8 min read · Published 20 September 2026
Most advice about sleep arrives as a list of rules, and most people abandon it within a week. Not because the rules are wrong, but because eleven things to do differently at 10pm is not a routine — it is a second job, arriving at the exact hour you have the least willpower left.
A bedtime routine works on a different principle. It is a short, repeatable sequence your body learns to read as a signal, so that falling asleep becomes something that happens to you rather than something you attempt. Sleep research is unusual in that the least glamorous interventions have by far the strongest evidence behind them, and nearly all of them are about consistency rather than effort.
This article walks through how to build that routine so it survives a real week — including the travel, the late dinners and the nights it does not go to plan. Toward the end it covers where melatonin genuinely fits, where it does not, and how a product like our SH Sleep Strip sits within a routine rather than in place of one.
Why the routine matters more than any single thing you take
Your body runs on an internal clock that is roughly, but not exactly, twenty-four hours long. It gets reset every day by external cues — light above all, then meal timing, activity, temperature and social schedule. When those cues arrive at consistent times, the clock stays locked in and the sleep pressure building through your day peaks when you want it to.
When the cues move around, the clock drifts. A 7am alarm on weekdays and an 11am lie-in on Saturday is, from your body's point of view, a modest time-zone change every Friday and another one every Monday. Around a third of US adults report routinely getting less sleep than they need, and for a large share of them the problem is not that they cannot sleep — it is that their schedule has stopped making a coherent request.
This is why regularity keeps appearing in the research as a factor in its own right. Studies looking at consistency of sleep timing, separately from total hours, keep finding that it tracks with health outcomes. That association is worth taking seriously, though observational work of this kind cannot prove which way the causation runs. The practical implication is the same either way: an average of seven hours slept at wildly different times is not the same thing as seven hours slept on a stable schedule.
Step one: fix the wake time, not the bedtime
This is the single highest-value change, and it is counterintuitive enough that most people skip it.
Bedtime is a poor anchor because you cannot reliably control when you fall asleep. Wake time you can control, because an alarm controls it. Holding a consistent wake time — including at weekends, within about an hour — does three useful things at once. It fixes the moment morning light reaches your eyes, it makes the following night's sleep pressure predictable, and it heads off the Sunday-night problem, where you go to bed early after a late Saturday and lie there wide awake.
Pick a wake time you can hold on a bad week, not an ideal one. Then leave it alone for a fortnight before judging it. Bedtime drifts toward the right place on its own as sleep pressure rebuilds; forcing it earlier before that happens mostly produces time spent awake in bed, which is the one habit you least want to teach.
Step two: build a wind-down hour you will actually do
The hour before bed is where most routines are won or lost. The aim is not relaxation for its own sake. It is to give your nervous system a consistent, recognizable runway, so the transition into sleep is gradual rather than abrupt.
Three principles make a wind-down stick:
- Keep it short. Thirty to sixty minutes. A routine that demands ninety minutes gets skipped on the nights you need it most.
- Keep the order fixed. The sequence is doing much of the work. The same three or four actions in the same order becomes a cue. The same actions in a random order does not.
- Make it portable. If your routine depends on a particular bath or a particular chair, it collapses the first night you spend in a hotel. Choose things you can do anywhere.
A workable version: dim the main lights, put the phone on charge in another room, do one quiet thing you enjoy for twenty to thirty minutes, then the same short bathroom sequence, then bed. That is all of it. The content matters far less than the repetition.
The four inputs worth getting right
Light. The most powerful cue by a wide margin. Get bright light — ideally outdoors — within an hour of waking, and let the evening get genuinely dim. Bright overhead lighting at 10pm is a stronger signal to your body clock than most people realize, and a far bigger factor than blue light from a screen held at normal distance.
Caffeine. Caffeine has a half-life of roughly five to six hours in most adults, so a 3pm coffee still has meaningful amounts circulating at bedtime. It does not always stop you falling asleep. It reliably makes the sleep shallower. A cut-off eight to ten hours before your target bedtime is a sensible default, earlier if you are sensitive.
Alcohol. The most commonly misunderstood one. A drink makes you fall asleep faster, then fragments the second half of the night as it is metabolized, suppressing REM sleep and pulling you awake early. It is a sedative, not a sleep aid, and those are not the same thing.
Temperature. Core body temperature has to fall for sleep to begin. A cool bedroom — most guidance lands somewhere around 65–68°F — supports that drop. It is also why a warm shower an hour before bed helps: the rebound cooling afterward works with the process rather than against it.
What to do when you are awake at 2am
Every routine has bad nights, and how you handle them decides whether they become a pattern.
The rule from cognitive behavioral therapy for insomnia — the intervention with the strongest evidence base for chronic sleep difficulty, stronger than any supplement — is deceptively simple. If you have been awake for roughly twenty minutes and are starting to get frustrated, get up. Go to another room, keep the lights low, do something undemanding, and come back when you feel sleepy.
The logic is that lying awake in bed teaches your brain to associate the bed with being awake. Breaking that association is worth more than the sleep you lose by getting up. It feels wrong the first few times and works anyway.
One more thing worth saying plainly. If you have been sleeping badly for more than three months, or you snore heavily and wake unrefreshed, that is a conversation to have with a doctor rather than a routine to optimize. Insomnia and sleep apnea both have real treatments, and no supplement substitutes for either.
Where melatonin actually fits — and where it does not
Melatonin is the most misunderstood product on the shelf, mostly because it is sold beside sedatives and assumed to work like one.
It does not. Melatonin is a hormone your own body releases as evening light fades, and its job is to tell the rest of your system what time it is. It is a timing signal, not a switch that powers you down. That single distinction tells you almost everything about when it is likely to help.
The evidence is strongest for circadian problems: jet lag, shift patterns, and a sleep phase that has drifted later than the life around it. In those cases you are asking melatonin to do exactly what it does — move the clock.
It is weaker for general insomnia in someone whose body clock is broadly fine. Effects in that group tend to be modest, and the American Academy of Sleep Medicine's clinical guidance has not recommended melatonin as a treatment for chronic insomnia in adults. That is genuinely useful to know before you buy, and you will rarely see it on a label.
Two further points get lost in the marketing. First, more is not better: the doses used in circadian research are typically well below the amounts sold in large US retail bottles, and higher doses mainly raise the odds of grogginess the next morning. Second, melatonin works best when the rest of the routine is doing its job. A timing signal sent into a bedroom with the overhead light on and a phone in your hand is competing with a much louder signal saying the opposite.
What melatonin will not do
- It will not sedate you through a racing mind or an unresolved stressful day.
- It will not compensate for a schedule that shifts by three hours at weekends.
- It will not fix sleep that is being broken by alcohol, untreated apnea or chronic pain.
Where the SH Sleep Strip fits
With that framing, here is the honest version of where our own product sits.
The SH Sleep Strip 3-Pack ($62.90) is a fast-dissolving oral strip carrying 1 mg of melatonin alongside valerian root, lavender, chamomile and hibiscus. It is built to sit at the end of a wind-down routine: one strip on the tongue 15 to 30 minutes before bed, dissolving in seconds, no water and nothing to swallow.
Three things make it a reasonable fit for the routine described above.
- The dose is deliberately low. 1 mg sits within the range circadian research actually uses, rather than the 5 or 10 mg that has become the US retail default. For a timing signal, that is a feature rather than a compromise.
- It removes the friction. The most common reason people abandon an evening supplement is not disbelief. It is that fetching water and swallowing a capsule breaks the wind-down they just spent forty minutes building. A strip does not.
- It travels. Individually sealed strips go in a jacket pocket and survive a carry-on, which matters precisely in the situations — travel, time-zone shifts, irregular schedules — where melatonin has the best evidence behind it.
The 3-Pack is a 90-day supply at our lowest cost per strip. If you have not used the strips before, though, the sensible order is the other way round: start with the single 30-day pack ($24.90), keep the rest of your routine unchanged for two to three weeks so you can actually tell what is doing what, and only then commit to a longer supply. Use code WELCOME15 on your first order.
A few practical notes, because they matter more than the marketing. Take it only when you have a full seven to eight hours available for sleep. Do not drive or operate machinery afterward, and do not combine it with alcohol. One strip a day, no more. It is intended for adults, and it is not suitable if you are pregnant, nursing or under 18.
And if your difficulty is a racing mind rather than drifted timing, melatonin may be the wrong tool entirely. Our piece on magnesium glycinate covers that case, and the non-melatonin options are worth reading before you buy anything at all.
A seven-day way to put this together
Change one thing at a time. The order below is deliberate — each step makes the next one easier.
- Days 1–2. Fix the wake time. Nothing else. Get outside within an hour of waking.
- Days 3–4. Add the caffeine cut-off, and start dimming the lights an hour before bed.
- Days 5–6. Build the short wind-down sequence, and keep the order identical both nights.
- Day 7 onward. Hold it steady for two more weeks before adding anything else — including a supplement. If you do add melatonin, add it last, so you can tell whether it is contributing.
Two to three weeks is the honest minimum before judging any of this. Sleep is noisy night to night, and one bad night after a change tells you nothing at all.
The routine itself is unremarkable, and that is rather the point. The parts with the strongest evidence behind them — a fixed wake time, morning light, a dim and predictable last hour, and getting out of bed when sleep is not coming — cost nothing and are available to you tonight. A well-chosen supplement can support the edges of that. It cannot do the middle.
If you would rather have the routine laid out as something to follow than something to read, our free guide The Better Sleep Reset walks through the light, timing and wind-down changes step by step, and is the best place to start.
This article is for educational purposes only and does not constitute medical advice. Always consult a qualified health professional before starting any supplement regimen, and speak to your doctor about sleep difficulty that persists for more than three months. These statements have not been evaluated by the Food and Drug Administration. These products are not intended to diagnose, treat, cure, or prevent any disease.
