Worldwide shipping, free over £80

30-day returns · 12-month warranty

No subscription, every metric included, forever

Sleeping while a smart ring tracks sleep stages overnight

What Is a Good Sleep Score? How Sleep Scores Work by Brand

A good sleep score is usually 80 or above on a 0 to 100 scale, and 85 or above counts as the top band on Oura and Whoop. But every brand builds its score differently, so a number from one app cannot be compared with another. What matters most is how your score moves against your own normal.

This guide explains how the main brands build their sleep scores, what typically goes into them, why an 82 on one device is not an 82 on another, and what to do with yours.

Key takeaways

  • Most sleep scores run from 0 to 100. On Oura, Fitbit and Garmin, 80 or more counts as good, and the top band starts at 85 or 90.
  • Scores mix several inputs, typically total sleep time, how long you lay awake, how often you woke, sleep stages and overnight heart data. Each brand weights them differently.
  • Because the formulas differ and are not published in full, scores are not comparable across brands or even across years on the same app.
  • Total sleep and a regular schedule are the inputs most within your control, and they are also the ones with the strongest evidence behind them.
  • Read the trend over a week or two. One low score is not a problem, and chasing a perfect number can make sleep worse.

What is a good sleep score?

On the scales the big brands publish, a score in the 80s or above is good, and the bands below that suggest room for improvement. The exact cut-offs vary, as this comparison of each brand's own help pages shows.

Brand Scale Top band Good Lower bands
Oura 0 to 100 85 to 100: Optimal 70 to 84 60 to 69 Fair, 0 to 59 Pay attention
Fitbit (Google Health app) Up to 100 90 to 100: Excellent 80 to 89 60 to 79 Fair, below 60 Poor
Garmin 0 to 100 90 to 100: Excellent 80 to 89 60 to 79 Fair, below 60 Poor
Whoop Percentage 85% or more: Optimal 70 to 85%: Sufficient Below 70%: Poor

Two things stand out. First, "good" on Oura starts at 70, while on Fitbit and Garmin a 70 sits in the Fair band. Second, the average person does not score in the top band. Google's help page says most users see an average Sleep Score between 72 and 83. If you sit in the 70s most nights, you are in common company, not failing.

How do different brands build a sleep score?

Every brand starts from the same raw signals: movement from an accelerometer and heart data from optical sensors. What differs is which measures they pull out, how they weight them, and what they compare you against.

Oura

Oura's help page says its score analyses resting heart rate, average body temperature, movement and time in each sleep stage. Named contributors include total sleep time, sleep efficiency (the share of the night you spend asleep) and latency (how long you take to fall asleep). Oura describes total sleep time as the most significant contributor, and suggests giving yourself an 8 to 10 hour window in bed to reach 7 to 9 hours of actual sleep.

Fitbit (Google Health app)

Google's page lists six metrics: sleep duration, time to sound sleep, time in sound sleep, restlessness, full awakenings (wake periods longer than five minutes) and interruptions. It says fewer than two full awakenings is considered good for most people, and that targets vary with age, gender and the time you spend trying to sleep. It does not publish point weights for each metric.

Garmin

Garmin's score, built on Firstbeat Analytics, uses heart rate, heart rate variability, respiration rate, body movement and other data from a watch worn overnight. In the same 2023 post, Garmin reported that only 5 percent of its users averaged an Excellent score over the final quarter of 2022.

Whoop

Whoop takes a different approach. Its Sleep Performance score is a percentage built from four parts: sleep sufficiency, sleep consistency, sleep efficiency and sleep stress. Sufficiency compares the sleep you got with a personal sleep need that moves with your recent training, sleep debt and naps. Consistency rewards regular bed and wake times, and Whoop says it needs three nights of data before it can score it. Whoop's page also notes that the score used to be based only on hours slept against hours needed, before the update.

What usually goes into a sleep score?

Strip away the branding and most scores draw on the same handful of ingredients.

  • Total sleep time. Almost always the biggest single factor. The American Academy of Sleep Medicine and the Sleep Research Society recommend that adults get seven or more hours of sleep per night.
  • Efficiency. Time asleep as a share of time in bed. Whoop treats 90% or more as optimal.
  • Latency. How quickly you fall asleep.
  • Wake-ups and restlessness. How broken the night was.
  • Sleep stages. Time in light, deep and REM sleep. Our guide to sleep stages explains what each one does.
  • Timing and regularity. Whether you went to bed and got up at similar times to usual.
  • Overnight heart data. Some brands fold in resting heart rate, heart rate variability or breathing rate.

Why can't you compare sleep scores across brands?

A sleep score is a summary of estimates, run through a private formula. There are three reasons two devices can give you different numbers for the same night.

The formulas are different

Whoop's score leans on regularity and how much sleep you needed. Oura weights total sleep time most heavily. Fitbit counts full awakenings over five minutes. A short but regular night could score well on one and poorly on another, even if both devices measured it identically.

The underlying measurements differ too

Wearables estimate sleep from movement and heart signals, not brain waves, so the inputs themselves are estimates. A study in the journal Sleep tested seven consumer sleep trackers against polysomnography, the lab test that records brain activity. The devices were good at detecting sleep, but their ability to spot time awake was low to medium, and their sleep stage results were mixed, with deep and REM sleep often recorded as light sleep. If two devices disagree about how long you lay awake, their scores will disagree as well.

The formulas change

Brands update their algorithms. Whoop's recent update added three new components to its score. A score from two years ago on the same app may not mean quite the same as one from this week. Compare like with like: the same device, the same app version, over the same few weeks.

PulseRing Pro reads your heart rate, HRV, blood oxygen and sleep stages overnight, builds your personal baseline over roughly two weeks, then gives you one readiness score and one plain sentence each morning: push, maintain or rest. There is no subscription, so every metric is included in the one-off £89 price for as long as you own it.

See what PulseRing Pro tracks overnight

What should you do with your sleep score?

Use it as a prompt to look at your habits, not as a grade. A simple routine works well.

  1. Look at the weekly average. Single nights jump around. A run of lower scores over a week tells you far more than one bad night.
  2. Open the contributors. A low total tells you little. A low "total sleep time" or "consistency" component tells you what to change.
  3. Fix time in bed first. If you are in bed for six and a half hours, no habit will get you to seven hours of sleep. Oura's own suggestion is an 8 to 10 hour window to land 7 to 9 hours asleep.
  4. Keep your wake time steady. The NHS advises going to bed and getting up at roughly the same time, including on days off. This is UK guidance, but it matches advice from sleep bodies elsewhere.
  5. Test one change at a time. Move caffeine earlier, skip the late drink or finish screens an hour before bed, then watch the trend for a week. Our guide to getting more deep sleep covers the habits with the best evidence.

How your sleep score relates to a readiness score

Many devices give you two morning numbers. The sleep score describes last night. A readiness or recovery score uses sleep plus heart rate variability, resting heart rate and recent activity to estimate whether you are ready to train. Sleep feeds into readiness, but a good night does not guarantee a high readiness score if you are run down from hard training or illness. Our guide to readiness scores explains how to use the second number.

Can a sleep score do more harm than good?

For some people, yes. Sleep researchers have described "orthosomnia", where people become preoccupied with achieving perfect sleep data. In case reports published in the Journal of Clinical Sleep Medicine, Baron and colleagues described patients seeking treatment for sleep problems they had inferred from their tracker, and noted that the data often felt more convincing to them than validated lab tests.

If checking your score makes you anxious, or you lie in bed worrying about it, take a break from looking for a week. Wear the device, ignore the number, and judge your nights by how you feel during the day. Then come back to the weekly trend.

When is a low sleep score worth taking seriously?

A consumer sleep score is a wellness measure. It cannot diagnose insomnia, sleep apnoea or anything else. What it can do is show a pattern. Speak to your GP or doctor if you regularly struggle to get to sleep or stay asleep, feel exhausted during the day despite enough time in bed, or a partner notices loud snoring, gasping or pauses in your breathing. Those are reasons for a proper assessment, whatever your score says.

Frequently asked questions

What is a good sleep score out of 100?

On Fitbit and Garmin, 80 to 89 is Good and 90 or more is Excellent. On Oura, 70 to 84 is Good and 85 or more is Optimal. Google says most Fitbit users average between 72 and 83, so a score in the high 70s or low 80s is normal. The trend against your own average matters more than any single number.

Is a sleep score of 70 good?

It depends on the brand. On Oura, 70 is the start of the Good band. On Fitbit and Garmin, 70 sits in the Fair band, and on Whoop, 70% is the bottom of Sufficient. A 70 is not a cause for concern on its own. If you are regularly below your usual level, look at the contributors to see what changed.

Why is my sleep score different on two devices?

Each brand uses its own formula and weights inputs differently. One may focus on total sleep time, another on regularity or how much sleep you needed. The devices may also estimate your time awake and sleep stages differently, since wearables infer sleep from movement and heart signals rather than brain waves. Compare each device only with itself.

How can I improve my sleep score?

Start with the basics that most scores reward. Give yourself enough time in bed to get at least seven hours of sleep, keep your wake time steady across the week, and cut back on late caffeine, alcohol and screens. Change one thing at a time and watch your weekly average rather than reacting to a single night.

Can a sleep score detect sleep apnoea?

No. A consumer sleep score is a wellness measure and cannot diagnose sleep apnoea or any other condition. If you snore loudly, a partner notices pauses in your breathing or gasping, or you feel very sleepy during the day despite enough sleep, speak to your GP or doctor, who can arrange a proper sleep assessment.

This article is for general information and is not medical advice. PulseRing Pro is a wellness device, not a medical device, and does not diagnose or treat any condition. If you are worried about your health, speak to your GP or a qualified clinician.

Free guide

Read your HRV like a pro athlete.

Get The Peak Readiness Guide free: six pages on building your baseline, the 60-second morning read and the four decisions it leads to. After that, one email a week on sleep, recovery and training.

Free. Unsubscribe any time.