A while back I did a big post that lays out what you need to know about the science behind sleep and sleeping better. But if you read it, you’ll probably ask the same question I do:

Hey, does this really work?

I don’t post this research so we’re all better prepared for Jeopardy. If it’s stuff that only works in a lab, well, I don’t live in a lab.

So I’ve been testing a few things. Like any mad scientist, on myself.

Here’s what I learned:

Seeing Trends Can Tell You A Lot

By using a cool little app called Sleep Cycle I was able to track a lot of data regarding how I sleep.

Merely seeing trends gave me more insight than you might expect into some bad habits.

I realized that the transition from the weekend back to the workweek was rough. I always screwed myself on Sunday nights:

help me sleep

What was happening? My weekend schedule was very different than my workweek schedule and this meant a lot less sleep on Sunday night.

As if I really needed something to make Monday mornings worse.

Studies show how you start the day affects the entire day and research on daylight savings time and jet lag shows sleep inconsistency hurts performance a lot more than you might expect.

This was also visible in the app’s scoring system. My sleep quality was lowest on Sunday nights.

help-me-sleep

What Gets Measured Gets Managed

Some of you may be wondering about the Hawthorne effect*. (Knowing a study is being performed makes people change their behavior.)

Actually, I was counting on it. After all, my goal is to improve, not to publish research.

I’ve posted research before that reminders are one of the most powerful things you can do to promote improvement. Sure enough, knowing this little app was watching gamed me into behaving better.

My guilt-driven-improvement is visible in the results. Over the course of measuring my sleep, I went to bed earlier:

help-me-sleep

Interesting that sleep quality got worse before it got better (By the way, 7/29 and 7/30 were pretty crazy days so I’m okay ignoring them.)

help-me-sleep

So What Did Help Me Sleep?

As I mentioned in my post on research regarding good sleep, these were the big four study findings from the excellent book Dreamland: Adventures in the Strange Science of Sleep:

Exercise during the day.
Create a cold sleeping environment.
Avoid light before bed.
Avoid alcohol and caffeine at night.

I already keep the bedroom cold as a meat locker and I rarely drink.

Though the numbers aren’t huge, it was interesting to see what variables had an effect on my sleep:

help-me-sleep

Here’s what the results showed and my analysis:

I drink an inhuman amount of coffee. It had minimal effect because, frankly, I don’t think there was a day when I didn’t have any.
Working out, as predicted by the research, definitely improved my sleep. I didn’t have much time to exercise so I feel pretty confident if my “study” had been longer and my workouts more frequent that number would be higher.
Eating late produced a small negative result. I’m pretty sure I know what that was: eating late also means more beverages late. Those nights I woke up to pee more often. No big thing but probably not optimal for sleep.
Stressful days had the biggest positive impact of anything; not that I’ll be looking to increase those, thank you. My guess is there was a vicious circle: sleep less, feel more stress the next day because of it and because I’m sleep deprived, sleep harder that night.

A couple other things worth noting:

I set an alarm every night to remind me to GO TO BED. I find this is almost more valuable than an alarm clock in the morning because when I go to bed early, getting up early is easier.
I never had any trouble getting to sleep. Part of this was due to being sleep deprived.
No, this wasn’t NIH funded and no, I didn’t have a twin Eric for a control group. But it was tested by a real me on a real me with some reasonable (though not rigorous) parameters, so it falls firmly into the category of good enough for now.

What Did I Learn?

A few things:

First off, I need more sleep. Within reason, more sleep is better sleep.
For both attitude and performance, Mondays would be better if my weekend schedule was more consistent with my workweek schedule.
Reminders are powerful. Knowing the app was watching kept me in line and a GO TO BED alarm made a much bigger difference than you might guess.
Exercise and a bit of stress (let’s stick to the good kind) really improve my sleep.

What’s next?

Keep doing what seems to be working: exercise, GO TO BED alarm, cold bedroom… and more sleep.
I always have a bright screen blaring into my eyes before I konk out. Need to try cutting that out.
Eating earlier so I’m not getting up to hit the bathroom.
Not drinking coffee. Haha, just kidding. That won’t happen anytime soon.

Your Turn

Merely reading this stuff won’t help you much.

Watching Monday Night Football does not make you a great quarterback and 60 years of sitcoms has not made Americans funnier, folks.

Check out what worked for me above and give something a shot for a week.

I’ll be following up with more results and tips on my mailing list, which you can join here.

Happy sleeping. More about getting better sleep in a recent post on the Buffer blog.

About the author:

Eric Barker uses the latest findings in the science of human behavior to improve our performance at work and at home. His blog, Barking Up The Wrong Tree, is listed on blogrolls at the New York Times, the Wall Street Journal and became a column in Wired Magazine.

You can get a collection of his most valuable insights by joining his free weekly email update here.

Eric Barker

How I Experimented on Myself: Here are 4 Things That Helped Me Sleep