Few habits are condemned as confidently as the snooze button. Sleep experts say it fragments your rest. Productivity writers say it starts your day with a failure. The advice is always the same: set one alarm, get up when it rings.
Here is the awkward part. When researchers finally brought habitual snoozers into a sleep laboratory, wired them up, and compared snoozing mornings with get-up-at-once mornings, the snoozers did fine. On some tests they did slightly better. The case against snoozing turns out to rest almost entirely on reasoning about what ought to happen, and hardly at all on measuring what does.
Almost everyone does it
If snoozing is a vice, it is a majority one. In 2025, researchers at Mass General Brigham analysed data from the Sleep Cycle app: 3,017,276 sleep sessions from 21,222 users around the world. Around 55.6 percent of sleep sessions concluded with snooze alarms, with the average sleeper spending 10.8 minutes snoozing after their initial alarm.
Some people lean on it harder. Around 45% of study subjects hit the snooze button on more than 80% of mornings, and these heavy users snoozed, on average, 20 minutes a day. Snoozing is largely a response to an alarm set by an employer’s schedule rather than by the body’s clock.
An earlier Swedish survey of 1,732 adults found the same thing from a different angle. There, 69% reported using snooze or multiple alarms at least sometimes, with an average of 22 minutes per morning. Snoozers tended to be younger and were more likely to be evening types.
The theory of why it should be bad
The argument against snoozing is coherent. The final hours of the night are rich in REM sleep. An alarm drags you out of it, you doze for nine minutes, and another alarm drags you out again. Sleep researchers argue that repeated alarms interrupt these critical stages and leave only light sleep in between.
The recommendation that follows is to set your alarm for the latest workable time, then get out of bed when it first rings.
It sounds right. But notice what the app study measured. It counted how often people snooze. It did not measure whether snoozers were more tired, performed worse or had poorer health. Sleep Cycle’s own press release says so: the study does not yet determine whether the behavior is helpful or harmful to health. The warning attached to the study came from the theory, not from the study.
What happened when someone tested it
The test came from Stockholm University, published in the Journal of Sleep Research under the title “Is snoozing losing?”.
The researchers recruited 31 habitual snoozers and had them sleep in a laboratory with full polysomnography (the brain, eye and muscle recording used to stage sleep). On one morning they could snooze for 30 minutes. On another they had to get up at the first alarm. Each person did both, so everyone served as their own comparison.
The results, from the paper’s summary: 30 min of snoozing improved or did not affect performance on cognitive tests directly upon rising compared to an abrupt awakening. The total cost in sleep was small. Snoozing resulted in about 6 min of lost sleep, while preventing awakenings from slow-wave sleep.
That last clause is the interesting one. Waking from deep slow-wave sleep is what produces the worst grogginess, the state sleep scientists call sleep inertia. A few rounds of light dozing appear to lift people out of deep sleep gradually, so the final awakening comes from a lighter stage.
Everything else came up blank. There were no clear effects of snoozing on the cortisol awakening response, morning sleepiness, mood, or overnight sleep architecture. The cognitive edge was brief too: the advantage on one test had gone within 40 minutes.
The lead author, Tina Sundelin, said she was surprised. Her summary: snoozing is common and probably not that bad for you, as long as you are getting the sleep you need beforehand.
The caveats, which are substantial
This is one small study, and its author says so: it is a rather small study and we only focus on one morning of snoozing.
Three limits matter most.
It only tested snoozers. All 31 were people who snooze regularly and fall back asleep easily. The results may not apply to people who do not usually snooze. If you lie awake anxious between alarms, you are not getting the gentle transition these participants got.
It tested one dose. Thirty minutes, with alarms at fixed intervals. Whether 10 minutes or 60 minutes behaves the same is unknown.
It tested one morning. Nobody has followed snoozers and non-snoozers for months to see whether anything accumulates.
So “snoozing is harmless” would be overclaiming. The accurate statement is narrower: the only controlled test found no harm and a small short-lived benefit, in people who already snooze.
What snoozing is a symptom of
The survey data point to the more useful question. Snoozers in the Swedish study slept slightly less and felt more drowsy in the morning than people who never snooze, and they were disproportionately evening types.
That suggests the snooze button is not the problem. It is what a late body clock does when it is made to get up early. An evening type woken at 6:30 is being pulled out of the middle of their biological night, and snoozing is a way of softening the landing. Taking the button away does not fix the mismatch.
The app data fit the same picture: snoozing tracks work-driven wake times more than the body’s own rhythm.
What to do about it
If you snooze, fall back asleep easily and feel fine by mid-morning, there is no evidence that you need to stop. The guilt is unsupported.
If you snooze and still feel awful, look upstream. The likelier culprits are too little sleep, irregular bedtimes, or a wake time that fights your body clock. Those have far better evidence behind them than anything about the alarm.
If you lie awake between alarms, snoozing is giving you nothing. You are losing sleep at the front end to buy wakefulness at the back. Set the alarm for the last possible minute and take the unbroken sleep instead.
And if you can shift your start time later, even by half an hour, that is worth more than any alarm strategy.
Common questions
Is it bad to hit the snooze button?
The only laboratory test, in 31 habitual snoozers, found that 30 minutes of snoozing improved or did not affect cognitive performance on waking and cost about 6 minutes of sleep. No effects were found on cortisol, mood, sleepiness or overnight sleep structure.
Does snoozing make you more tired?
Not in the controlled study. Snoozers in surveys do report more morning drowsiness, but they also sleep slightly less and are more often evening types, so the tiredness appears to come before the snoozing and not from it.
How many people use the snooze button?
In an analysis of 3,017,276 sleep sessions from 21,222 app users, about 55.6 percent of sessions ended with a snooze alarm, averaging 10.8 minutes. Around 45% of users snoozed on more than 80% of mornings.
Why do sleep experts say not to snooze?
Because the last hours of sleep are rich in REM, and repeated alarms interrupt it and offer only light sleep in between. That reasoning is plausible, but the large app study that accompanied the advice measured how often people snooze, not whether it harms them.
The bottom line
The snooze button has been convicted on theory. The one experiment that actually measured it found about six minutes of lost sleep, no change in stress hormones or mood, and marginally sharper thinking in the first minutes of the day. It is a small study in committed snoozers and proves nothing about the long term. But if your mornings work, keep them. If they do not, the fix is more sleep at a more regular time, not a sterner relationship with your alarm.
Did this answer your question?
Thanks. That helps us decide what to fix.
Leave a comment