You wake up, reach for your phone and check last night’s number before you have decided how you feel. Some mornings it says you slept badly when you feel fine; other mornings it looks good and you are still tired. Either way, the number tends to set the tone for the day.
That mismatch is not surprising. Trackers are good at telling sleep from wake, weaker at catching the moments you are awake, and inconsistent on deep sleep and REM.1 A single night says little, and for some people chasing better numbers makes sleep worse.3
This guide walks you through what your tracker can and cannot measure, how to step back if it worries you, the patterns worth acting on and how to read two weeks of your own nights, and when a doctor matters more than any app.
Start with the device itself: what it measures well, what it only estimates, and why a single night tells you so little.
Good at sleep versus wake, weaker on the rest#
A consumer tracker is reasonably accurate at telling sleep from wake across a night, less so at catching the moments you are awake, and inconsistent on sleep stages.1 Its sleep times and totals are worth more than its stage breakdown.
In a laboratory study, seven consumer sleep trackers were compared night by night with polysomnography, the full sleep study used in hospitals. Most performed as well as or better than a standard research wristband at separating sleep from wake, and all of them were very good at recognising sleep when it happened.1
They were much weaker at recognising wake: many brief awakenings were scored as sleep, and the devices did worse on disrupted nights.1 The volunteers were 34 healthy young adults, so accuracy in older adults or poor sleepers is less certain.1
Polysomnography, in one sentence
The full overnight sleep study, which records brain waves, eye movements, muscle activity, breathing and oxygen levels. It is the reference that sleep trackers are measured against.
Deep sleep and REM figures are rough estimates#
Measured against a laboratory sleep study, tracker estimates of sleep stages were inconsistent, so they cannot be trusted night by night.1 A low deep sleep figure is not, on its own, a reason to worry or to change anything.
Sleep stages are defined by brain activity, which a wrist or bedside device cannot record. It infers stages from signals such as movement and heart rate, and in the validation study those stage estimates were mixed from one device to the next.1
The American Academy of Sleep Medicine’s position is that consumer sleep technology cannot be used to diagnose or treat sleep disorders, but that its data can support a conversation with a doctor when it is looked at within a proper evaluation.2 That is a fair way to hold it: useful notes, not a verdict.
REM sleep, in one sentence
Rapid eye movement sleep, the stage most associated with vivid dreaming. A sleep laboratory measures it from brain waves and eye movements; a tracker can only estimate it.
One bad night on the app says little on its own#
Single nights vary for many ordinary reasons, and trackers are least accurate on exactly the disrupted nights that produce the worst readings.1 A pattern across two weeks tells you far more.
In the laboratory comparison, devices performed worse on nights of poorer or disrupted sleep.1 A low score after a restless night is therefore the score most likely to be off, in either direction.
What a tracker does offer is consistency: it measures you the same way every night, so it is good at comparing your own nights. We suggest fourteen nights because that covers two working weeks and two weekends, enough to compare nights after a late coffee, a glass of wine or a late dinner with nights without. Fourteen is a practical choice, not a threshold from research.
Knowing what one night cannot tell you also matters for how the numbers make you feel.
Understand your sleep data better
Eight short questions, and one of our experts reviews your case personally, then goes through it with you on a free call.
For some people, the morning number does more than inform: it starts to shape the night itself.
Chasing perfect sleep data can make sleep worse#
Sleep researchers coined a term for it, orthosomnia: a perfectionist pursuit of ideal tracker data that ends up making insomnia worse in some people.3
In 2017, sleep researchers described patients who sought treatment for sleep problems they had identified from their tracker, such as light or restless sleep, and whose quest for the ideal night had become part of the problem.3 For some of them, the tracker’s data felt more believable than a laboratory sleep study.3
If the morning score sets the mood for your day, or you lie in bed thinking about tonight’s number, that is a signal. Look at weekly averages only, or put the tracker in a drawer for a fortnight and notice how you feel.
The app and the way you feel often disagree#
The most common thread is a mismatch between the number and how people actually feel, in both directions. These are the phrases we hear most often.
- My score says I slept badly but I feel fine
- I check my sleep score before I get up
- I never get enough deep sleep
- My watch says I was awake for an hour
- Good score, still exhausted
- Same bedtime, completely different score
- I sleep worse when I wear it
- Is my sleep tracker accurate?
Setting the daily number aside leaves room for the patterns that are really worth your attention.
What a tracker shows well is your timing, and how your nights change with what you do. That is where the useful patterns are.
Act on patterns tied to things you can change#
Bed and wake times that move around, nights regularly under seven hours, and worse nights after late caffeine, evening alcohol, bright screens or a late, heavy meal all come down to something you can change.45678
| Pattern in your data | What the evidence says |
|---|---|
| Bed and wake times that shift, especially between workdays and days off | In a large cohort, the least regular sleepers had higher mortality, and regularity predicted it better than sleep length. An association, not proof of cause.4 |
| Most nights under seven hours | Adults are advised to sleep seven hours or more on a regular basis.5 |
| Later, shorter nights after an afternoon coffee | 400 mg of caffeine disturbed sleep even when taken six hours before bed.6 |
| Falling asleep fast after drinks, then a restless early morning | Alcohol consolidates the first half of the night and disrupts the second.7 |
| Later sleep after evenings on a bright screen | A light-emitting e-reader before bed delayed sleep and lowered melatonin.8 |
| More waking after a late or heavy dinner | Linked in early, small studies only.910 |
| Better nights on active days | Regular exercise has small to moderate benefits for sleep.11 |
Regular sleep times may matter as much as length#
In a large study using wrist sensors, people with the most irregular sleep timing had a higher death rate than more regular sleepers, and regularity predicted it better than sleep duration did.4
The study calculated a sleep regularity score from more than ten million hours of wrist data in 60,977 UK Biobank participants. Over about six years, the more regular four fifths had a 20 to 48% lower risk of death from any cause than the least regular fifth, after adjusting for age, lifestyle and health factors.4
It is an observational finding, so it cannot show that changing your timing changes your risk. But regularity is one of the patterns a tracker shows most reliably, because it rests on sleep and wake times rather than stages.1
Late coffee and evening alcohol leave visible traces#
In your data, late caffeine tends to show up as later, shorter nights, and evening alcohol as falling asleep quickly followed by a restless second half.67 Both are easier to spot across two weeks than in a single night.
In a study at home, 400 mg of caffeine taken at bedtime, three hours before or six hours before bed all disturbed sleep compared with a placebo, and the authors concluded that substantial caffeine should be avoided for at least six hours before bed.6 Comparing your own days with and without an afternoon coffee shows where your cut-off sits.
Alcohol works the other way round. It helps you fall asleep and steadies the first half of the night, then sleep becomes more disrupted in the second half, and the first REM period is delayed at every dose studied.7 Trackers miss many brief awakenings, so a restless early morning can look calmer in the app than it felt.1
Screens, late meals and exercise each play a part#
Bright screens late in the evening delay sleep, late and heavy meals may fragment it, and regular exercise helps modestly.8911 Your own data shows which of them matters most for you.
In a crossover study, people who read on a light-emitting device in the hours before bed took longer to fall asleep, released less melatonin, had their body clock shifted later and were less alert the next morning than when they read a printed book.8
The evidence on meals is thinner. In a survey of 793 university students, eating within three hours of bedtime was linked with waking during the night, and the authors call it preliminary.9 In a small controlled feeding study, more fibre went with more deep sleep, and more saturated fat and sugar with lighter sleep and more arousals.10
A meta-analysis of 66 studies found that regular exercise brings small benefits for total sleep time and moderate benefits for sleep quality, varying with age, sex, fitness, type of exercise and time of day.11
Each of these is easier to spot once you know how to read a fortnight of your own nights.
Read averages and spread, then change one thing#
Averages and spread tell you more than any single night: how much your bed and wake times vary, how often you sleep under seven hours, and how nights differ after the habits you noted. From there, change one thing at a time.
Patterns in your data can guide your habits, but they have limits. This last part covers when a doctor comes first, and where our work sits.
A tracker cannot rule out a sleep disorder#
Loud snoring with gasping or pauses in your breathing, sleepiness while driving, or months of struggling to sleep call for a doctor rather than an app. A reassuring reading does not mean there is no sleep disorder.
Obstructive sleep apnoea is common, and many people with it do not feel especially sleepy: in the general population, only 15 to 50% of people with it report excessive sleepiness.13 Untreated sleep apnoea clearly raises the risk of road crashes.14
For insomnia that has lasted months, the American Academy of Sleep Medicine strongly recommends cognitive behavioural therapy for insomnia, and advises against relying on sleep hygiene advice alone.12 It is worth asking your doctor about.
We help you read your data next to your days#
Our part sits alongside medical care, never instead of it: making sense of your own two weeks of data with you, and working on the sleep, nutrition, movement and stress habits behind the patterns.
A tracker gives you numbers. What it rarely gives you is the context: what you ate, when you trained, the late meeting, the second glass. Our app is built for exactly that: connect your wearable, log what you do and eat, and see the two side by side over fourteen days.
If something in your data or your days puzzles you, talk it through with our care team on a free call. It is a coaching conversation, not a medical consultation, and when a pattern points to a doctor, we will tell you so.
Quick answers12 questions this guide covers, in short
How accurate is a sleep tracker?
Reasonably accurate at telling sleep from wake across a night, less accurate at catching the moments you are awake, and inconsistent on sleep stages. Its sleep times and totals are more useful than its stage breakdown. Read the section
Can you trust the deep sleep and REM numbers on your tracker?
Not night by night. Compared with a laboratory sleep study, tracker estimates of sleep stages were inconsistent, so a low deep sleep figure is not, on its own, a reason to worry or to change anything. Read the section
Why does one bad sleep score not mean much?
Because single nights vary for many ordinary reasons, and trackers are least accurate on exactly the disrupted nights that produce low scores. A pattern across two weeks tells you far more. Read the section
Can tracking your sleep make it worse?
Yes, for some people. Sleep researchers coined the term orthosomnia for a perfectionist pursuit of ideal tracker data that ends up making insomnia worse. Read the section
What do people say about their sleep tracker?
Mostly that the score and the way they feel do not match, in both directions. These are the phrases we hear most often. Read the section
Which sleep tracker patterns are worth acting on?
Patterns tied to things you can change: bed and wake times that move around, nights regularly under seven hours, and worse nights after late caffeine, evening alcohol, bright screens or a late, heavy meal. Read the section
Why does a regular sleep schedule matter so much?
Because regularity may matter as much as length. In a large study using wrist sensors, people with the most irregular sleep timing had a higher death rate than more regular sleepers, and regularity predicted it better than sleep duration did. Read the section
How do coffee and alcohol show up in your sleep data?
Late caffeine tends to show up as later, shorter nights, and evening alcohol as falling asleep quickly followed by a restless second half. Both are easier to see across two weeks than in one night. Read the section
Do screens, late meals and exercise change your sleep?
Yes, to different degrees. Bright screens late in the evening delay sleep, late and heavy meals may fragment it, and regular exercise helps modestly; your own data shows which one matters most for you. Read the section
How do you read two weeks of sleep data?
Look at averages and spread rather than single scores: how much your bed and wake times vary, how often you sleep under seven hours, and how nights differ after the habits you noted. Then change one thing at a time. Read the section
When should you see a doctor about your sleep instead of tracking it?
When you snore loudly with gasping or pauses in your breathing, feel sleepy while driving, or have struggled to sleep for months. A tracker cannot rule out a sleep disorder, and a good score does not mean you do not have one. Read the section
Where does a functional approach fit with sleep tracking?
Alongside medical care, never instead of it. We help you read your own two weeks of data next to your days, and work with you on the sleep, nutrition, movement and stress habits behind the patterns. Read the section
- Chinoy ED, Cuellar JA, Huwa KE, et al. Performance of seven consumer sleep-tracking devices compared with polysomnography. Sleep. 2021;44(5):zsaa291. PMID : 33378539. Link
- Khosla S, Deak MC, Gault D, et al. Consumer sleep technology: an American Academy of Sleep Medicine position statement. J Clin Sleep Med. 2018;14(5):877-880. PMID : 29734997. Link
- Baron KG, Abbott S, Jao N, Manalo N, Mullen R. Orthosomnia: are some patients taking the quantified self too far? J Clin Sleep Med. 2017;13(2):351-354. PMID : 27855740. Link
- Windred DP, Burns AC, Lane JM, et al. Sleep regularity is a stronger predictor of mortality risk than sleep duration: a prospective cohort study. Sleep. 2024;47(1):zsad253. PMID : 37738616. Link
- Watson NF, Badr MS, Belenky G, et al. Recommended amount of sleep for a healthy adult: a joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society. Sleep. 2015;38(6):843-844. PMID : 26039963. Link
- Drake C, Roehrs T, Shambroom J, Roth T. Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed. J Clin Sleep Med. 2013;9(11):1195-1200. PMID : 24235903. Link
- Ebrahim IO, Shapiro CM, Williams AJ, Fenwick PB. Alcohol and sleep I: effects on normal sleep. Alcohol Clin Exp Res. 2013;37(4):539-549. PMID : 23347102. Link
- Chang AM, Aeschbach D, Duffy JF, Czeisler CA. Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. Proc Natl Acad Sci U S A. 2015;112(4):1232-1237. PMID : 25535358. Link
- Chung N, Bin YS, Cistulli PA, Chow CM. Does the proximity of meals to bedtime influence the sleep of young adults? A cross-sectional survey of university students. Int J Environ Res Public Health. 2020;17(8):2677. PMID : 32295235. Link
- St-Onge MP, Roberts A, Shechter A, Choudhury AR. Fiber and saturated fat are associated with sleep arousals and slow wave sleep. J Clin Sleep Med. 2016;12(1):19-24. PMID : 26156950. Link
- Kredlow MA, Capozzoli MC, Hearon BA, Calkins AW, Otto MW. The effects of physical activity on sleep: a meta-analytic review. J Behav Med. 2015;38(3):427-449. PMID : 25596964. Link
- Edinger JD, Arnedt JT, Bertisch SM, et al. Behavioral and psychological treatments for chronic insomnia disorder in adults: an American Academy of Sleep Medicine clinical practice guideline. J Clin Sleep Med. 2021;17(2):255-262. PMID : 33164742. Link
- Gottlieb DJ, Punjabi NM. Diagnosis and management of obstructive sleep apnea: a review. JAMA. 2020;323(14):1389-1400. PMID : 32286648. Link
- Tregear S, Reston J, Schoelles K, Phillips B. Obstructive sleep apnea and risk of motor vehicle crash: systematic review and meta-analysis. J Clin Sleep Med. 2009;5(6):573-581. PMID : 20465027. Link
Articles published by FunctionAlps are educational and written within ASCA and EMR scope. They do not replace personalised clinical advice. When findings warrant medical evaluation, we coordinate with a referring physician.