🌿 What melatonin actually is, and is not

Ten milligrams was the wrong lever

August 17, 2026 • Estimated Reading Time: 5 minutes

The bottle kept getting bigger

I started at one milligram, which felt sensible.

When that did nothing, I moved to three, because three is what the shelf offers next. Then five. Then, somewhere around the second week of a bad stretch, I was standing in a drugstore aisle holding a bottle of ten milligram gummies and thinking that this was a reasonable response to a problem.

It was not. I slept about the same, woke at three most nights, and spent the following mornings feeling like I had been assembled out of spare parts.

What I had been doing was turning the only dial the label mentions. Nobody prints the other one on the bottle, and the other one is the one that matters.

In partnership with

The professional secret to dough is out

Ever tried to make dough in your regular mixer?

The results aren’t pretty: Clumpy texture. Stodgy bread. Overheated motors.

So, how do the pros get it right?

Its secret is the same dough-making technology that professional bakers have used for decades.

Until now, most spiral mixers have been industrial-sized machines with industrial-sized price tags.

They also only made dough. And that gets a no from us.

To put the same professional tech into the hands of home bakers, we rebuilt the spiral mixer — creating a mighty little machine that can whip, beat and knead to professional high standards.

*Ad

It was never a sleeping pill

Here is the thing I had wrong at a very basic level.

Melatonin is not a sedative. It does not switch off your brain the way an actual sleep medication does. It is a message. Your pineal gland releases it when the light goes down, and the rest of your body reads it as a timestamp: the night has started, begin the shutdown.

Blood pressure drifts down. Core temperature drops. Alertness fades on a schedule that was set hours earlier.

So a pill of it is not a knockout punch. It is a forged timestamp. You are telling your body that dusk arrived, and the body, which is not stupid, weighs that message against everything else it knows. Which is why a bigger message does not automatically buy you more sleep, and why the timing of the message does an enormous amount of work.

Immediate release melatonin also clears fast, with a half life under an hour. Swallow it thirty minutes before bed and the signal peaks and collapses while you are still in the first stretch of the night. Then it is gone, at three in the morning, which is roughly when I kept opening my eyes.

The meta-analysis that reorganized my thinking

A team pooled 26 double blind randomized trials, 1,689 observations and did something most reviews skip. Instead of asking whether melatonin works, they asked which variables in the schedule predict whether it works.

Two things came out.

The dose response curve rises and then flattens, peaking around 4 mg a day. Past that you are not buying anything. My ten milligram gummies were on the wrong side of the hill.

The second finding is the one I keep repeating to people. The gap between swallowing the pill and going to bed was a significant predictor of how fast people fell asleep. Their conclusion, stated plainly, is that taking it about three hours before your intended bedtime, rather than the customary thirty minutes, along with a moderate dose, appears to work better than the schedule almost everyone actually uses.

Three hours. Not thirty minutes. That is a completely different instruction than the one printed on every bottle I have ever bought.

1,000,000 Men Use This. You Probably Know One.

Most men don't think about their skin until eye bags, dark spots, and wrinkles show up at once. Particle Face Cream is a groundbreaking men's formula that fights all three, while also restoring firmness and hydrating deeply. No complicated routine. Trusted by over a million men. 30-day money-back guarantee. 20% off with code BH20.

*Ad

The camp that says go smaller

Now the complication, because there is always one, and skipping it would be dishonest.

There is a serious body of work arguing the opposite of "more." A team at MIT ran a double blind study with three doses, 0.1, 0.3, and 3.0 mg, given half an hour before bed, in people whose own nighttime melatonin output had fallen off and whose sleep had gotten patchy.

The 0.3 mg dose, which is the amount that puts your blood level roughly where a healthy night would put it on its own, restored sleep efficiency. It worked mostly through the middle third of the night, which is exactly the stretch I was losing.

The 3 mg dose helped too, but it dropped body temperature and left melatonin circulating into the following day. The signal that says "it is night" was still being broadcast at lunchtime.

And people in the same study whose sleep was already fine got nothing from any dose. Which tracks. You cannot deliver a message the body has already received.

How both answers can be right

These two results look like a contradiction and are not, once you notice they are answering different questions.

The small physiological dose is replacement. It refills a signal that has thinned out, at the hour that signal belongs, and it goes away by morning like the real thing does.

The larger dose taken hours early is doing a different job. It is shifting the clock itself, dragging your whole internal schedule earlier. Researchers mapped this directly by giving people melatonin at different hours and measuring how far their body clock moved. In that work, the biggest advances came from 0.5 mg taken in the afternoon, several hours before the body's own release would have started, not at bedtime.

So the question is not "how much." It is "which problem." If your clock runs late and you lie there wide awake at midnight, you want the early, clock shifting version. If you fall asleep fine and come apart at three, you want the small replacement dose at bedtime. Same molecule, two entirely different instructions, one bottle that mentions neither.

Wake Up Smarter About AI.

The president of OpenAI. The CEO of Google. The founder of LinkedIn. We talk to the people building AI. Every morning, we put what they told us in your inbox.

What they're actually worried about. What they're betting on. What's coming next. Delivered in five minutes, before your first meeting.

With one email, and only five minutes, you can stay ahead of 99% of the world.

*Ad

The label is doing its best

There is a practical wrinkle underneath all of this that deserves its own paragraph.

Researchers bought 31 melatonin products off ordinary store shelves and measured what was in them. More than seven in ten missed their own label by more than 10 percent. The actual content ranged from 83 percent below the stated amount to 478 percent above it. Within a single product, one lot differed from another by 465 percent.

And serotonin, which is not supposed to be in there at all, showed up in about a quarter of the samples.

That study is from 2017 and covers one country's shelves, so do not treat it as a description of every bottle sold. But it does mean that fine tuning between 0.3 and 0.5 mg is a slightly comic exercise when the pill itself may not know what it contains. Buy something that carries a third party testing mark, and hold your precision loosely.

The part that deflates me

I owe you this one.

In its 2017 guideline on insomnia drugs, the American Academy of Sleep Medicine looked at the evidence and recommended against melatonin for chronic insomnia in adults. They graded that as a weak recommendation, and weak is doing real work in that sentence, but the direction is clear enough.

Read the fine print, though, and their judgment rests largely on trials of roughly 2 mg taken shortly before bed. Which is precisely the schedule the 2024 analysis identifies as the poorly chosen one. That does not overturn the guideline, and I am not going to pretend it does. It does suggest the guideline may have graded the wrong protocol, and I would like to see the question asked again with trials built around the clock instead of the milligrams.

Either way, the honest summary is that melatonin is a modest tool. It is not the reason anyone sleeps badly, and it will not repair a bedroom that is too warm, a phone at midnight, or a glass of wine at nine. Those cost nothing and outperform the pill.

Please read this bit

Melatonin interacts with blood thinners, blood pressure medication, diabetes medication, immunosuppressants, and some antiseizure drugs. If you take anything from those categories, this is a conversation with your pharmacist, not a solo experiment. Your pharmacist will do it in three minutes and will not charge you.

Daytime grogginess, vivid dreams, and headaches are the common complaints, and they usually point to a dose that is too large or too late rather than to melatonin being wrong for you.

Do not drive after taking it, including the afternoon timing described above. And if the problem is that you wake gasping or your partner reports that you quit breathing, no supplement addresses that and a sleep study will.

Nothing here diagnoses, treats, or cures anything. If you are pregnant, nursing, on medication, or managing a health condition, run any change past the person who knows your chart. This is one friend reading the papers out loud.

What my evenings look like

The ten milligram gummies went in the bin, which felt good in a small petty way.

I use 0.5 mg most nights, taken around seven thirty rather than at eleven, and I dim the lamps at the same time so the pill and the room are telling the same story. That combination did more in a week than four escalating months of bigger bottles.

Waking at three has mostly quit happening. I am one person with one set of eyes on one clock, which is worth exactly what you would expect, so run your own version. Keep a note of your lights out time and your wake ups for two weeks before you change anything, then change one variable and watch two more weeks.

And if you are already holding a bigger bottle, try moving the clock before you move the milligrams. The clock is free.