research that does exist is older and smaller, but it at least counted people. In a 1988 study, John Norcross and Dominic Vangarelli of the University of Scranton followed 200 resolvers for two years: about three-quarters were still keeping their pledge after one week, and about one in five after two years.²

Myth

80 percent of New Year's resolutions fail by February.

Fact

That figure traces to a 2015 magazine column that cited no study. In a 1988 follow-up of 200 resolvers, a little over half were still keeping their pledge after one month.

The drop was real, but it was a slope, not a cliff. As Martin Oscarsson and colleagues summarize that 1988 data in their own 2020 paper, a little over half were still keeping their resolution at one month, and the share kept falling slowly through the following months.³

1. 1 week77% still keeping
2. 1 month55%
3. 6 months40%
4. 2 years19%

These are small samples from one university decades ago, and success was what people reported, so no single number here is the true failure rate. The practical point is about timing: a wobble in late January puts you in the ordinary middle of the curve, not at the end of it.

1. The story people quote: most resolutions collapse by February; the figure traces to a 2015 magazine column that named no study
2. What follow-up studies found: a gradual decline over months, with slips along the way, including among people who kept their resolution

Do New Year’s resolutions work better than waiting?

People who made a New Year’s resolution did far better than people who meant to change later, in a 2002 University of Scranton study by John Norcross, Marci Mrykalo and Matthew Blagys that followed both groups for six months by telephone. At six months, 46 percent of the resolvers reported continuous success, compared with 4 percent of similar people who planned to tackle the same kinds of problems later.⁴

The two groups had similar backgrounds, problem histories and goals, mostly weight loss, exercise and stopping smoking. What set the successful resolvers apart showed up before the year even began: their confidence that they could changeself-efficacy: A person's belief in their ability to carry out a particular task or reach a particular goal. It is specific to the task, so someone can feel capable at work and unsure with money, and it goes with how readily people start and keep going.Full entry in the glossary, their skills for changing and their readiness, all measured before January 1, predicted who would succeed. Once January came, the successful resolvers used more cognitive and behavioral techniques, and fewer that raise awareness or stir up emotion about the problem.⁴

The comparison was not randomized. People chose whether to resolve, so the resolvers may simply have been readier, and all outcomes were self-reported. The study shows that resolving and succeeding went together, not that a date on the calendar made the difference.

Picture two colleagues who both want to start running. One picks January 2, buys shoes in December and books three morning slots; the other plans to start “once things calm down.” The lesson from the Scranton data is not that January has special power. It is that the people who succeeded arrived at the start date already confident and equipped, so build those in the weeks before, not on the day.

Why resolutions fade: false hopes, stress and the first slip

Resolutions tend to fade when expectations are unrealistic and when a stressful stretch knocks the new behavior off course. In a 2002 review, Janet Polivy and Peter Herman describe a “false hope syndrome”: people expect self-change to be faster, easier and larger than it turns out to be, fall short, and then try again.⁵

The 1988 Scranton study shows what the stressful stretch looks like. Slips were usually set off by a loss of personal control, excessive stress or negative emotion, and slipping was common even among the successful: 53 percent of those who kept their resolution over two years had slipped at least once. The successful resolvers also reported relying more on controlling their surroundings, rewarding themselves and willpower than the unsuccessful did.²

So a slip, on its own, tells you little about how the year will end. Take a resolution to walk every lunchtime. A deadline week arrives, you skip four days, and the false-hope reading says the resolution is dead. The 1988 data say slips like this were common, including among the people who kept going.

Controlling your surroundings is the tactic with the clearest practical form. If the lunchtime walk dies whenever meetings run over, book the walk in your calendar like a meeting and leave your shoes by the door, so the setting prompts you when your willpower is spent.

Further reading

• 

  How to Change: The Science of Getting from Where You Are to Where You Want to Be

  by Katy Milkman

  Katy Milkman, co-author of the fresh start research cited here, on timing a change well and keeping it going after a lapse.

• 

  Get It Done: Surprising Lessons from the Science of Motivation

  by Ayelet Fishbach

  Ayelet Fishbach, a co-author of the 2025 resolutions study, on why enjoying the pursuit of a goal helps you stick with it.

As an Amazon Associate WiserHours earns from qualifying purchases.

Resolutions to do something outlast resolutions to stop

In a 2020 Swedish study that its authors describe as probably the largest on New Year’s resolutions so far, people who framed their resolution as something to do or approach were more often still keeping it after a year than people who resolved to stop or avoid something.³

The study

Limited evidence

Sweden, 2017: 1,066 resolvers followed for a year

At one year, 55% of those who responded considered themselves successful. Resolutions phrased as something to approach were kept by 58.9%, against 47.1% for resolutions to avoid something. The group randomly given some support (information on social support, one named supporter, monthly check-ins and one email of exercises) reported somewhat more success than the group given none, and also more than a group given extensive goal-setting instruction.³

Only the support comparison was randomized. People picked their own wording, so those who wrote “start” resolutions may have differed in other ways too, and one group had been coached to write approach goals. Success was self-rated, the volunteers were probably more motivated than most, and three of the authors receive royalties from a book on resolutions based partly on this study.³ Treat the wording advice as a cheap bet with some evidence behind it, not a proven lever.

In practice, turn a “stop” into a “start”. “Stop snacking at my desk” becomes “eat fruit at three o’clock”; “stop checking email at night” becomes “close the laptop at seven and read”. For a habit you want to drop, the case for naming a replacement behavior instead of a bare refusal has its own evidence.

More instruction did not mean more success. The group that also learned SMART goal-setting and set six interim goals did no better than the group with no support. The authors offer one possible reason: specific goals and interim deadlines make every shortfall visible, so a resolution with six deadlines has six chances to feel like failure.³ The package that helped included naming one supporter, and the study cannot say which part of it did the work.

Enjoyment matters too: in a year-long 2025 study of US adults by Kaitlin Woolley, Laura Giurge and Ayelet Fishbach, finding the pursuit enjoyable predicted sticking with a resolution better than finding it useful.⁶

Why January feels like a clean slate

January 1 is one of the biggest “fresh start” moments, and far from the only one. In a 2014 paper, Hengchen Dai, Katherine Milkman and Jason Riis found that US Google searches for “diet” rose at the start of each year, and that students at a university gym were about a third more likely to visit at the start of a week than on ordinary days.⁷

The authors’ proposed explanation is that a landmark such as a new year, month, week or birthday opens a new “mental accounting period”: past failures feel as if they belong to the old one, and people take a big-picture view of their lives, which nudges them toward their aspirations. Their evidence was records of behavior, not experiments, and it captured starts rather than staying power. By the authors’ reading of their own plots, the lift after a holiday faded quickly, while the lift at the start of a week, month or year wore off more gradually.⁷

In one later test, timing and framing alone did not help. In a 2017 randomized trial with some of the same researchers, funded by the US health insurer Humana, reminder letters sent to its members around their birthdays or after New Year’s Day and framed as a fresh start did no better than ordinary reminders; the authors note many New Year letters arrived in late January.⁸

So January is a good starting line but not the only one, and the extra motivation is a window, not a fuel supply. Someone whose plan to learn Spanish lapsed in February does not need to wait eleven months: the first Monday of March is a landmark too. Because the lift fades, set up the first week before the date arrives, so the burst of motivation lands on a plan that is already in place.

Spend part of December looking back

The weeks before January are where much of a resolution’s outcome seems to be set: in the 2002 Scranton study, confidence, skills and readiness measured before January 1 predicted who was still succeeding months later.⁴ A year-end review has not, as far as our searches found, been tested directly, so this is an inference from what predicted success, not a tested method.

A useful review looks for the weeks that went wrong last year. The 1988 study found that slips usually followed a loss of personal control, excessive stress or negative emotion.² If your resolution died during last March’s reorganization, this year’s plan needs an answer for the next reorganization, and a supporter who knows it is coming.

Then choose one resolution you expect to enjoy, phrase it as an action, and decide who will check in with you. The checklist below gathers the steps this evidence supports.

WiserHours

A resolution with better odds

• One resolution, written as something to do
• A version of it you expect to enjoy
• A start date on a landmark: New Year, a Monday or the first of a month
• One named person who checks in monthly
• The likeliest bad week named, with a plan for it
• A restart plan for the day after a slip

Find this checklist online:
https://wiserhours.com/goals/new-years-resolutions/#checklist-new-years-resolutions-plan

What the evidence on resolutions supports, claim by claim

The evidence on resolutions is thin: a handful of follow-up studies, mostly with self-rated success, and one large randomized comparison of support. The table sorts the main claims by the strength of what stands behind them.

The rows point to where your effort goes furthest. The cheapest moves, a supporter, a “do” wording and a restart plan, rest on real follow-up data, while the most repeated claim, the February cliff, rests on none.

The bottom line

Resolutions fade slowly rather than fall off a February cliff, and much of what predicted success was in place before the year began. Pick one resolution you would enjoy, phrase it as something to do, name one person to check in with you, and plan the restart for the day after your first slip. If January passes you by, the next Monday will do.