Experiment guides → Supplements → Melatonin not working
Melatonin not working? Here's how to find out why
Quick answer: there are four well-documented reasons, and none of them is "you're immune". The product: a lab analysis of 31 melatonin supplements found actual content from 83% below to 478% above the label, and a quarter contained serotonin. The timing: taken at lights-out, melatonin arrives after the signal it's supposed to send. The dose: 5–10 mg mostly buys grogginess, not sleep. The problem: melatonin is a clock signal, not a sedative — if you can't sleep because you're wired, stressed or caffeinated, there's no clock error for it to fix. The pooled trial effect is only about 7 minutes anyway, so "nothing happened" is the most common honest outcome.
"I will take a low dose, 1 hour before bed, with a fixed wake time, for 14 nights."
What the research says
What's in the bottle may not be what's on the label. Erland and Saxena analysed 31 melatonin supplements sold in Canada. Measured melatonin ranged from −83% to +478% of the labelled amount, lot-to-lot variability within a single product reached 465%, and 26% of products contained serotonin, which isn't supposed to be there at all. If your melatonin "stopped working", the bottle may simply have changed (Erland & Saxena, 2017).
The expected effect is small to begin with. Across 19 placebo-controlled trials, melatonin shortened sleep onset by about 7 minutes and added about 8 minutes of sleep. That's the average benefit in people with primary sleep disorders. If you were expecting the feeling of a sleeping pill, the disappointment is with the expectation, not the molecule (Ferracioli-Oda et al., 2013).
When the clock is the problem, low and early works. In 116 adults whose body clocks ran late, 0.5 mg of melatonin an hour before the target bedtime — combined with a fixed sleep schedule — advanced sleep onset by about 34 minutes versus placebo on the same schedule. Small dose, early timing, held-still wake time: that's the recipe the evidence supports, and it's rarely how people take it (Sletten et al., 2018).
What this evidence doesn't say
The content analysis was Canadian products in 2017; regulation differs by country, and prescription melatonin in Europe is standardised in a way over-the-counter products are not. The trials don't identify who is a "non-responder" — the four reasons above are the plausible ones, not a diagnosis. And where melatonin is prescription-only, don't change what your doctor prescribed on the strength of a web page.
Run it as an experiment
- Five nights off, with a log. Stop (if it's not prescribed) and rate sleep quality, restedness and time-to-sleep each morning. Many people discover the off nights look exactly like the on nights — which is the answer.
- Fix the three variables the trials fixed. A low dose (0.5–1 mg if available; otherwise the smallest you have), taken 60 minutes before your target bedtime, with the same wake time every day including weekends. Dim the lights in that last hour — bright light overrides the signal.
- Make the pact. "I will take a low dose, 1 hour before bed, with a fixed wake time, for 14 nights." Keep caffeine to before early afternoon; it's the most common confounder.
- Morning check-ins, plus one extra question. Did you wake groggy? Grogginess at low doses is unusual; at high doses it's common and often mistaken for "deep sleep".
- Compare, and route. Better than the off nights? Keep it, and retest later. Same? Melatonin isn't your lever — the problem probably isn't circadian, and the next experiment is caffeine timing, an earlier dinner, or a screen-free wind-down.
What to expect
If timing and dose were the issue, you'll usually notice an earlier, easier drift off within the first week, growing slightly over the second as the clock shifts. If the problem was never circadian, the on and off blocks will look the same — and that's a useful result, because it points you at the behavioural experiments that actually address a wired mind or a caffeinated afternoon.
Who should skip this
If melatonin was prescribed to you, ask your prescriber before stopping or changing the dose. Don't run this test if you're pregnant or breastfeeding, take anticoagulants, immunosuppressants or seizure medication, or have an autoimmune condition without medical advice. If your sleep has been broken for months, insomnia treatment (CBT-I) has stronger evidence than any supplement.
Frequently asked questions
Why did melatonin stop working for me?
Three candidates: the product changed (content varies several-fold between lots), your schedule changed (a later wake time or brighter evenings cancel the signal), or it never did much and the first few nights were expectation. The on/off test above distinguishes them.
Does melatonin work for everyone?
No. Meta-analyses show meaningful benefit mainly in circadian and primary sleep disorders, with no significant effect in some groups. It's a timing signal; if your problem isn't timing, the evidence doesn't predict a response.
Should I take more melatonin if it isn't working?
The research says the opposite: clock-shifting trials used 0.5 mg, and higher doses add grogginess without adding sleep. If a low dose at the right time does nothing, more of it at the wrong time won't either.
Related experiments
Sources
- Erland LA, Saxena PK. Melatonin natural health products and supplements: presence of serotonin and significant variability of melatonin content. J Clin Sleep Med. 2017;13(2):275–281. PubMed 27855744
- Ferracioli-Oda E, Qawasmi A, Bloch MH. Meta-analysis: melatonin for the treatment of primary sleep disorders. PLoS One. 2013;8(5):e63773. PubMed 23691095
- Sletten TL, Magee M, Murray JM, et al. Efficacy of melatonin with behavioural sleep-wake scheduling for delayed sleep-wake phase disorder: a double-blind, randomised clinical trial. PLoS Med. 2018;15(6):e1002587. PubMed 29912983
Track this experiment in N of 1
A few baseline nights, then the change — and your own mornings deliver the result. No streaks, no guilt. Your first experiment is free.
Downloadon the App StoreThis page is general information about published research, not medical advice. Melatonin is a prescription medicine in many countries; never stop or change a prescribed dose for an experiment without asking your doctor.