Sleep Efficiency: What the Number Means and What It Doesn’t

Sleep efficiency describes how much of your time in bed was spent asleep. It is usually expressed as a percentage. A night with little wakefulness after you get into bed tends to produce a higher value. More time trying to fall asleep, waking during the night, or lying awake before getting up can lower it.

The metric is useful because “eight hours in bed” and “eight hours asleep” are not the same measurement. Someone might go to bed at 10:30 p.m., take 30 minutes to fall asleep, wake for 25 minutes overnight, and spend another 20 minutes awake before getting up. Their time in bed would be longer than their total sleep time.

Sleep professionals often consider sleep efficiency alongside other measures such as sleep onset latency, wake after sleep onset, total sleep time, symptoms, and daytime function. The American Academy of Sleep Medicine includes sleep efficiency among sleep-continuity outcomes used in insomnia research and treatment evaluation. That context matters. The percentage is one part of the picture rather than a stand-alone verdict on whether you slept “well.”

Time in bed and total sleep time are different inputs

Time in bed usually covers the interval you intend to spend in bed for sleep. Total sleep time is the amount of that interval estimated to have been spent asleep. The exact values can differ depending on whether they come from your memory, a sleep diary, actigraphy, a consumer wearable, or polysomnography in a sleep lab.

That measurement difference is easy to miss. If your watch says 91% and your sleep diary works out to 84%, the two numbers may be based on different sleep/wake estimates. It does not automatically mean one is “wrong.”

The basic formula divides total sleep time by time in bed, then multiplies the result by 100. The arithmetic is simple. Getting reliable inputs is the harder part.

The formula stays the same, but the way sleep and wake are detected can change the result. A diary uses your report. Actigraphy infers sleep and wake mainly from movement. Polysomnography records signals such as brain activity, eye movement, muscle activity, breathing, and other physiologic data during a sleep study.

Why wearable efficiency can run high

Many wrist and bedside devices are good at identifying sleep but have more difficulty identifying quiet wakefulness. A 2025 validation study of six commercial wrist-worn devices found that most differed significantly from polysomnography on total sleep time, sleep efficiency, and wake after sleep onset. The devices detected more than 90% of sleep epochs, while wake specificity ranged from about 29% to 52%.

A separate study of more than 400 nights found that an under-mattress consumer sensor overestimated sleep efficiency by an average of 9 percentage points compared with polysomnography. Performance was more variable during daytime sleep and among people with sleep disorders.

These findings do not make wearables useless. They show why a wearable percentage is better treated as an estimate and trend signal than a clinical measurement. Look at patterns over time and consider how you actually feel and function.

A sleep efficiency value describes one feature of a sleep period. It does not identify the cause of wakefulness. Two people can have the same percentage for very different reasons.

One person may spend extra time in bed reading before sleep. Another may take a long time to fall asleep. Someone else may wake repeatedly because of pain, hot flashes, caregiving, noise, reflux, breathing problems, medication effects, or an irregular schedule. The resulting percentage can look similar even though the underlying situations are not.

Clinical research sometimes uses particular sleep-efficiency thresholds within a defined population or treatment protocol. Those cut points should not be turned into a universal diagnosis for everyone using a smartwatch. The American Academy of Sleep Medicine’s insomnia guidance treats sleep efficiency as one sleep-continuity outcome alongside sleep latency, wake after sleep onset, symptom severity, daytime outcomes, and other information.

Anything that increases awake time while you are in bed can lower the calculation. Sometimes that reflects a sleep problem. Sometimes it reflects behavior, schedule, or the way the measurement was collected.

A bad night can simply be a bad night

A single percentage is especially easy to overread. Travel, stress, illness, an unusual bedtime, alcohol, a noisy neighbor, or a child waking overnight can change one night’s result. A short run of sleep-diary entries usually gives more context than repeatedly checking one morning score.

NHLBI recommends that people preparing to discuss insomnia symptoms with a clinician consider keeping a sleep diary for one to two weeks. Recording bedtime, wake time, naps, daytime sleepiness, caffeine, alcohol, and exercise can make a pattern easier to see.

The percentage matters most when it matches a persistent problem you can feel in daily life. Consider discussing your sleep with a healthcare professional if difficulty falling asleep or staying asleep keeps recurring, you regularly wake unrefreshed, or daytime sleepiness affects work, school, driving, or other routine activities.

NHLBI notes that insomnia evaluation may include questions about how long it takes you to fall asleep, how often you wake, how long you stay awake, how refreshed you feel, your schedule, medicines, caffeine or alcohol use, and symptoms such as loud snoring or waking short of breath. A sleep diary can help organize that information.

A clinician may consider sleep efficiency as one piece of the assessment. Depending on the symptoms, they may also look for another cause of disrupted sleep rather than trying to “fix the percentage” itself.

The sources below were used for the measurement, wearable-accuracy, and clinical-context statements on this page. Manufacturer or marketplace content was not used.