The CDC estimates that roughly a third of US adults don't get the sleep they need on a regular basis.1 The supplement industry's response has mostly been to sell "sleep blends" — five or six ingredients under one undisclosed number. Here's what actually has trial evidence behind it, separated from what's just packaging — including what deep sleep is actually doing in your body, and the one lever most of those blends never mention.
Light exposure and timing, not a pill, is the strongest lever
The single best-evidenced input into sleep quality isn't a supplement at all — it's light. Reviews of circadian research consistently find that bright light exposure earlier in the day and reduced light (particularly blue-spectrum light) in the hours before bed measurably shift sleep timing and quality, by directly influencing the body's circadian pacemaker.2 Consistency of sleep and wake times matters as much as any single ingredient does. This isn't a satisfying answer if you're looking for something to buy, but it's the most reliably evidenced one.
What deep sleep actually does
"Deep sleep" isn't marketing shorthand — it's a specific, measurable stage. Sleep researchers stage a night's sleep using polysomnography (simultaneous EEG, eye-movement, and muscle-tone recording), which sorts sleep into light stages (N1, N2), deep or "slow-wave" sleep (N3, marked by large, synchronized delta brainwaves), and REM. Deep sleep is concentrated in the first third to half of the night and shrinks with each successive sleep cycle, which is part of why a short night disproportionately costs you deep sleep rather than REM.
What actually happens during N3 is where it gets interesting, and where most sleep marketing quietly stops being able to cite anything. The clearest mechanistic evidence comes from a 2013 study in mice published in Science: using real-time imaging of the brain's extracellular space, researchers found that the interstitial space between brain cells expanded by roughly 60% during sleep (and under anesthesia) compared with wakefulness, and that this expansion nearly doubled the rate at which the brain cleared metabolic byproducts — including beta-amyloid, a protein implicated in Alzheimer's disease — through what's now called the glymphatic system.5 That's a mouse study, not a human trial, and it does not show that sleeping more prevents Alzheimer's in people — we're not going to claim that. What it does establish, mechanistically, is that sleep isn't simply "rest" for the brain; it appears to be an active housekeeping window, and slow-wave sleep specifically is when it's thought to run.
Temperature: the lever most sleep blends skip
If light and timing set the strongest circadian input, temperature is probably the strongest overlooked one. Your core body temperature naturally drops by roughly half a degree Celsius in the hours before and during sleep, and that decline is tightly linked to how quickly you fall asleep and how much slow-wave sleep you get.
A 2024 three-site, blinded, crossover study in 72 healthy adults (young men, middle-aged men, and post-menopausal women, tested in Turin, Berlin, and Chicago) put this to a direct test. Participants slept a full night on a mattress engineered for much higher conductive heat loss, and a separate night on a standard mattress, with full polysomnography recording both nights. On the high-heat-loss mattress, slow-wave sleep increased by about 1.9 percentage points as a share of total sleep time, and average heart rate during the night dropped by roughly 2.4 beats per minute — both statistically significant, and consistent across all three very different participant groups.6 The effect tracked with a bigger gap between core and skin temperature, and it showed up most clearly in the second half of the night, which lines up with when deep sleep is normally winding down.
This is consistent with older, smaller studies on pre-sleep warm baths: a warm bath or shower 1–2 hours before bed causes skin blood vessels to dilate, which paradoxically speeds up the core temperature drop once you're out of the water and into a cooler room — the same physiological direction the mattress study measured directly, just a different route to it. None of this is a supplement, and none of it is being sold to you here; it's a genuinely evidence-backed, free lever that most "sleep stack" marketing skips over entirely because there's no ingredient to put on a label.
Melatonin: real evidence, and where it sits in our range today
Melatonin is one of the few sleep ingredients with genuinely strong trial support. A 2024 dose-response meta-analysis of randomized controlled trials found melatonin progressively reduces the time it takes to fall asleep, with the effect peaking around a 4mg dose — and, notably, found that taking it roughly 3 hours before the target bedtime outperformed the more common practice of taking a smaller dose 30 minutes before bed.3
Melatonin is in two products in our range. Sleep Formula lists it as one of several individually-dosed ingredients — 2mg, disclosed on the label, not folded into a blend total — alongside valerian, chamomile, GABA, L-tryptophan, lemon balm and passion flower, each with its own stated dose. Sleep Support also contains melatonin, currently without an individually disclosed dose; that's part of the same disclosure work underway for GABA, 5-HTP and St. John's Wort in that formula. Neither product is dosed at the roughly 4mg peak-effect level from the meta-analysis above — Sleep Formula's 2mg is a well-tolerated starting dose, not a claim that it matches the highest-effect research dose.
| Intervention | Evidence strength | What the trials actually show |
|---|---|---|
| Light exposure & timing | Strong | Consistently shifts circadian timing and sleep quality across multiple reviews; no single trial number, but the most replicated finding in the field. |
| Melatonin (4mg, ~3h before bed) | Strong | Dose-response meta-analysis of RCTs; reduces time to fall asleep, effect peaks around 4mg. |
| Accelerating core temperature drop (cooling mattress, warm bath) | Moderate | Multi-site RCT: +1.9 percentage points of slow-wave sleep, -2.4 bpm heart rate overnight. |
| Magnesium | Modest | Meta-analysis of 3 small trials in older adults; ~17 minutes faster sleep onset, authors rate evidence limited. |
Magnesium's more modest, real role
Magnesium is heavily marketed for sleep. The honest read: a meta-analysis of three trials in older adults found it reduced time to fall asleep by around 17 minutes versus placebo — real, but modest, and the authors themselves rated the evidence limited.4 We cover this in more depth in our complete magnesium guide.
Bottom line
The best-evidenced sleep interventions, roughly in order of how strong the evidence actually is: consistent light exposure and sleep timing, properly dosed and timed melatonin (already in Sleep Formula at a disclosed 2mg, and in Sleep Support at a dose we're still working to disclose), helping your core body temperature drop faster at night through a cooler room or a warm pre-bed bath, and a modest assist from magnesium if you're running low. Deep sleep itself looks less like passive downtime and more like a housekeeping shift — measurable by EEG, concentrated early in the night, and apparently doing real metabolic clearance work, at least in the animal studies that have looked closely. None of that requires a six-ingredient blend.
This article is for informational purposes only and has not been evaluated by the FDA. It is not intended to diagnose, treat, cure, or prevent any disease. Speak with a healthcare provider before starting any new supplement, especially if you take medication or have an existing health condition.
References
- CDC: 1 in 3 adults don't get enough sleep. cdc.gov
- Recommendations for daytime, evening, and nighttime indoor light exposure to best support physiology, sleep, and wakefulness in healthy adults. PLOS Biology. journals.plos.org
- Optimizing the Time and Dose of Melatonin as a Sleep-Promoting Drug: Systematic Review of RCTs and Dose-Response Meta-Analysis. Journal of Pineal Research, 2024. pubmed.ncbi.nlm.nih.gov
- Oral magnesium supplementation for insomnia in older adults: systematic review & meta-analysis. ncbi.nlm.nih.gov
- Xie L, et al. Sleep Drives Metabolite Clearance from the Adult Brain. Science, 2013. pmc.ncbi.nlm.nih.gov
- Enhanced conductive body heat loss during sleep increases slow-wave sleep and calms the heart. Scientific Reports, 2024. nature.com
Related reading: Sleep Hygiene Tips vs Real Treatment: What the Evidence Hierarchy Actually Looks Like, Caffeine and Sleep: What the Dose-and-Timing Trial Actually Found, CBT-I: The Sleep Treatment With Better Evidence Than Any Supplement and Sleep Difficulties in Autism: What the Melatonin Research Shows.
Proco Magnesium Glycinate is dosed at 275mg elemental magnesium, third-party tested every batch. Shop Magnesium →