How Fitbit Created the Sleep Profile Experience

Fitbit is always looking for ways to help users sleep better. You may already be aware of some of the ways your Fitbit can help you get a better night’s rest, especially if you check your sleep score regularly, but did you know we dial in deeper sleep insights to help you understand your sleep health? I did A whole new way?

Personalized Sleep Profile with Fitbit Premium goes beyond nightly tracking to analyze your sleep over a month. It consists of three components: 10 key sleep metrics, education and guidance, and your monthly sleep analysis based on your sleep animal. Curious about how Fitbit researchers went about creating this important new feature? Read on to learn more—straight from the source.

We chatted with Karla Gleichauf, a research scientist at the Consumer Health Research team, which uses Fitbit biometric data to help users sleep better. In addition to inventing and developing sleep profiles, his recent work includes studies on sleep population trends and health outcomes. With nearly 7 years under her belt as a Fitbiter, Carla has worked in a variety of research areas including physical activity, diabetes and now sleep. Prior to Fitbit, Carla received a PhD in Environmental Fluid Mechanics from Stanford, working on boats in coastal environments and modeling their flow. Carla moved from working with sensors in the water to sensors on the wrist.

Where did the idea of ​​the sleep profile come from?

Carla: We wanted to help users learn about their sleep patterns and understand how they can sleep better. That means creating an experience that assesses users’ sleep health with their Fitbit data. Healthy sleep is not the absence of sleep problems but promotes physical and mental well-being and is associated with positive health outcomes.

If we could measure users’ sleep health, we would also have the opportunity to see whether there are distinct sleep patterns, a question that is an active area of ​​research in sleep science. We thought that asking a user about their sleep patterns would be a fun, accessible and non-judgmental way to learn about their sleep. It will also provide a benchmark for people who sleep like them. With Sleep Profile, we are able to provide scientific rigor packaged in an enjoyable way to provide insight and guidance to users.

Why is Fitbit well positioned to take on this challenge?

Fitbit has been tracking sleep with metrics like Sleep Score since 2009, helping users get a more granular look at their overall sleep and including tips for how to get more rest and recharge. There are key benefits to using a Fitbit to track sleep, even from a sleep lab For one thing, you can see as much of a holistic view of sleep metrics as you want. Thanks to Fitbit’s history, and the proliferation of devices, we set out to analyze 22 billion hours of sleep data.

What challenges did the team face?

There is no phenotyping in the literature that includes the data we capture on Fitbit. Aside from known sleep patterns—like night owls and morning larks—there aren’t universally recognized healthy sleep patterns that we can pinpoint, so we turned to the data to see if we could find distinct patterns. It was a cross functional effort with people from Google and Fitbit’s data visualization, product, design, user research (UXR), clinical, and research and development teams.

How did the team determine sleep characteristics?

Fitbit researchers developed thousands of features inspired primarily by academic research and observations of Fitbit’s largest longitudinal sleep dataset available. For example, we developed several metrics describing a user’s difficulty falling asleep, such as the probability of a user waking up in the first hour of sleep and the time it takes them to reach “sound” sleep, inspired by the consistency of sleep literature. We also created sleep architecture features, such as average sleep length and how long it takes you to go through different stages of sleep.

We then narrowed down the features to the most important ones using unsupervised machine learning and statistical techniques. We divided the population into subpopulations and found that there were 6 sleeper type clusters. We then investigated which characteristics were most associated with each sleep cluster, and worked with sleep experts at home (sleep neurologists, clinicians, and researchers) and in academia (such as Dr. Michael Grandner and Dr. Alison Seaburn, as well as Dr. Logan Snyder). To determine which metrics are most important for getting good sleep. We ultimately landed on 10 longitudinal sleep characteristics in the “Monthly Sleep Analysis.”

These five metrics are all new to Fitbit, describing users’ sleep habits (sleep schedule variability, days with sleep), sleep maintenance (nights with long wakefulness, sleep consistency) and sleep onset (time before sound sleep).

What did the team discover?

We wondered whether a higher incidence of short wakefulness, which was 30 seconds or longer, was associated with better health and fitness, such as lower BMI, lower RHR, and higher active minutes. We discovered that three minutes or more of wakefulness was the threshold when health outcomes began to become more negative. Additionally, we found that there are 6 distinct sleeper types. The most common type of sleeper, giraffes, don’t sleep much but sleep soundly when they do. The least common sleep type, dolphins, usually go to bed at variable times, they sometimes stay awake longer, and they nap more frequently than most.

How did you turn education into a sleeping animal?

Once we identified sleep patterns, we had to decide: What is the appropriate metaphor? How can we express the metaphor of a globally available product in a culturally sensitive way? We tested animals vs. other options and decided that animals are the most relatable. Then, we had to figure out which animals to use. Our UXR team spoke to academics and experts on spokecharacters, animism, totemism and animal sleep patterns.

do you know

  • Animals with the highest proportion of females are turtles, who fall asleep more slowly and sleep a reasonable amount overall.
  • The sleeping animal with the oldest users is the hedgehog, which typically gets less deep and REM sleep. Males also have the highest number of hedgehogs.
  • The most common type of sleeper is the giraffe, who doesn’t sleep much but sleeps soundly when they do.
  • The least common sleep type is dolphins, who usually go to bed at variable times; They sometimes have long wakefulness and take naps more frequently than most.

What can users do with this information?

Users can learn how to sleep better from their sleep profiles. Your sleep animal highlights your overall sleep patterns and can help you understand how you sleep compared to others. Your monthly sleep analysis assesses 10 dimensions of your sleep health and shows you where you’re doing well and where you can improve.

For example, last month my sleep animal was a giraffe because I tended to go to bed late, slept less than most users, but didn’t have much time to stay awake. My monthly sleep analysis showed that my “sleep onset time” and “sleep schedule variability” were above the ideal range, highlighting that these are areas I could improve.

Fitbit continues to innovate in sleep tracking. In research, design and beyond, we see sleep as one part of a broader set of health and wellness patterns. The amount or duration of exercise affects sleep, nutrition interacts with sleep quality, long-term changes in daily habits can affect key sleep metrics, and Fitbit helps us integrate all of those areas to help our users live their best lives.

Leave a Reply

Your email address will not be published.