Skip to content
What to Drink Before Bed: Sleep Drinks Ranked by Evidence
Only two bedtime drinks have placebo-controlled trials as drinks: tart cherry juice, three crossover trials of 20, 15 and 8 people with more total sleep in two and less time awake in the third, and chamomile, where the one double-blind trial in insomnia (34 adults, 540 mg of extract a day) found no change on any sleep-diary measure and an unblinded tea trial in 80 postnatal women found fewer sleep complaints at two weeks and none at four. Warm milk has a review of 14 mixed studies, lemon balm tea has no sleep trial, and sleep powders are judged by their ingredients, of which magnesium has three placebo-controlled trials of its own. The one drink the evidence and the NHS advise against is alcohol. The sleep supplements evidence page grades the ingredients one by one.
Does what you drink before bed change how you sleep?
Only a placebo drink separates a ritual from its ingredients, so the table ranks drinks by the trials behind them.
| Drink | Trial dose | Trials, people and what was found |
|---|---|---|
| Tart cherry juice | 240 mL of juice twice a day for 2 weeks; concentrate for 7 days | 3 placebo-controlled crossover trials: total sleep and efficiency up (20 adults); minutes awake after sleep onset down (15); 84 more minutes by polysomnography (8) |
| Chamomile | 270 mg of extract twice a day for 28 days; tea for 2 weeks | Extract, 34 adults with insomnia, double-blind: no significant change on any sleep-diary measure. Tea, 80 postnatal women, unblinded: fewer sleep complaints at 2 weeks, none at 4. 12-trial review: pooled sleep quality SMD -0.73 |
| Sleep powder in water | Glycine 3 g; tryptophan 1 g or more (GB supplement cap 220 mg) | No trial of a finished powder. Magnesium, 3 trials, 151 older adults: sleep onset about 17 minutes sooner, low-quality evidence. Tryptophan: less time awake at 1 g or more |
| Warm milk | Varied; a glass supplies a small fraction of a gram of tryptophan | Review of 14 studies, 8 randomised: better sleep quality within a balanced diet including dairy; mixed results |
| Lemon balm, rooibos, other herbal teas | Not established | No placebo-controlled sleep trial as a tea; lemon balm trials used extract and measured calmness; rooibos contains no caffeine |
| Water | Timing, not volume | Review of 8 bladder studies: cutting caffeine reduced nocturia; cutting fluid as well did not |
| Alcohol (the nightcap) | REM disrupted from about 2 standard drinks (0.5 g/kg) | Review of all healthy-volunteer studies and a 27-study meta-analysis: falling asleep shortened only at about 5 drinks; second half of the night disrupted at every dose |
Does tart cherry juice help you sleep?
In the UK, 20 adults took a tart cherry juice concentrate for seven days: urinary melatonin rose and time in bed, total sleep time and sleep efficiency increased.4 Fifteen older adults with chronic insomnia drank a tart cherry juice blend for two weeks: minutes awake after sleep onset fell against placebo, the other sleep measures did not, and the authors judged the effects "considerably less" than those of sleeping tablets or cognitive behavioural therapy.5 In the smallest, eight people over 50 drank 240 mL of juice twice a day for two weeks and slept 84 minutes longer by polysomnography.6 Montmorency cherries carry 13.46 nanograms of melatonin per gram,7 so the juice is not a melatonin dose. Do sour cherries help you sleep? has the trials and the 30 mL serving; the tart cherry warnings page covers sugars, loose stools and potassium.
Does chamomile tea help you sleep?
The only double-blind, placebo-controlled chamomile trial in diagnosed insomnia gave 34 adults 270 mg of standardised extract twice a day for 28 days and found no significant difference from placebo on any sleep-diary measure, from total sleep time to number of awakenings.1 The tea trial is different in kind: 80 women with poor sleep after childbirth drank chamomile tea for two weeks or received usual care with no placebo drink; their sleep-inefficiency and depression scores were lower at two weeks and no different at four.3 In a review of 12 randomised trials, pooled sleep quality was better on chamomile (standardised mean difference -0.73), while the one insomnia trial showed no change.2 A cup cannot be converted into milligrams of extract. Does chamomile help you sleep? has the daisy-family allergy and the anticoagulant question.
Do lemon balm, rooibos and other herbal teas help you sleep?
Lemon balm's trials used extract rather than tea and mostly measured calmness; none has measured sleep after the tea. Rooibos contains no caffeine and has no sleep trial; the rooibos page covers what it has been studied for. Any caffeine-free herbal tea serves as the warm ritual.
Does warm milk help you sleep, and what about its tryptophan?
A 2020 systematic review found 14 studies between 1972 and 2019, eight of them randomised, and concluded that a well-balanced diet including milk and dairy improved sleep quality, with mixed results across studies.8 The tryptophan explanation is arithmetic: a glass of milk supplies a small fraction of a gram of tryptophan, while the pooled tryptophan trials found less time awake after sleep onset at 1 g or more a day (28.9 minutes against 56.6 minutes below 1 g) and no effect on the other sleep measures.9
What is in a sleep powder you mix with water?
Sleep powders are sachets stirred into water before bed, usually magnesium with amino acids and botanical extracts. No finished powder has a trial of its own, so the evidence is the ingredients'. Magnesium: three trials in 151 older adults brought sleep onset 17.36 minutes sooner than placebo, on evidence rated low to very low quality;11 magnesium contributes to a reduction of tiredness and fatigue and to normal psychological function.12 Glycine: the sleep studies used 3 g in water before bed; the glycine page has them. L-tryptophan: in Great Britain a food supplement may recommend no more than 220 mg a day;10 the pooled trials found their effect at 1 g or more.9 L-theanine: trial doses are stated as milligrams of theanine, so a label listing a green tea leaf extract "providing L-theanine" without a figure does not state the theanine amount; see the L-theanine explainer. Their labels carry cautions: not with sedatives or antidepressants (tryptophan and theanine), not in pregnancy or breastfeeding without advice, and ask a pharmacist if you take any medicine. How to choose a melatonin-free sleep supplement reads a label line by line.
Does a nightcap help you sleep?
No. A review of every study in healthy volunteers found that at all doses alcohol shortens the time to fall asleep, consolidates the first half of the night, increases disruption in the second half and delays the first REM period.13 A 2025 meta-analysis of 27 studies found REM sleep was disrupted from about two standard drinks (0.5 g of alcohol per kilogram), while the time to fall asleep and to reach deep sleep shortened only at about five drinks (0.85 g/kg).14 That second-half disruption is the 3am waking many drinkers describe (why do I wake up at 3am). The NHS insomnia advice: "do not smoke or drink alcohol, tea or coffee at least 6 hours before going to bed".15 Needing a drink to get to sleep is a reason to talk to a GP (when to see a GP about sleep).
What about caffeine in tea, green tea, matcha and decaf?
Across 24 studies, caffeine cut total sleep by 45 minutes and sleep efficiency by 7%, added 9 minutes to falling asleep and 12 minutes awake after sleep onset, and took 11.4 minutes off deep sleep; by their arithmetic a 107 mg coffee should be finished at least 8.8 hours before bed.16 NHS estimates: 100 mg in a cup of instant coffee, 75 mg in a cup of black or green tea; in pregnancy, no more than 200 mg a day.17 Green tea and matcha are tea, so an evening cup counts; decaf contains less caffeine, not none.
Will a bedtime drink make me get up in the night?
Timing matters more than volume: in a review of eight studies in adults with an overactive bladder, cutting caffeine reduced nocturia and incontinence episodes, while restricting fluid as well as caffeine did not.18 The NHS line: "If you have to urinate frequently during the night (nocturia), try drinking less in the hours before you go to bed."19 One cup an hour to 90 minutes before bed, with caffeine and alcohol moved much earlier, fits both; new or worsening night-time weeing is a GP conversation.
Frequently asked questions
What can I drink to get deep sleep?
Nothing in the drink trials added deep (slow-wave) sleep. Caffeine took 11.4 minutes off it across 24 studies,16 and alcohol increases first-half slow-wave sleep only to fragment the second half,13 so the drinks that protect deep sleep are the ones you leave out of the evening.
What drink makes you fall asleep quickly?
Only magnesium has a pooled figure for falling asleep: about 17 minutes sooner in 151 older adults, on low-quality evidence.11 Melatonin is available on prescription only in the UK,20 so no drink or supplement sold here contains it; see the melatonin explainer and does magnesium help you sleep?
Does the sleepy girl mocktail work?
It is tart cherry juice, a magnesium powder and sparkling water. No trial has tested the combination; cherry's three small trials and magnesium's three are above, and the melatonin in the cherries is nanograms, not a dose.7
Sources
- Zick SM et al. Chamomile extract for chronic primary insomnia: placebo-controlled pilot study. 2011. PubMed 21939549
- Hieu TH et al. Chamomile for anxiety, insomnia and sleep quality: meta-analysis. 2019. PubMed 31006899
- Chang SM, Chen CH. Chamomile tea, sleep quality and depression in postnatal women. 2016. PubMed 26483209
- Howatson G et al. Tart cherry juice, melatonin levels and sleep quality. 2012. PubMed 22038497
- Pigeon WR et al. Tart cherry juice beverage and sleep in older adults with insomnia. 2010. PubMed 20438325
- Losso JN et al. Tart cherry juice for insomnia: pilot study. 2018. PubMed 28901958
- Burkhardt S et al. Melatonin in Montmorency and Balaton tart cherries. 2001. PubMed 11600041
- Komada Y et al. Milk and dairy products and sleep: systematic review. 2020. PubMed 33339284
- Sutanto CN et al. Tryptophan supplementation and sleep quality: meta-analysis. 2022. PubMed 33942088
- The Tryptophan in Food (England) Regulations 2005, regulation 5. legislation.gov.uk
- Mah J, Pitre T. Oral magnesium for insomnia in older adults: meta-analysis. 2021. PubMed 33865376
- Great Britain nutrition and health claims (NHC) register. gov.uk
- Ebrahim IO et al. Alcohol and sleep I: effects on normal sleep. 2013. PubMed 23347102
- Gardiner C et al. Alcohol and subsequent sleep in healthy adults: meta-analysis. 2025. PubMed 39631226
- NHS. Insomnia. nhs.uk/conditions/insomnia
- Gardiner C et al. Caffeine and subsequent sleep: meta-analysis. 2023. PubMed 36870101
- NHS. Foods to avoid in pregnancy (caffeine). nhs.uk/pregnancy/keeping-well/foods-to-avoid
- Park J et al. Fluid and caffeine modification in overactive bladder: systematic review. 2023. PubMed 37015722
- NHS. Urinary incontinence. nhs.uk/conditions/urinary-incontinence
- NHS. About melatonin. nhs.uk/medicines/melatonin/about-melatonin
