Sleep efficiency

The real measure of restful sleep

Sleep efficiency answers a simple question: how much of the time you spend in bed are you actually sleeping? The formula is straightforward: total sleep time divided by time in bed, times 100. It is the same definition sleep clinics use in polysomnography reports and the American Academy of Sleep Medicine uses in its scoring manual, which makes it one of the few wearable metrics with a clinical twin.

If you go to bed at 11pm, fall asleep at 11:20, wake briefly twice for ten minutes each, and get up at 7am, you were in bed 8 hours and asleep 7 hours 20 minutes: efficiency 92 %. That's solid. Lie in bed scrolling until midnight and wake at 5:30 to lie there until the alarm, and the same seven hours of sleep are spread over nine and a half hours in bed: 74 %.

What the numbers mean

• 90 % and above: excellent. You're falling asleep quickly and staying asleep
• 85 to 90 %: the healthy range for most adults; 85 % is the threshold the National Sleep Foundation's consensus panel called appropriate across all ages (Ohayon 2017)
• Under 80 %: something's getting in the way. Could be screen habits, caffeine timing, stress, or an underlying sleep issue

Two cautions. Efficiency above 95 % night after night is not a trophy; combined with short sleep it usually means you are so sleep-deprived that you fall asleep instantly and never surface, which is a sign of debt rather than of good sleep. And efficiency declines a little with age: 85 % at 30 and 80 % at 70 are the same story.

The three parts of a lost night

Efficiency falls for three reasons, and the fix depends on which one it is:

• Sleep latency, the time to fall asleep. Under 20 minutes is normal; 30 minutes or more most nights is the textbook marker of sleep-onset insomnia. Late caffeine, an early bedtime that is ahead of your body clock, and a racing mind are the usual drivers
• Wake after sleep onset (WASO), the minutes awake between falling asleep and the final wake-up. Under 30 minutes is healthy. Alcohol, a full bladder, a warm room, apnea and pain all push it up
• Early waking, then lying in bed. Common in depression, with age, and simply when bedtime is earlier than the sleep you need

How wearables measure it

A watch has no EEG. It infers sleep from movement and heart rate, and the boundary decisions are where the error lives: lying still while awake often scores as light sleep, which inflates efficiency, and restless sleep can score as awake, which deflates it. Against polysomnography, consumer devices agree on sleep versus wake about 90 % of the time but miss a large share of wake epochs (Chinoy 2021). "Time in bed" is also a guess about when you lay down. The absolute percentage is therefore soft by a few points; the night-to-night comparison on one device is what carries information.

How to raise it

The insomnia treatment with the best evidence, CBT-I, is built on sleep efficiency: sleep restriction means limiting time in bed to roughly the hours you actually sleep, then adding 15 minutes a week as efficiency climbs above 85 to 90 %. The bed becomes a place where sleep happens quickly. Alongside that, the levers in the sleep hygiene entry: a fixed wake time, no screens in bed, caffeine before early afternoon, a cool dark room and alcohol kept away from bedtime. If you lie awake for more than 20 minutes, get up and return when sleepy.

In MyBodyAI

Sleep efficiency is one of the seven components of the Sleep Quality index, computed as total sleep time over time in bed, with time in bed taken as sleep plus the awake minutes the device recorded plus the time it took to fall asleep where the device reports it. Your wearable tracks when you fall asleep, when you wake, and every restless moment in between, giving you an honest picture night after night.