A case for the smart alarm

Posted on Posted in Uncategorized

A case for the smart alarm

On average, people sleep around 7 hours a night. But according to our research, it differs a lot with the location. For example, people around the equator tend to sleep less – up to 30 minutes less than the average American!


  • Using smart alarm has a positive effect on the morning mood
  • About 30% of alarms get snoozed for longer than 30 minutes
  • Setting up smart alarm prevents oversleeping


We have also found that the longer people sleep, the happier morning they have.

However, you probably don’t want to sleep your life away. Most of us need to work, want to spend some time with children, with their hobbies and with all the nice things in life. How can you make your sleep better without waking up in the afternoon? Use smart alarm. We have hard evidence that it works, Watson.

First argument for the smart alarm: people woken up by smart alarm are rating their sleep better. For smart period = 30 minutes, there is a small tendency to rate the sleep better if the smart alarm did wake you up in the light sleep. If it woke you up at the end of the period (so you were in deep sleep at the time), the rating tends to be lower. We can say that using a smart alarm has a positive effect on the morning mood. See for yourself:


Looking at the chart, 0.03 difference in the rating might not seem like a lot. But the data behind it are very consistent, and with a standard error of the mean less than 0.01, we can say we have a very good idea of the underlying reality.

Most people are using the smart period of 30 minutes – it is the default in Sleep as Android, and it makes most sense. Half an hour almost guarantees that a part of the smart period will take place in light sleep, thus enabling the smart alarm to gently wake you up.

Second argument for the smart alarm: It turns out that a whole lot of people snooze their mornings away.


This chart is very revealing, since it shows that people are snoozing a lot. 2% of mornings are getting a solid hour surplus, and lot of mornings even more, up to almost 150 minutes. If we count all snoozes longer than 30 minutes, we get around 30% of mornings!

Let’s break that down a little bit in the following table.

Average snooze duration Total sleep duration Average oversleep
(Time from first alarm ringing to dismiss)
No alarm X 7 hours 39 minutes X
Fixed alarm time 9 min 16 sec 7 hours 43 minutes 9 min 16 sec
Smart alarm 24 min 34 seconds 7 hours 15 minutes 15 min 52 seconds

It’s interesting that people do sleep longest on average when they have set up a fixed alarm.
With a fixed alarm time, there is a tendency to wake up earlier than the alarm would sound, indicating that many people are capable of waking up smart by themselves. But on the other hand, the average snooze time is about 10 minutes. So people who do not wake up by themselves will try to stay in the bed a little longer. This might have a lot of causes. One of the main causes will probably be that people just set their alarms for a little earlier than they really need to.

With smart alarm, people generally sleep less (no wonder!), they snooze much more, and rate their sleep slightly better than with fixed alarm time. If we may interpret a little bit, a possible cause for the snoozing rise is that they just have more time. Even then, on average they get 5 minutes before the designated alarm deadline. Which means that with respect to the fixed alarm time, they have on average 15 minutes out-of-bed time more.

Other articles in SleepCloud study series<< Snoring: an apology for the pathologyHow do we sleep post-brexit? >>

Leave a Reply

Your email address will not be published. Required fields are marked *