Sleep supplements: ingredients before the promise

Open a sleep category in an online smartshop and you get a very calm periodic table: melatonin, CBD, CBN, 5-HTP, hops, chamomile, lavender, magnesium. Same shelf, same soft-focus moon on the box. These ingredients aren't doing the same job, though, and the amount of research behind each one varies enormously. So let's start with the only part you can read straight off the packaging (what's in it), then work outwards to what each ingredient is known to do and what the rules say about selling it.

Here's the shape of the answer. A sleep product's label tells you what is in it and in what amount. It doesn't tell you whether that amount does anything for sleep. For most ingredients in this category the human evidence is small, mixed or missing; melatonin has the most data, and even there the average gain in studies is modest. Melatonin and CBD rules in the Netherlands also move, so we've dated what we checked below. And if poor sleep has become a regular thing, the evidence-based first step isn't a product at all (more on that near the end).

The shelf in five lines

• Keep three questions apart: what does the label say is in the product, what does research say that ingredient does for sleep, and what are sellers allowed to do with it in your country? Each has a different kind of source.
• A list of ingredients without amounts tells you very little. A list with amounts tells you something once you can compare the amount with what was studied.
• Melatonin has the most human evidence for sleep, though the average effect in a 2013 meta-analysis was small (about 7 minutes faster to fall asleep). Evidence for CBD and CBN as sleep aids in people is thin, and for the doses in these products thinner still.
• In the Netherlands, melatonin supplements are judged on the amount per dose and on whether the pack carries a health claim. CBD in food still has no EU authorisation. Neither point is a statement about any particular product.
• Persistent insomnia is a job for cognitive behavioural therapy for insomnia (CBT-I) and a GP, not for the shopping basket.

Every sleep box asks three questions at once

Marketing tends to blend them into one confident sentence. Pulling them apart makes a product page much easier to read.

• Question: What is in this product?; Where the answer comes from: The label and the seller's product page; What it can't tell you: Whether the amount on the label is what's in the pot, or whether it works
• Question: What does this ingredient do for sleep?; Where the answer comes from: Human trials and systematic reviews, ideally at a comparable amount; What it can't tell you: Whether this product's mix, extract or amount behaves like the studied one
• Question: Can it be sold here, and with which claims?; Where the answer comes from: Regulators, courts and official guidance, with a date; What it can't tell you: Whether a given product is compliant, which only the product's paperwork or the regulator can answer

A seller's page is good evidence for the first row and no evidence for the second. A trial of 300 mg of something says little about a capsule containing 12.5 mg of it. And a product being on sale tells you nothing about the third row. We'll keep coming back to this table.

Reading the box: amounts, forms and a three-capsule surprise

If you've read our guide to reading supplement labels, the same logic applies here, plus a few sleep-specific checks.

• Amount per serving, and what a serving is. Is it per capsule or per daily dose? One of the Zamnesia products described below lists its amounts "per daily serving (3 capsules)" while advising 2–3 capsules a day, so the per-capsule number is a third of what you'd first assume.
• The form of the ingredient. "Griffonia extract" is not "5-HTP". Cibdol's Stay Asleep lists 167 mg griffonia extract standardised to 30% 5-HTP, which works out at roughly 50 mg of 5-HTP (167 × 0.30). Extracts of valerian, chamomile or lavender can't be compared milligram for milligram unless the extract ratio or standardisation is given.
• Timing instructions. Melatonin products give a before-bed window; others don't. Some brands' own pages disagree with each other on this (see below).
• Claim wording. In the EU, "melatonin contributes to the reduction of time taken to fall asleep" is a set wording tied to a stated amount (details in the rules section). Wording like "supports deep sleep" isn't the same thing.
• Cautions that matter for sleep products: drowsiness and driving, pregnancy and breastfeeding, medication, and "short-term use" statements.
• Allergens and carriers: soy lecithin, alcohol (one liquid below contains 2.3% ethanol by volume), sweeteners.

Amounts matter most for melatonin, where the label-versus-pot question has been tested. A 2017 analysis of 30 commercial melatonin supplements in Canada found measured melatonin ranging from 83% below to 478% above the label claim, varying by as much as 465% between lots of the same product, and eight of them also contained serotonin (Erland & Saxena, Journal of Clinical Sleep Medicine, 2017). In 2023, lab analysis of 25 US melatonin gummy products found 22 inaccurately labelled, with measured melatonin between 74% and 347% of the label (Cohen and colleagues, JAMA). A 2025 analysis of melatonin supplements sold online in the EU (50 products, half from legal supply chains) found that 12% of those legal-chain products broke the rules, either through unauthorised health claims or by containing at least triple the melatonin permitted in Belgium (Molecules, 2025). None of these studies tested the products discussed here, so they say nothing about them. They are why "ask for batch testing" is a fair question for any melatonin product. Cibdol's own pages describe batch analysis and a certificate lookup, which is supplier information we haven't checked independently, but it is the sort of document the question is asking for.

A real sleep shelf: Cibdol's range as listed on 8 October 2026

Cibdol's sleep range is a handy example because it covers three different formulas: a melatonin-plus-CBD pair, a melatonin-free CBD-and-CBN capsule with herbs, and a liquid that adds chamomile and lavender. The table shows what its public pages said on 8 October 2026. It is supplier information about what the seller says is in the products, not independent testing and not a statement about effectiveness.

• Product (cibdol.com sleep pages, 8 Oct 2026): Fall Asleep capsules; Format: Hard capsule, liposomal; Listed per serving: 1.5 mg melatonin and 2.5 mg CBD (from hemp extract); Listed instructions: 1 capsule 30 minutes before bed on the product page; "1 to 2 hours" on the category listing; Gaps on the page: Why the two timings differ
• Product (cibdol.com sleep pages, 8 Oct 2026): Fall Asleep liquid ("Meladol"); Format: Liposomal liquid with pipette; Listed per serving: Per 1 ml: 1.5 mg melatonin, 2.5 mg cannabinoids, 0.3 mg terpenes, 0.7 mg vitamin B6; Listed instructions: 1 ml measured by pipette, 30 minutes before bed; contains soy lecithin and 2.3% ethanol; Gaps on the page:
• Product (cibdol.com sleep pages, 8 Oct 2026): Stay Asleep capsules; Format: Hard capsule; Listed per serving: 12.5 mg CBD, 12.5 mg CBN, 50 mg hops extract, 167 mg griffonia extract (30% 5-HTP); Listed instructions: 1 capsule 30 minutes before sleep; Gaps on the page: Melatonin-free; extract ratio for hops not given
• Product (cibdol.com sleep pages, 8 Oct 2026): Complete Sleep; Format: Liquid, MCT oil base; Listed per serving: CBD and CBN, chamomile extract, lavender extract; Listed instructions: 1 ml held under the tongue for 60 seconds, 30 minutes before sleep; Gaps on the page: Milligrams per serving for any ingredient

Three things stand out.

The brand's own pages don't fully agree. The category listings (both the UK and the EU versions) describe the capsule as taken "1 to 2 hours before bedtime", while the capsule product page says "30 minutes before bedtime". The pages don't explain the difference, so neither figure should be read as the answer. We've left prices out: they change, and the UK pages are priced in pounds, not euros.

Fall Asleep is a melatonin-plus-CBD pair. The Meladol page quotes the authorised EU melatonin claim and says the beneficial effect needs 1 mg of melatonin shortly before bed; both Fall Asleep products list 1.5 mg. That is above the 0.3 mg figure the Dutch regulator uses, and the rules section explains why that figure matters. Classification depends on more than the number, so the natural question for the seller is which market the pack is sold in and on what classification basis. Nothing on the pages answers it either way. Both pages say to speak to a doctor if pregnant or breastfeeding, warn about drowsiness and driving, and describe the products as short-term.

The other two lines are CBD plus companions, with sparse detail on amounts. Stay Asleep combines 12.5 mg each of CBD and CBN with hops and a 5-HTP-standardised extract. Complete Sleep lists its ingredients but not their amounts.

What studies say about each ingredient

Now the second question: what research says. Here is the whole survey at a glance, with a little detail on each ingredient underneath for anyone who wants it.

• Ingredient: Melatonin; What human studies show: Small average benefit in sleep-onset time; a Cochrane review supports use for jet lag; helps shift body-clock timing when taken at the right time; The catch: Benefit depends on timing and the situation; the European Insomnia Guideline does not recommend fast-release melatonin for insomnia
• Ingredient: CBD; What human studies show: Reviews find insufficient evidence for sleep disorders; one 4-week trial found a small change in sleep stages; The catch: The trial used 300 mg; the capsules above list 2.5 mg and 12.5 mg
• Ingredient: CBN; What human studies show: A single-night trial in 20 adults with insomnia; The catch: The primary outcome did not improve; the doses were 30 mg and 300 mg
• Ingredient: Magnesium; What human studies show: Three small trials in older adults suggest faster sleep onset; The catch: Low-quality evidence in a specific population
• Ingredient: Glycine; What human studies show: Small studies in healthy people; The catch: Not an ingredient of the Cibdol lines above
• Ingredient: L-theanine; What human studies show: A 2026 systematic review of 13 trials; The catch: Many small trials; few in clinical insomnia
• Ingredient: Valerian (and valerian with hops); What human studies show: Reviews disagree; The catch: Dutch guidance advises against it
• Ingredient: Chamomile, lavender; What human studies show: Some reviews suggest benefit; The catch: Lavender evidence is mostly inhaled essential oil, not oral extracts
• Ingredient: 5-HTP; What human studies show: One small, unblinded trial in older adults; The catch: Interaction risk with antidepressants

Melatonin: modest on average, stronger for jet lag

A 2013 meta-analysis of 19 placebo-controlled trials in 1,683 people with primary sleep disorders found melatonin shortened the time to fall asleep by about 7 minutes (95% CI 4.4 to 9.8) and increased total sleep time by about 8 minutes (Ferracioli-Oda and colleagues, PLoS ONE). The authors called the effects modest. Melatonin is better supported where the body clock is the problem: a Cochrane review of ten trials (Herxheimer & Petrie, 2002) found melatonin taken close to destination bedtime effective for jet lag, with benefit greater the more time zones were crossed and less for westward flights.

The European Insomnia Guideline 2023 (Riemann and colleagues, Journal of Sleep Research) recommends CBT-I as first-line treatment for chronic insomnia, allows prolonged-release melatonin for people aged 55 and over for up to three months, and does not recommend fast-release melatonin or phytotherapeutics for insomnia. Dutch GP guidance goes further: Thuisarts advises against valerian and says melatonin does not effectively treat sleep problems and can worsen them, with jet lag after long-distance flights as the exception.

CBD: the trial dose and the capsule dose are far apart

A 2020 systematic review of cannabinoids for sleep disorders (12 clinical studies) concluded there was insufficient evidence to support routine clinical use for any sleep disorder, given the limited research and the risk of bias (Suraev and colleagues, Sleep Medicine Reviews). A 2022 review of CBD trials for anxiety (58 studies) found anxiety-reducing effects in healthy people and certain clinical groups "without prominent effects on sleep" (Narayan and colleagues, Neuroscience & Biobehavioral Reviews). A 2025 randomised crossover trial in 125 people with insomnia gave 300 mg CBD plus terpenes for four weeks and found a small increase of 1.3 percentage points in time spent in slow-wave plus REM sleep as measured by a wrist device, with no effect on total sleep time (Wang and colleagues, Journal of Clinical Sleep Medicine).

That trial dose was 300 mg, and the Fall Asleep capsule lists 2.5 mg. The smaller amount may or may not do something; what the comparison shows is that the trial can't be borrowed as evidence for it.

CBN: the "sleepy cannabinoid" has one small trial so far

CBN is a cannabinoid that forms as THC ages, and it is often described as the "sleepy" one. A 2026 single-night crossover study of 20 adults with insomnia tested cannabinol at 30 mg and 300 mg (Lavender and colleagues, Journal of Sleep Research). Neither dose significantly reduced the primary outcome, time awake after sleep onset. At 300 mg some secondary measures improved, including time to fall asleep and subjective sleep quality. The authors called for larger and longer trials. Stay Asleep's listed 12.5 mg is below either tested dose.

Magnesium: cheap, plausible, weakly supported

A 2021 systematic review of oral magnesium for insomnia in older adults pooled three randomised trials in 151 people and found sleep onset about 17 minutes shorter than placebo (Mah & Pitre, BMC Complementary Medicine and Therapies). All three trials were at moderate to high risk of bias and the authors graded the evidence as low to very low quality. Whether it translates to younger adults isn't something that review can tell us. The magnesium forms differ in more than name, which our guide to magnesium forms covers.

Glycine: a real finding in a narrow setting

In a small study of healthy volunteers whose sleep was restricted to 75% of normal for three nights, 3 g of glycine before bed reduced next-day fatigue and improved a vigilance test compared with placebo (Bannai and colleagues, Frontiers in Neurology, 2012). It wasn't a trial in people with insomnia, and none of the Cibdol products above lists glycine.

L-theanine: encouraging and early

A 2026 systematic review found 13 trials (550 people) of L-theanine used on its own, two open-label and eleven randomised, at 50–900 mg a day (Cotter and colleagues, Nutritional Neuroscience). The authors concluded that 200–450 mg a day appeared safe and effective for supporting healthy sleep in adults, while saying more high-quality trials, including in people with clinical insomnia, were needed. The included trials are small, so treat the result as early.

Valerian, and valerian with hops: reviewers disagree

Reviewers have reached different conclusions from overlapping trials. A 2006 meta-analysis (16 studies, 1,093 patients) said valerian might improve sleep quality but flagged publication bias (Bent and colleagues, American Journal of Medicine). A 2007 review titled "safe but not effective" found most trials showed no difference from placebo (Taibi and colleagues, Sleep Medicine Reviews). A 2020 review of 60 studies concluded valerian could be a safe and effective sleep herb, while noting that results varied with extract quality (Shinjyo and colleagues, Journal of Evidence-Based Integrative Medicine). Dutch guidance (Farmacotherapeutisch Kompas, reviewed January 2025) says valerian is "hardly or not at all effective" for insomnia.

Hops appear in sleep products mostly alongside valerian, and the trials in this area tested the combination. For example, a 184-person trial of a valerian-hops extract found modest subjective improvements, with few comparisons against placebo reaching statistical significance (Morin and colleagues, Sleep, 2005). Evidence for the 50 mg hops extract in Stay Asleep is therefore indirect at best.

Chamomile and lavender: one has mixed trials, the other mostly smells

A 2024 meta-analysis of chamomile for sleep (10 studies, 772 people) found a lower Pittsburgh Sleep Quality Index score but with very high inconsistency between trials (Kazemi and colleagues, Complementary Therapies in Medicine). For lavender, recent reviews of sleep outcomes largely concern essential oil used by inhalation or aromatherapy, not an oral extract in a capsule or oil, so they are weak support for an ingestible lavender product.

5-HTP: a step on the way to melatonin

5-HTP is the step before serotonin, which is itself a step before melatonin, which is why it turns up in sleep formulas. A 2024 trial in 30 older adults in Singapore gave 100 mg of 5-HTP daily for 12 weeks against no supplement and reported better sleep scores in participants who started as poor sleepers (Sutanto and colleagues, Clinical Nutrition). It was single-blinded and used a no-treatment comparison, with just 30 participants.

The interaction side is the part to take seriously. A 2017 case report described a man taking the antidepressant sertraline who also used several dietary supplements, including a 5-HTP product, and developed serotonin syndrome that progressed to acute compartment syndrome (American Journal of Case Reports). One case can't show how common this is, but it is the kind of combination worth a pharmacist's attention. If you take an antidepressant or any other serotonergic medicine, this is a conversation for a pharmacist or doctor, not a label check. See our guide to supplements and medication.

What sellers may say and sell (rules checked on 8 October 2026)

This is the third question, and the one that changes fastest. Dutch rules are not EU rules, and none of this is advice about a particular product.

Melatonin

• Netherlands. Under the government's reply to parliamentary questions of 8 October 2019 (Kamervragen 2019-2020, nr. 297), melatonin below 0.3 mg per dose is allowed as a food supplement only if the pack carries no health claim at all ("geen enkele gezondheidsclaim"). A sleep-onset claim is a health claim, so a low dose on its own doesn't settle the question. At 0.3 mg and above, the reply says a product can be a medicine ("kan er sprake zijn van een geneesmiddel") if it also meets other criteria set by case law, such as the EU Hecht-Pharma ruling, and it notes that melatonin has long been sold as a supplement above 0.3 mg too. The health inspectorate IGJ takes the position that from a daily dose of 0.3 mg a product has a notable pharmacological effect. Medicines must be registered, and IGJ's FAQ tells consumers to check registration in the medicines database, geneesmiddeleninformatiebank.nl. On 1 April 2026, the Council of State (Raad van State, ECLI:NL:RVS:2026:1846) is reported to have upheld the 0.3 mg approach in a case about two companies (read in a law firm's summary, not the judgement). A Dutch sleep-specialist working group stated in October 2025 that an appeal by manufacturers was still pending.
• EU. There is no single EU melatonin limit for supplements; the 2025 analysis above describes melatonin as permitted in supplements "with country-specific limits". At EU level, two health claims are authorised under Regulation (EU) No 432/2012: reducing the time taken to fall asleep (tied to 1 mg per portion, taken close to bedtime) and alleviating subjective feelings of jet lag (at least 0.5 mg). Those two conditions come from secondary summaries, so check the Official Journal text before relying on them.

CBD

• EU. CBD is treated as a novel food. As of 21 September 2026, trade press reports that no CBD novel food has been authorised, though the European Food Safety Authority (EFSA) issued its first positive opinion on one application that month. The Commission and member states still have to decide.
• What EFSA has said so far. In an update dated 9 February 2026, EFSA proposed a provisional safe intake of about 2 mg a day for a 70 kg adult (0.0275 mg per kg body weight) for CBD food supplements of at least 98% purity, and said safety cannot be established for people under 25, for pregnancy or breastfeeding, or for people taking medicines. Trade press reporting on the September 2026 opinion gives 0.026 mg/kg; "about 2 mg a day" is consistent with both.
• Netherlands. The food authority NVWA's supplement page says novel foods must be assessed and approved in the EU before they are marketed, and points sellers to the EU Novel Food Catalogue and Union list (page read 8 October 2026). That page doesn't single out CBD, and a 2022 industry write-up described lenient practice at the time, so today's enforcement picture is unconfirmed. The NVWA is the place to check.

What that means in practice: the EFSA figure is about authorising a food ingredient, not about sleep, and not a ruling on any product on a shelf. It is still a reason to read the CBD amount on the label and to take the "under 25" and "medication" caveats seriously, because they apply to a lot of readers of this site.

Other ingredients here (5-HTP, herbal extracts) have their own national rules, which this page doesn't cover.

Zamnesia: two non-CBD labels, as listed on 8 October 2026

Zamnesia's sleep category is a long run of products, some of them CBD-based. These two non-CBD pages show how different a "relaxing" label and a "sleep" label can look. Everything below is supplier information.

• Dream Tablets 3-in-1 Multi Formula (Zamnesia). One tablet is listed as one daily dose: valerian extract 120 mg, lemon balm extract 75 mg, GABA 50 mg, passionflower extract 100 mg, chamomile extract 50 mg, melatonin 0.99 mg, magnesium 120 mg, L-tryptophan 80 mg, plus vitamins B3, B6, B12, D3 and folate. Instructions: 1 tablet with water 30–60 minutes before bed. It's a good example of the "small alphabet of ingredients": thirteen items, and most of the plant extracts sit at amounts that can't be compared with a trial without knowing the extract ratio. The melatonin amount is the one to read first, with the Dutch rules above in mind.
• Natural Relax (Zamnesia). Per daily serving of three capsules: valerian root extract 399 mg, rhodiola extract (1%) 300 mg, passionflower 300 mg, chamomile extract 75 mg, hop extract (0.4%) 51 mg and L-theanine 150 mg; instructions say 2–3 capsules a day. The page says it is not suitable during pregnancy or breastfeeding. It's positioned as relaxing rather than as a sleep aid.

Category placement on a shop page isn't a recommendation, and Zamnesia's other sleep products aren't covered here.

When a GP appointment is the better first step

Occasional bad nights are normal and you probably don't need an appointment. But if sleep problems persist or affect your days, a product is the wrong first move.

• The European Insomnia Guideline 2023 recommends CBT-I as the first-line treatment for chronic insomnia in adults, in person or digital, and says medicines can be offered when it isn't enough.
• The NHS says to see a doctor if changing your sleep habits hasn't helped or if insomnia is affecting your daily life in a way that makes it hard to cope. Thuisarts, the Dutch GP information site, says the same in plain terms: if the sleep advice isn't helping enough, make an appointment with your GP. The Dutch Farmacotherapeutisch Kompas calls insomnia treatment mainly non-drug.
• Sleep trouble can have causes a supplement won't touch, such as other sleep disorders (the guideline suggests assessment when something like sleep-related breathing problems is suspected), low mood, medicines, and shift patterns. A clinician can sort those; a label can't.

Medicines and pregnancy. Several ingredients here interact with medication or need caution in pregnancy: melatonin (the Dutch working group names fluvoxamine as a strong slow-down of melatonin clearance, raising levels about twelve-fold, and does not recommend melatonin in pregnancy or breastfeeding because the data are inadequate), CBD (EFSA could not establish safety with medicines or in pregnancy), and 5-HTP (antidepressants). If you take prescription medicines, check with a pharmacist before adding any of them, and have a look at the sleep routines guide for the non-product basics. Sleep products can make you drowsy; the Fall Asleep pages advise against driving or operating machinery after taking it.

Three questions to take to the shelf

If you only keep three things: the label tells you what's in the product and nothing about whether it works; the evidence for most sleep ingredients is smaller than the shelf suggests, and for CBD and CBN products it's thinner than the marketing; and the rules vary by country and keep changing, especially for melatonin and CBD. Run any product, whether it's from Cibdol, Zamnesia or anywhere else, through the same three questions, and be wary of any page, ours included, that answers them with a single sentence. We'll update this guide as product details, amounts and Dutch and EU availability get verified.

Next reading: our guide to reading supplement labels, supplements and medication, sleep routines and magnesium forms.

Written by Smart Supplements with AI assistance from the sources linked in the text. It has not been reviewed by a clinician.

Last updated: 12 October 2026

This article is for informational purposes only and is not intended as medical advice. Always consult a healthcare professional before starting any new supplement, especially if you take prescription medication.