How to better measure your sleep with Apple Watch.

[et_pb_section fb_built=”1″ _builder_version=”3.22.3″ custom_padding=”0px|||||”][et_pb_row _builder_version=”3.22.3″ background_size=”initial” background_position=”top_left” background_repeat=”repeat” custom_padding=”0px|||||”][et_pb_column type=”4_4″ _builder_version=”3.0.47″][et_pb_post_title _builder_version=”3.22.7″ title_font=”||||||||” title_text_color=”#000000″][/et_pb_post_title][et_pb_text _builder_version=”3.22.7″ text_text_color=”#000000″]

It seems everyone with an Apple Watch is running mini experiments to monitor their sleep quality but is the technology accurate? Is the methodology accurate?

Follow my experiment

Over the next 30 days I will be conducting an experiment on myself measuring the impact of the Can’t Sleep App on my sleep. As a 3rd year PhD student with a research topic of music & sleep, I am already aware of healthy sleep hygiene and am a good sleeper. The question is, can music improve my sleep even further? Music has shown to be effective for poor sleepers but there has not been a lot of research on normal sleepers. I hope to educate other individuals who are interested in bio-hacking to run more rigorous experiments using the technology that is available to the public.

Some of the most important elements of research are knowing the purpose of the study and the limitations. Whilst every research paper state both areas, unfortunately they are not often represented in citations or when presented to the public. To increase the accuracy of any experiment you must be aware of these.


  1. Ask your GP before performing any experiment.
  2. If you are a researcher and would like something added to this blog post please email me at

Purpose of the experiment

This experiment will measure the effect of the Can’t Sleep App on sleep for 30 days alternating music (intervention) and silence (control). Sleep will be measured using a series 3 apple watch and the sleep app “Pillow” for Deep sleep, Time to fall asleep, REM sleep, Number of Awakening and Total time sleeping. I hypothesize that music will improve each of these measures and continue to improve each measure with greater exposure. At the end of this experiment I will post the results, analysis and any findings.

Limitations and variables to be aware of

Study design limitations

A single participant experiment means minor influences (such as going to bed late one night) can greatly impact on the results. I am attempting to minimize this impact by conducting the experiment for 30 days which means more points of measurement.

I am the participant and researcher which means personal bias could impact the implementation, measurements and analysis. Being the founder of the music sleep app and studying music & sleep will unintentionally lead to bias. My experience in building the app and working with my developer will provide a different experience of the music compared to others. This means the experiment could go either way.

Technology limitations

From the research I did in the Appstore, “Pillow” provided the most variables. Whilst the gold standard for measuring sleep quality is polysomnography (think brain activity, eye movement, heartrate) in combination with participant self-report measures, these aren’t financially viable and require professionals. I do not believe that Apple Watch and Pillow are viable for accurate sleep tests but may present some indication of sleep quality. If you are interested in finding out why? Let me know at, otherwise an easy way to check accuracy of a product is to search the product in “google scholar” and read any articles. If no articles show up then it’s not accurate.


There are a lot of variables that impact on sleep, but I am going to limit them to keep the experiment more manageable. These include

  • Going to sleep before or after normal bedtime
  • Caffeine consumption
  • Headaches
  • My noisy neighbour
  • Blue light prior to sleep & screen time
  • Issues with the technology including Pillow App, iPhone or Apple Watch
  • Alcohol

Other variables might include exercise, stress/anxiety, your pet waking you up, jet lag, shift work, medication/drugs, sleeping in an unfamiliar bedroom, light shining in your bedroom, temperature, diet, having a late meal, getting sick/flu and the list goes on…

Is it even worth running experiments on yourself when it can be so inaccurate?

Adding up all the study design limitations, technology limitations and potential variables you might be thinking to yourself “I’ll just leave it to professional researchers and follow their advice”. This isn’t necessarily a bad idea, but I believe a lot can be gained by measuring yourself. The good news is that technology is getting more affordable, more available and more accurate. The great news is that running these experiments creates an awareness of factors that might influence your sleep, and this can shape your behaviour in positive ways. For me personally, as a research I’m really looking forward to be a test subject for once, to truly experience the other side and I hope this will give insight into my future studies.