Sleep efficiency: formula, calculation and interpretation

Sleep efficiency is one of the most cited metrics in sleep research: it relates how much time you spend in bed to how much of that time is actually sleep. It underlies many cognitive behavioral therapy for insomnia (CBT-I) protocols, and it's the metric many wearables report as part of their sleep score. This guide covers the exact formula, a full worked example, and why two different tools can give you slightly different results for the same night.

The formula

Sleep Efficiency (%) = Total Sleep Time ÷ Time in Bed × 100. Total Sleep Time (TST) is how much you actually slept. Time in Bed (TIB) is how long you stayed in bed in total, sleep plus wakefulness: sleep onset latency, night-time awakenings, and the time spent awake in bed before getting up in the morning.

Full worked example

You go to bed at 11:30 PM and get up for good at 7:15 AM: time in bed = 7 hours 45 minutes (465 minutes). It takes you 15 minutes to fall asleep, you're awake 20 minutes during the night, and you stay awake in bed 5 minutes after your final wake-up before getting up: total awake time = 40 minutes. Sleep time = 465 − 40 = 425 minutes. Sleep efficiency = 425 / 465 × 100 ≈ 91.4%.

You can redo this exact calculation, and try your own numbers, with the FitMesh Labs Sleep Efficiency Calculator: simple mode (just time in bed and time asleep) or advanced mode (with clock times, latency and awakenings), all calculated in your browser, no data sent to a server.

Time in bed vs. time asleep: why the difference matters

Low efficiency doesn't necessarily mean "little sleep": it can mean you spend a lot of time in bed awake, from difficulty falling asleep, frequent awakenings, or simply the habit of staying in bed awake before getting up (reading, on your phone). Efficiency isolates this from absolute sleep time: two nights with the same number of hours slept can have very different efficiencies depending on how much extra time you spent in bed not sleeping.

Why the denominator is debated

The definition this guide uses (TST / TIB × 100) is the most common, but it's not the only one in the literature. A study published in the Journal of Clinical Sleep Medicine (Reed & Sacco, 2016) shows that choosing a different denominator (for example "sleep episode duration" instead of raw time in bed) can shift the result by several percentage points for the same night. That's not a flaw in the formula: it's one more reason to know exactly which definition the tool you're looking at is using, before comparing numbers across different devices.

How wearables estimate sleep time

• •Actigraphy: estimates sleep from movement (accelerometer): less movement = likely sleep. The historical method, also used in clinical research, but can confuse being still while awake with sleep.
• •PPG (photoplethysmography): adds heart rate/HRV signal to movement to classify sleep stages more precisely, typical of modern rings and watches.
• •Sleep diary (self-report): what you perceive when you wake up, often less precise for exact latency and brief awakenings you don't remember, but it's the only method that needs no device.

Limits

Sleep efficiency is not a diagnosis and doesn't replace a clinical assessment (polysomnography, professional actigraphy, a consultation with a sleep physician). A single number, a single night, says little: the sleep literature treats efficiency as most useful as a trend over time, under comparable conditions, not as an isolated value to judge.