Eleven ingredients are sold for sleep. Two have evidence that holds up across independent trials, three are worth a conversation, and the rest are on the shelf because the shelf sells them.
The effect is clearest in people whose intake is genuinely low. In those with adequate intake, the trials mostly show nothing — an honest result that rarely makes it onto a label.
Shortens time to fall asleep by roughly seven minutes on average — real, replicated, and much smaller than the packaging implies. It shifts timing rather than sedating.
A handful of small trials found faster sleep onset and better next-day ratings, plausibly through a drop in core temperature. Small, consistent, not yet repeated at scale.
Trials consistently report calmer subjective ratings; objective sleep measures move much less. That gap is the whole story.
Studied for decades with stubbornly inconsistent results. Extracts vary so much between manufacturers that trials are not really testing the same substance.
A proprietary blend lists ingredients without amounts, so you cannot tell whether a studied dose is present or a trace. It is not that these fail — it is that they cannot be evaluated.
| Ingredient | Evidence | Studied dose | Typical on shelf | Watch for |
|---|---|---|---|---|
| Magnesium glycinate | Strong | 200–350 mg | 200–400 mg | Accumulates in reduced kidney function; interacts with some… |
| Melatonin | Strong | 0.5–3 mg | 5–10 mg | Grogginess at higher doses; interacts with some blood… |
| Glycine | Moderate | 3 g | 1–3 g | Trial base is small. |
| L-theanine | Moderate | 200–400 mg | 100–200 mg | Frequently underdosed in blends. |
| Valerian root | Limited | 300–600 mg | 300–900 mg | Extract varies by brand; sedation with alcohol. |
Sleep broken for months, loud snoring with daytime sleepiness, or waking gasping are signs to see a doctor rather than shop. Sleep apnoea does not respond to melatonin, and treating it as a supplement problem delays the diagnosis.
How we make money. Some links on this page are affiliate links and we may earn a commission if you buy through them. That never affects the grade — grading happens before any product is attached.
Chosen because the elemental amount is printed on the label and the batch certificate is public. Both are rarer than they should be.
Most of the shelf sells 5 and 10 mg. The trials that found an effect used a fraction of that.
Left out on purpose. A page that ranks ten products when two have evidence is a shopping page wearing a lab coat.
No. It works as a timing signal rather than a sedative, so past a low dose you are not adding sleep, you are adding next-morning grogginess.
Because the effect is largely about correcting a shortfall. If your intake is already adequate, there is nothing to correct, and the trials reflect that.
A few nights for melatonin. Three to eight weeks for magnesium. If nothing changed after a month at a studied dose, the honest conclusion is that it is not working for you.
What changed in the evidence, what got downgraded, and what quietly turned out to be nothing.