Fitbit launched Sleep Profile last year as a Premium feature, which provides users with a detailed monthly analysis of their sleep. This feature requires users to use it for at least 14 nights over a month. Fitbit research scientist, Karla Gleichauf, explains that this feature was the result of a multi-year effort to answer the question of whether there are different types of sleepers. While this question had been of interest in academia, wearables like Fitbit provided the capability to answer it.

Fitbit’s team set out to gather useful data by engineering over 1,000 sleep features to capture users’ sleeping behaviors. These sleep features included factors such as the probability of waking up in the first hour of sleep, sleep cycle length, wake time consistency, and weekend versus weekday bedtime discrepancy. Using machine learning technology, the team clustered the user data and discovered that there are indeed different types of sleepers. In fact, they identified six different types.

However, the goal was not only to identify these different sleep types but also to help people improve their sleep based on this information. This led the Fitbit team to collaborate with colleagues on the product side to find a way to translate this mathematical information into something more relatable. Fitbit product manager, Elena Perez, explains that they brainstormed different ways to represent these sleep types, including using natural elements like storms or oceans. However, they found that people had a difficult time connecting to these representations.

Eventually, the team settled on using animals to represent the different sleep types. They worked with ethnographers and animal experts to determine the appropriate animals for each sleep type, eventually deciding on the bear, dolphin, hedgehog, giraffe, parrot, and tortoise. The team also delved deeper into the data to determine the optimal number and types of sleep profiles to help people understand their sleep health.

Overall, Fitbit’s Sleep Profile feature provides users with valuable insights into their sleep patterns and offers suggestions for improvement based on their sleep type. By using machine learning technology, Fitbit has been able to identify six different sleep types and translate this information into relatable representations using animals. This feature aims to help users better understand their sleep health and make positive changes to support better sleep.