Sleep Efficiency: Why 8 Hours in Bed Isn't 8 Hours of Sleep

You go to bed at 11. You get up at 7. That's eight hours, which sounds like a win.

But time in bed is not the same as time asleep, and the gap between them is one of the most useful numbers in sleep science. It's called sleep efficiency, and it answers a simple question: of all the time you gave to sleep, how much of it did you actually spend asleep?

The answer matters more than most people think. Two people can each spend eight hours in bed, one sleeping for seven and a half of them and the other tossing through six, and they'll feel completely different the next day. Same hours, different biology.

Sleep efficiency is the metric that separates those two nights. Here's what it is, what a normal number looks like, why your watch probably flatters you, and the counterintuitive fix that reliably improves it.

TL;DR: Sleep efficiency is total sleep time divided by time in bed, expressed as a percentage. If you're in bed for eight hours and asleep for seven, that's 87.5%. Around 85% or higher is generally considered good, 90%+ is excellent, and consistently under 85% suggests fragmented or inefficient sleep — the pattern clinicians look for in insomnia. The number is driven less by how long you sleep than by two things: how quickly you fall asleep and how often you wake up. The most effective lever isn't going to bed earlier — it's spending less time in bed, which is the logic behind sleep restriction therapy. And there's a catch with wearables: watches tend to overestimate total sleep time, so they tend to inflate your efficiency score.

YouTube: Related video

What sleep efficiency actually measures

The formula is simple:

Sleep efficiency = total sleep time ÷ time in bed × 100

If you get into bed at 11pm, fall asleep at 11:30, wake twice for a total of 20 minutes, and get up at 7am, you were in bed for 480 minutes and asleep for about 430. That's a sleep efficiency of roughly 90%.

Three things make that number move:

• Sleep onset latency — how long it takes you to fall asleep once you're in bed. Thirty minutes of lying awake costs you six percentage points straight away.
• Wake after sleep onset (WASO) — every minute you spend awake in the middle of the night.
• Time in bed itself — the denominator. This is the one people forget, and it's the one you have the most control over.

Notice what that means. A low sleep efficiency can come from sleeping badly, or from giving yourself an unnecessarily large window to sleep in. Someone who goes to bed at 9pm "to get more sleep" but doesn't fall asleep until 11pm has just manufactured a 65% efficiency night without doing anything wrong.

What counts as normal

There's no single official cut-off, but the research and clinical conventions cluster tightly.

The 85% figure shows up repeatedly because it functions as a working threshold in sleep medicine. It's the number used in behavioural sleep therapy to decide whether to expand someone's time in bed, and it's the rough line the general guidance draws between consolidated and broken sleep.

Here's the important caveat: these are population conventions, not personal verdicts. One night at 82% means almost nothing. Five nights in a row at 78%, with days that feel foggy, means something worth looking at.

Why the denominator is more slippery than it looks

There's a genuine definitional wrinkle here, and it's worth understanding because it changes how you read your own data.

The clean version of the metric uses time in bed as the denominator — from the moment you get into bed to the moment you get out. But "time in bed" isn't always sleep time. If you read in bed for an hour, or watch TV, or scroll on your phone before lights out, that hour is technically time in bed but has nothing to do with your ability to sleep.

That's why sleep researchers have argued for decades about the right denominator. A 2016 paper in the Journal of Clinical Sleep Medicine laid out the problem plainly: some studies use time in bed, others use total recording time (from lights out to lights on), and the same night's data can produce meaningfully different efficiency numbers depending on which one you pick. The inconsistency makes it genuinely hard to compare results across studies and clinical settings.

For your purposes, the practical takeaway is this: be consistent about what you're measuring, and be honest about non-sleep time in bed. If you regularly read for 45 minutes in bed before lights out, your watch will see that as time in bed and quietly mark you down — which is arguably correct, because you weren't sleeping.

What low sleep efficiency means for your health

Sleep efficiency isn't just an academic curiosity. It tracks with outcomes.

A 2023 study of 90,398 UK Biobank participants used accelerometers — device-measured, not self-reported — to look at sleep duration and sleep efficiency together against mortality risk. Lower sleep efficiency was associated with increased risk of all-cause and cause-specific mortality, independent of how long people slept. In other words, how consolidated your sleep is carries information beyond how much sleep you get.

That's a strong finding, and it fits with what we know about fragmentation. Every time you wake, you interrupt the architecture of the night — the progression through light sleep, deep sleep and REM that does the actual restorative work. Being in bed for eight broken hours is not the same as being asleep for eight continuous ones, and your body can tell the difference even if your sleep app rounds them together.

This is why frequent awakenings deserve real attention rather than a shrug. Fragmentation is often the mechanism behind a bad efficiency number.

The wearable problem: your watch probably flatters you

Now the part that complicates tracking.

Sleep trackers, including Apple Watch, infer sleep from signals like movement and heart rate rather than from brain activity. That's a genuinely clever approach, and in validation studies against lab-grade polysomnography, the better trackers land surprisingly close on total sleep time.

The direction of the error matters, though. Wearables tend to overestimate total sleep time, because quiet, still wakefulness is easy to mistake for light sleep. If you lie awake motionless for 40 minutes, your watch may score a good chunk of that as sleep.

The consequences are predictable:

• Total sleep time gets inflated — you appear to have slept more than you did.
• Sleep efficiency gets inflated — the numerator is too big.
• Wake time gets underestimated — which is exactly the thing you're trying to measure.

So if your watch says 92% but you remember waking several times, the discrepancy is probably not your memory being wrong. It's the tracker having a known blind spot in one direction.

The good news: even with that bias, the trend is still useful. If your efficiency score drops five points over two weeks, that change is meaningful even if the absolute number is generous. Consistency and direction beat precision, which is why tracking over weeks tells you far more than any single night does. If you want to know how much sleep you actually need by age — and whether your target is even right — that's the frame to start from.

How to raise sleep efficiency

Here's where it gets counterintuitive. Most people try to improve sleep by adding time — going to bed earlier, staying in bed later, chasing the eight-hour badge. That usually makes sleep efficiency worse, because you're extending the denominator while the numerator stays flat.

The most effective approach goes the other way.

1. Spend less time in bed (the sleep restriction principle)

This is the core of behavioural sleep therapy. You estimate how long you're actually sleeping, then set your time in bed to roughly match it — and no more.

If you're averaging six hours of sleep but spending nine hours in bed, you're teaching your body that bed is a place for lying awake. Compress that window to about six and a half hours and two things happen: sleep pressure builds higher, and you fall asleep faster and stay asleep longer. Efficiency climbs. Once it's consistently above roughly 85%, you extend the window by 15–20 minutes at a time and keep going as long as efficiency holds.

The American Academy of Sleep Medicine's 2021 clinical practice guideline recommends sleep restriction therapy as a treatment for chronic insomnia disorder, though it's a conditional rather than strong recommendation, and it's usually delivered as part of full CBT-I. A key point from the guideline: sleep restriction is not for everyone. It's contraindicated if you drive or operate heavy machinery, and it needs care in people with bipolar disorder or poorly controlled seizure disorders. If you're dealing with chronic insomnia, this is a conversation to have with a clinician or a properly trained therapist, not a DIY protocol.

For everyone else, the milder version is enough: stop treating an oversized time-in-bed window as insurance.

2. Don't get into bed before you're sleepy

This is one of the stimulus control instructions in the same AASM guideline, and it's simple. Bed should be strongly associated with sleep, not with lying there awake. If you're not sleepy, you're not ready for bed.

3. Get out of bed if you can't sleep

The guideline is specific: if you can't sleep, leave the bed. Staying for an hour, frustrated, teaches your brain that bed equals frustration. Ten to twenty minutes of lying awake is a reasonable tolerance before getting up and doing something quiet and dim until you feel sleepy again.

4. Fix your wake time, not your bedtime

A consistent wake time anchors your circadian rhythm far more reliably than a consistent bedtime. It's also the lever with the strongest evidence behind it for improving sleep quality. Same wake time, seven days a week, is the foundation everything else sits on.

5. Cut naps and late caffeine

Naps reduce sleep pressure, which is exactly the currency sleep restriction therapy relies on. If your efficiency is poor and you nap during the day, you've borrowed from the night. Likewise, caffeine's long half-life means an afternoon coffee is still active at bedtime, quietly extending how long it takes you to fall asleep.

Alcohol deserves its own warning. It's a sedative, so it can shorten sleep onset latency — making your efficiency look better — while fragmenting the second half of the night and crushing REM sleep. That's a case where efficiency improves while sleep quality gets worse, which is a good reason not to judge your sleep on one number alone.

6. Get the room right

Cool, dark and quiet is the boring but reliable combination. A cooler bedroom supports the drop in core temperature that helps sleep begin and continue, and the same principle sits behind the warm bath before bed trick: heat the periphery, cool the core. Blackout, earplugs and a thermostat do more for consolidation than most gadgets.

7. Rule out the medical causes

If your efficiency stays stubbornly low despite good sleep hygiene, look for a reason rather than pushing harder. Loud snoring, gasping, morning headaches and unrefreshing sleep point toward sleep apnea. Restless legs, chronic pain, acid reflux and prostate-related night waking all fragment sleep mechanically. And if low efficiency is driven by chronic insomnia, the first-line treatment is CBT-I — not a supplement, and not another sleep tracker.

Reading efficiency in your own data

If you want to use this number properly rather than obsess over it:

1. Track it for two to three weeks, not two nights. Efficiency is noisy night to night and only meaningful as a trend.
2. Compare against your own baseline, not the population figure. 87% might be excellent for you and a red flag for someone else.
3. Look at onset and wake time separately. A 78% score from a 45-minute sleep onset is a different problem from a 78% score from five brief awakenings. The fix is different too.
4. Change one thing at a time. Fix wake time for two weeks before touching anything else. Otherwise you'll never know what moved the number.
5. Take the watch number as directional. Given the known tendency to overestimate sleep, treat it as a relative signal, not a lab measurement.

This is where a tool like Century AI earns its place. Rather than chasing a single score each morning, it puts sleep, resting heart rate and HRV side by side against your own history, so you can see whether a change in sleep efficiency is showing up in your recovery too. A night at 82% after alcohol looks very different from a night at 82% after a hard training day, and the surrounding metrics are what tell you which is which.

And if you want to sanity-check whether you're sleeping enough in the first place, our guide to how much sleep you need by age is the right starting point.

Summary

• Sleep efficiency = total sleep time ÷ time in bed × 100. Eight hours in bed, seven hours asleep, is 87.5%.
• 85% or higher is generally good, 90%+ is excellent, and consistently under 85% points to fragmented or inefficient sleep.
• The number is driven by how fast you fall asleep, how often you wake, and how large a sleep window you give yourself — the denominator matters more than most people realise.
• A 2023 study of 90,398 UK Biobank participants found lower device-measured sleep efficiency was linked to higher all-cause mortality, independent of sleep duration.
• Wearables tend to overestimate total sleep time, which inflates efficiency. Use the trend, not the absolute number.
• The counterintuitive fix is less time in bed, not more — the logic behind sleep restriction therapy, which the AASM guideline supports for chronic insomnia.
• Fix your wake time, get out of bed when you can't sleep, cut naps and late caffeine, and check for medical causes like apnea if nothing moves.

Century AI helps you understand your body with a daily health score, recovery score, and sleep insights — using the watch you already wear.