What is this post about?
Over the summer of 2020, I slept pretty badly for a long period of time and it had pretty noticeable effects on my productivity, perceived energy levels, and well-being in general. This was mostly due to the fact that my room was under the southern roof of the house and it had around 28 degrees Celcius during “cool” summer nights which made it effectively impossible for me to get more than 6 hours of light sleep per night. In autumn, I moved to a new flat where I had to get a new bed, mattress, pillow, etc.
Since I had felt the negative consequences of bad sleep pretty recently, I figured I could conduct a bit of research and use science to improve my own sleep. In the following, you can find a summary of what I found out about the right equipment and behavior to improve sleep mixed with my own considerations.
One piece of framing that I want to provide before diving into the main part is about the money and effort one should be willing to spend to improve our sleep. The longer I thought about this question the more I thought the right answer should be: “a lot”. Just think of the effect that good vs. bad sleep has on your life and quantify it roughly. Let’s assume the difference between good vs. bad sleep results in 30 more productive minutes per day. If you value your time at the level of the minimum wage (as a conservative estimate) which is around 10€ in Germany this means you should be willing to pay up to 5€ per day. Over a year that makes 5 * 365 = 1825€. If you additionally consider all the negative side effects of bad sleep this number gets larger very quickly. If you have, for example, a mattress that is way too soft for you, your spine might suffer which can have large consequences for the rest of your life, e.g. while sitting, driving a car, or during exercise. Treating these outcomes, which are often only revealed after years and take a long time to fix, is often very expensive and emotionally draining or painful. A second example would be our perceived well-being throughout the day. I think we all know that there is a noticeable difference between a day after a good night and a bad one. One day feels slow, annoying, and unproductive while the other feels faster, more energetic, and more productive. While this is harder to quantify financially it is still something that I would be willing to put a lot of effort in to improve consistently.
July 2021: added section on eightsleep; emphasized positive personal experiences with exercise.
The right bed
There are a lot of guides out there telling you how to choose the right mattress, pillow, blanket, bed frame, etc. to optimize your sleep. In this chapter, I want to give a broad overview of the different options and how confident we should be in the publicly available advice.
Unfortunately, though, I have to start this section with bad news. There is only very little solid and reliable research on the topic of mattresses. This Vice article gives a pretty concise summary of the state of research and the problems it has. Basically, it says that a) research about sleep is always very expensive since you have to distribute a lot of mattresses to a lot of people, b) it is therefore often funded by big mattress companies who have their own interests and the results should be approached with caution and c) private research or privately sponsored research often yields rather imprecise results such as “on average people preferred the medium-firm mattress to the firm one”. But then most mattress companies have their own rating system and there is no unified measurement so results from one company won’t necessarily translate into another. The one trustworthy study that the authors found concluded that no mattress or mattress type is clearly superior to any other in general but for every individual person there are drastically better or worse mattresses. This finding mixed with my personal experience leads me to believe that it makes sense to research mattresses to improve my quality of sleep.
Keeping in mind that they are not research-backed and should rather be interpreted as experience-based heuristics at best, we can look at the recommendations made by people who sell mattresses for a living. This guide (and this) gave me a good overview of the parameters that you can choose and should be aware of when buying a new one. The article’s main points are a) the parameters that matter most are your sleeping position, your weight, and your preferred firmness. Depending on them you should get a soft or firmer mattress. b) Mattresses are made of different materials. There are memory foam, latex, innerspring, and hybrid models. Each of which has its advantages and disadvantages with respect to sleep feeling and living age of the mattress. c) You shouldn’t buy mattresses online. Go to a local store and test out different models. Really lie down for five minutes or longer per mattress in contention. If you don’t know exactly what the parameters of your mattress are (which most people don’t) you have to try it out (which is harder to do online). d) Quality comes at a price. It might not be a perfect predictor and you will find some outliers but usually, a more expensive mattress will give you a better sleeping experience. As with most products, there are diminishing returns but given that this choice will have a non-negligible impact on the average quality of sleep over your next 8-10 years you should not aim too low.
There are also different fancy ergonomic pillows and blankets. In both cases, you will find wild claims about how the right choice of either can drastically improve your sleep and decrease negative conditions such as neck pain. However, once again, the research seems to be based to a large extent on privately sponsored or privately conducted research which we should treat with caution.
As stated above I have bought a new mattress following all the guidelines. I think my quality of sleep and my perceived energy levels have increased since and it was worth the price already. But as I have moved to a new flat there are also many other parameters that have changed and it is hard to isolate the impact of the mattress compared to all others. I have also bought a fancy pillow that promised to be especially ergonomic for belly sleepers. Honestly, I mostly bought it out of curiosity and because I low-key wanted to show that these are mostly empty promises without tangible effect. But I was wrong. I really like the new pillow. It fits exactly to my sleeping position and my neck is very relaxed every morning. If you have neck problems after sleeping I would recommend testing out a couple of different pillows, they are astonishingly cheap (20-50 bucks) for the effect that they can have.
From my perspective, the most sensible view of choosing the right mattress, pillow, etc. is that they will yield small improvements for most people but prevent large harms for some. So if your sleep is already pretty OK you can expect a small but tangible improvement. If you have neck pain after waking up or troubled sleep in general, a change of equipment might have a very noticeable positive effect on your sleep and life in general.
A note on your Sleeping position: Apparently some people believe that one sleeping position is clearly superior to all others and therefore you should actively switch to that. While there are advantages and disadvantages to all positions, e.g. certain sleep-related conditions arise less when you sleep on your back, there is no clear best position. Additionally, it is very hard to train a different position from the one that feels natural to you. So if you don’t have a condition that can clearly be linked to your position (e.g. some instances of snoring) the guides recommend adapting your mattress, pillow, etc. to your position rather than changing your position.
Even though you can get measurable improvements through a change of equipment the largest improvements probably come only with behavioral changes and putting in some effort - buying a good bike certainly improves your speed but you still have to actually train to get large improvements. In this part of the post, I will talk about everything that can broadly be attributed to a change of surroundings and behavior on our quality of sleep.
Biology behind Sleeping
To understand a bit better why certain behaviors or environments are desirable for better sleep we first need to look at the biology of sleep. If we know how it works we can more easily improve it. Obviously, I will simplify and my neuroscience professor would not be very happy but I will try to give a concise summary of the main concepts.
The Circadian Rhythm describes the 24-hours wake/sleep cycle that most humans experience. It is controlled by multiple parts of the brain one of which is the pineal gland. When the eyes receive sunlight the pineal gland does not produce melatonin and you feel awake. If it doesn’t receive sunlight it does produce melatonin and you feel tired. However, this is clearly not the only mechanism controlling your circadian rhythm - otherwise, we would not experience jetlag. The Suprachiasmatic Nucleus (SCN) is a small subpart of the Hypothalamus that functions as the “master clock”. It is responsible for keeping up the 24-hour rhythm independent of sunlight. This rhythm can be adapted, e.g. when we switch timezones, or forcefully changed in length but it takes time and comes with a potentially high cost to the person’s mental state.
There are different stages and types of sleep. Stage one describes a transition stage where we start to relax. People awakened during this stage often deny that they have been sleeping at all. Stage two describes light sleep but people can still easily be woken up. Stage three is moderate deep sleep and usually starts 20 minutes after stage one. Your muscles are relaxed and vitals (body temperature, blood pressure, and heart and respiratory rates) fall. Stage four is the last stage and it describes very deep sleep. Your muscles are relaxed, vitals are at the lowest level and it is very hard to wake you up. This stage is only reached during the two earliest 90-minute cycles of your sleep. These 90-minute cycles are always between awake and a rapid eye movement (REM) phase or between two REM phases. During every cycle, you will go from a lighter to a deeper stage and come back to a lighter one. The REM phases increase in length over the duration of a night and they are thought to be linked to memory consolidation. The current scientific hypothesis is that you relive memories during the REM phase, strengthen important and weaken unimportant ones.
I will try to use this knowledge to motivate parts of the following heuristics but this is not always possible as our scientific understanding of sleep is rather limited. Measuring stuff in the brain is always hard but gets harder when people can’t communicate because they are asleep. Additionally, many experiments that might yield insights into sleep would involve a lot of sleep deprivation and are often not granted ethical approval.
In which circumstances we sleep can already have a noticeable effect on its quality. Here are some environment-related parameters that one could improve.
- Light: As described above, your pineal gland doesn’t produce melatonin when your eyes receive too much light. So if your room isn’t sufficiently dark you will have a harder time falling asleep. If your window is next to a street lamp and you have a bad curtain or blind that you can’t change because your landlord doesn’t allow it I would recommend trying a sleeping mask. It’s a bit weird to get used to but can help wonders once you do. If you have a sleeping rhythm that is compatible with daylight times I would also recommend curtains or blinds that allow in some light to be gradually woken up by sunlight in the morning. In my experience, this is a less disruptive and more pleasant way of waking up.
- Noise: Noise can disrupt your sleep and thereby decrease the time you spend in stage three or four. However, this is primarily the case due to abrupt noises, e.g. the squeaking breaks of a car but not so much for regular noises such as the constant flow of cars on a highway as your brain quickly adapts and filters them out. If you sometimes wake up due to noise you should try to sleep with earplugs, they work wonders in hostels but can also improve your regular sleep at home.
- Temperature: If you sleep in conditions that are too hot or too cold your sleep will be affected negatively. Most people have probably experienced the upper ends of this scale when sleeping in a very hot room in the summer or the lower end when camping with a light sleeping bag under cold conditions. On average people prefer a room temperature of 18-22 degrees Celcius.
- Oxygen: Not much to say here. Your body needs oxygen. This doesn’t stop while we sleep. If possible open a window.
Honestly, I think most of us have been told these things since we were children but are either too lazy to care about them or lack the inertia to e.g. try sleeping masks or earplugs. Thus, this section is less about learning new things but rather a friendly nudge and reminder for you. If you have lacked the inertia to improve these really simple things, order the earplugs now and set yourself an alarm to open the windows before you go to bed.
Besides optimizing our environment we can and should also optimize our behaviour w.r.t. sleep. There are tons of good guides out there, e.g. by the sleep foundation, the NHS, headspace, James Clear’s blog, and many more which I will try to condense and summarize in the following.
Sleep the right amount: There exist recommendations on the optimal amount of sleep but their help is limited. They recommend that young adults should sleep between 7 and 9 hours and you should adapt your length of sleep based on many conditions including your work, your health, your stress level, the amount of exercise on a given day, your caffeine intake, etc. I guess the most important information is that you should try to prevent sleeping less than 7 hours over prolonged periods of time if you are a young adult. Otherwise, this information is a bit useless as they don’t give clear guidelines on how much I should shorten or extend my average amount of sleep for a given condition. In the ‘Fancy Gadgets’ section, I will explore whether this question can be meaningfully answered by technology.
Make a sleep schedule: All the guides basically agree on two things: a) Do not accumulate sleep debt. If your regular amount is 8 hours per night you can’t replace that with sleeping 6 hours for five nights and then 13 hours for two nights. Sleeping less to learn more in preparation for an exam or similar circumstance also leads to worse performance (see e.g. here). b) Don’t change your rhythm drastically. If you have found a working rhythm, e.g. from 11 pm to 7 am you should not change that. This includes weekends and holidays. The more you change the less rested you will be and the more sleep you need overall. As alluded to above your SCN adapts very slowly and too drastic changes will decrease sleep quality.
Get a pre-bed routine: This essentially boils down to getting yourself physically ready for bed and wind-down mentally. Possible strategies include: stop working, getting away from screens, relaxing exercises such as yoga, meditation or guided wind-downs. My girlfriend uses a sleep routine from headspace and it seems to work well for her. This seems to be different from person to person though. I usually get so bored from the wind-downs that I get annoyed by it and then can’t sleep because I’m annoyed. Another option is to distract your mind and thoughts that keep you awake by listening to podcasts.
Keep your bed for sleep and sex only: Your bed should be a comfort zone and not associated with anything that could cause stress or negative thoughts. If you work or watch Netflix on your bed, you might associate it with negative things or emotional experiences and thereby increase the time you need to get in the zone. While this is sound advice, whether you can actually follow it or not seems to mostly depend on your financial situation. While I was a student my rooms were so small that it was impossible to separate the categories entirely.
Apply pro-sleep behavior during the day: There are plenty of things you can do during the day to improve your sleep at night. You can exercise - it will reduce the time it takes to fall asleep. I think regular intense exercise (e.g. 1h of cycling) has had the biggest positive effect on my overall quality of sleep. See the light of day and reset the melatonin in your brain. Don’t eat too late. The exact reason for why it hampers with sleep quality seems to be unclear but many people report that it does. Don’t drink alcohol. While it may help you to fall asleep it will increase sleep disruptions later when liver enzymes metabolize the alcohol during the night (see e.g. here). Don’t smoke. It is generally terrible for your body, nicotine makes you wake up similar to caffeine and smokers have a higher risk of breathing-related sleep problems during sleep (see e.g. here). Don’t drink too much caffeine, especially close to bedtime.
Try to sleep only when you’re tired: If you can’t sleep there is no point in focusing on it too much. It will only make you feel bad for missing out on important minutes or hours of sleep while time seems to stand still. The general recommendation is that if you can’t fall asleep after 20 minutes of trying you should do something else. Read a book, listen to a podcast, meditate, etc. However, you should avoid bright screens and you should be able to stop the activity whenever you feel tired again and go back to bed.
Screens: While the general recommendation is to avoid screens entirely before sleeping, I think this is not always practical. Sometimes I want to read something online or as a PDF or maybe just watch a dumb YouTube video to wind down. I would recommend using programs like redshift or a native night mode on your computer. You can adjust the color brightness and the amount of red in it to simulate natural light. Since I have started using these programs I don’t have screen-related sleeping problems and I have gotten so used to them that I get headaches and my eyes hurt on computers without these programs. Most modern smartphones also have some sort of night mode that you should activate. It turns the screen black and white and disables most notifications. As soon as it turns on, I just don’t want to look at my phone anymore - it’s astonishingly effective for me.
Power naps: If you are very tired throughout the day you can do power naps. The two things that are always emphasized are a) Don’t power nap too late. Optimally, you should do it after lunch but before 4 pm. b) They should not last longer than 20 minutes. Otherwise, you get into the deeper sleep phases and will feel groggy for the rest of the day. For most people, it requires a bit of training to fall asleep quickly and get back up after 20 minutes but it is manageable with a bit of training. While I was skeptical a few years ago, I now use them regularly. Whenever I feel too tired to be productive I just sleep for 10-20 minutes and return fresh. It’s much preferable to an hour of low productivity after lunch in my personal experience.
There are many gadgets and apps out there that promise to improve your sleep. While I was very skeptical in the beginning I was still curious and wanted to try one out to see how well it works. I could very quickly exclude all apps that infer your quality of sleep through the motion sensors of the phone which you are supposed to lay on your mattress. While this probably provides some data, this channel seems way too noisy for me to be actually useful. Additionally, these apps don’t incorporate information about your day (e.g. exercise levels) and therefore can’t give you any actionable recommendations.
After reading a bit further I could boil the competition down to two options: the Oura Ring and Whoop. The Oura Ring is, as the name suggests a ring that you keep on all the time and it measures blood flow and other values in your finger to infer heart rate, breathing patterns, etc. Whoop uses a bracelet and basically infers similar information from your wrist instead of your finger. While the Oura Ring is solely focused on sleep, Whoop can also be used for exercising or computing general strain throughout the day. If you want to use the gadget solely for sleep I think none of them are clearly superior. Since I recently picked up cycling again, I also wanted to make use of the exercising functions and therefore decided in favour of Whoop. This video made me curious and convinced me to test Whoop.
I have been using the Whoop strap since October 2020 and so far I found two things. Firstly, the measurements and their interpretations seem to broadly align with my intuition. It detects pretty accurately when I sleep and when I wake up and most of the time when I feel well-rested the app says my recovery is high and when I feel tired the app tells me my recovery is low. However, Whoop uses heart rate variation as a proxy for recovery which is already imperfect for physical activity but definitely doesn’t capture mental activity as much as it should and might thereby skew the results a bit. Secondly, whenever my intuition and whoops measurement disagree, the app is far more often correct than I am. When I wake up and feel good but the app says I’m not well recovered I usually get tired and unproductive earlier than usual. I also think that it is not a self-fulfilling prophecy, i.e. I only get tired because I expect it due to whoop, because I often forget what Whoop told me in the morning and only remember once my unproductive phase hits me.
I don’t think the app is a must-have. The reason why I will probably keep using it and therefore am willing to spend 0.5-1€ per day for it is two-fold: a) Just tracking the amount of sleep I get is already helpful. Even though you might have a rough estimate of the length of last night’s sleep without a sleep tracker, you probably don’t have a good intuition for the last week or last month. Just having the actual numbers is already informative to evaluate your current need or see patterns. And having the app puts more attention on sleep already making it more likely that I prioritize it higher. b) The app is a good indicator of when I need to take a break. When I’m really tired and have an unproductive day, I usually feel bad for not getting things done. If Whoop says that my recovery is terrible after working long, I have to take time off. Since the app “made the decision for me” I don’t feel guilty and can focus on other tasks that need less attention.
After writing the original post, I found eightsleep. It’s a start-up that optimizes everything around sleep and their reviews indicate that they do so quite successfully. Unfortunately, their products are only available in the US and Canada so far and too expensive for me, but when I have more disposible income, I would be curious to try them out.
There is also a new (German) app called Mementor which seems to have pretty promising results in randomized controlled trials (see here). Unfortunately though, you have to get a prescription from an actual doctor to be allowed to use the app and it is only designed for more serious problems related to sleep so I can’t try it out. If you have personal experience with the app and want to share it please reach out to me.
From my research and my own experiences over the last month I find three main conclusions:
Sleep is very very individual but you can improve your own sleep after finding out what your preferences are. There are no one-fits-all solutions but by thinking about what type of sleeper you are realizing your own bad habits, etc. you can improve your sleep significantly. Especially if you have problems falling asleep, disrupted sleep, or pain in a body part like the spine or neck after sleeping there is a good chance you can solve them.
It makes sense to invest time and money to improve your sleep. Sleep just has a large influence on many things that are important to us. If you are well-rested you are more productive, you feel better, and you are less likely to be annoyed quickly and emotionally hurt the people you like. All of these are things we value highly and thus should be willing to invest time and money to improve.
Most of this seems pretty obvious you just gotta make it a priority and do it. Most of the insights presented in the literature and summarized in this post are not revolutionary. Sleeping the right amount improves how rested you are the next day - no shit. But the fact that they seem so obvious is dangerous because you nearly feel like “you do them already because they make intuitive sense”. However, at least I, find myself too often in a situation where I go to bed too late even though I know that I have to get up early the next morning or do other things that go against the classic advice. I think optimizing your sleep usually is more a problem of implementation rather than lack of knowledge. If you have trouble following the advice I can recommend to nudge yourself and thereby make your choice easier. Use technology to help you. Set yourself a time after which certain behavior is not allowed anymore or make a daily short walk part of your routine.
One last note
If you want to get informed about new posts you can subscribe to my mailing list or follow me on Twitter.
If you have any feedback regarding anything (i.e. layout or opinions) please tell me in a constructive manner via your preferred means of communication.