Bedtime routines for kids might be the least controversial topic in sleep. The research supports them, pediatricians recommend them, and exhausted parents who finally get one working tend to become evangelists. But the usual summary of that research is wrong in a specific and useful way, and the correction changes what you should expect.
A routine does not make your child sleep much more. It makes them fall asleep faster, wake less, and stop fighting you at the door. Those are different promises, and the second one is the one the numbers support.
The short version
- Same steps, same order, every night. A randomized trial of 405 families cut time-to-sleep from 20.8 to 12.4 minutes in infants and dropped night wakings from 1.6 to 1.0 a night in three weeks.
- Expect better sleep, not more sleep. In a survey of 10,085 families, kids with a nightly routine slept 1.6 hours more at night than kids with none, but only about 11 minutes more per 24 hours. The routine moves sleep out of the day and into the night.
- More nights per week is better, right up to seven. Night wakings in that survey lasted 67 minutes for children with no routine and 19 minutes for children with one every night.
- Start it 30 to 45 minutes before lights out, and work backwards from the wake-up time to set bedtime.
- For curtain calls, try a bedtime pass. One card, good for a single trip out of the room, cut the calling out and the room-leaving in a small randomized trial.
- After age 2, an afternoon nap starts costing you. A review of 26 studies found napping past that age tracks with later bedtimes and shorter night sleep.
How much sleep kids need
These are the consensus figures from the American Academy of Sleep Medicine, published in 2016 and since endorsed by the American Academy of Pediatrics, per 24 hours with naps included.
| Age | Recommended sleep (per 24h) |
|---|---|
| Infants (4–12 months) | 12–16 hours |
| Toddlers (1–2 years) | 11–14 hours |
| Preschool (3–5 years) | 10–13 hours |
| School age (6–12 years) | 9–12 hours |
| Teens (13–18 years) | 8–10 hours |
Work backwards from the wake-up time. If a 7-year-old is up at 6:45 for school, a 9 to 12 hour target puts bedtime between 6:45 and 9:45 the night before, and the routine needs to start 30 to 45 minutes before the lights-out end of that window.
The trial everyone cites, and who paid for it
The study behind almost every bedtime routine article is Jodi Mindell’s, published in Sleep in 2009. It is a genuine randomized trial: 405 mothers and their children, split into two age groups, 7 to 18 months and 18 to 36 months, running three weeks each with the first week as a baseline.
The results were large. Among infants on the routine, time to fall asleep fell from 20.8 to 12.4 minutes, night wakings from 1.6 to 1.0, and the length of those wakings from 21.8 to 12.6 minutes. The longest unbroken stretch of sleep grew from 7.6 to 9.3 hours. Toddlers improved on wakings and on bedtime difficulty, though their time to fall asleep did not clearly move. Control families changed on nothing. Mothers’ mood improved.
Two things about it rarely get mentioned. Total nighttime sleep barely budged in either group, the pattern that repeats everywhere in this literature. And the study was funded by Johnson & Johnson Consumer Companies, three of the four authors were employees there, and the tested routine was a bath with a provided wash product, a massage with a provided massage product, then quiet time. That does not make the findings false. The design was randomized, the control group got the same attention, and the effects held across two separate age cohorts. It does mean the most-cited evidence for “have a bedtime routine” was produced by a company selling baby bath products, and you would not learn that from the pages citing it.
More nights is better, and here is how much better
Mindell followed up in 2015 with something much larger and much weaker: a survey, not a trial. Mothers of 10,085 children aged 0 to 5 across 14 countries answered how often they ran a bedtime routine, published in Sleep in 2015. The pattern was a clean staircase.
| Routine frequency | Bedtime | Minutes to fall asleep | Night wakings | Minutes awake at night | Night sleep |
|---|---|---|---|---|---|
| Never | 10:18pm | 23.5 | 3.17 | 67.2 | 8.2 h |
| 1–2 nights a week | 10:13pm | 27.5 | 2.94 | 46.8 | 8.5 h |
| 3–4 nights | 9:56pm | 24.9 | 2.78 | 34.3 | 8.9 h |
| 5–6 nights | 9:10pm | 20.2 | 2.45 | 23.8 | 9.6 h |
| Every night | 8:53pm | 18.2 | 2.20 | 18.9 | 9.9 h |
Those are infants and toddlers. Now do the arithmetic the paper prints but nobody quotes. Night sleep rises by 1.6 hours from the top row to the bottom. Total sleep across the whole 24 hours rises from 12.46 to 12.65 hours, which is about 11 minutes. So roughly an hour and a half of sleep moved out of the daytime and into the night, and almost none of it was new sleep.
That reframes the promise. A routine buys an earlier bedtime, a shorter struggle, and a child who stays down, which is most of what a tired parent is asking for. It does not buy a child who sleeps more in total. Hold it loosely, though: this is a survey, families were recruited through a parenting website, most mothers had been to college, and children who never have a routine may differ from children who always do in ways nobody measured. The 2015 paper also carries a Johnson & Johnson co-author.
What goes in the routine
The specific activities matter less than their character: calm, screen-free, same order nightly. A workable template:
- A warm bath or shower. Beyond the hygiene, the body cools afterward, and that cooling is one of the signs that primes sleep.
- Teeth, pajamas, and any “last calls.” Water, bathroom, favorite stuffed animal located. Handling these inside the routine stops them becoming stalling tactics later.
- A book or quiet talk, in the bedroom, lights low. This is the wind-down core. For older kids, a few minutes of chatting about the day works as well as reading.
- A consistent goodnight. Same phrase, same hug, same finality. Kids are lawyers, and ambiguity is an opening.
What stays out: screens. The hour before bed is the wrong time for tablets and TV, both for the light and because nothing on a screen is designed to make a child less alert. If removing them cold causes a revolt, swapping in something engaging goes down better than a bare ban. Audiobooks are the classic move.
Age by age
The evidence above is almost entirely from children under six, so what follows past that age is reasoning, not trial data.
Toddlers, 1 to 3. This is where the research sits and where routines do the most work. Keep it short, keep it identical, and expect bedtime itself to drift earlier as the routine settles.
School age, 6 to 12. The bath stops being the anchor and homework, screens and activities start eating the evening. The routine becomes a wind-down block rather than a sequence of chores, and reading in bed does most of the lifting. Caffeine also hides in more kid-adjacent products than parents expect, including chocolate, many sodas and most iced teas.
Teens. Here biology fights you. Puberty pushes the body clock later, so a 15-year-old told to sleep at 9pm often physically cannot, and the American Academy of Pediatrics has argued for years that middle and high schools should start no earlier than 8:30am. A routine still helps, but the lever that works is a fixed wake time and morning daylight, not a lights-out you cannot enforce.
Curtain calls and the bedtime pass
Every parent of a preschooler knows the pattern: lights out, then a drink, then a question, then a monster. There is a tested fix for this, and it is almost absurdly simple. Brie Moore and colleagues gave nineteen 3- to 6-year-olds a bedtime pass in a randomized trial published in the Journal of Pediatric Psychology in 2007: one card, exchangeable for a single trip out of the room or one parental visit, after which the parents ignore further bids for attention.
Children with the pass left their rooms and called out less than controls, settled faster, and were still doing better three months later. Parents liked it, which counts for more than it sounds, because harsher versions of this advice get abandoned in week one. Nineteen children is a very small trial, so treat it as a good idea with modest evidence rather than a proven protocol.
The nap problem
If bedtime keeps sliding later and your child is over two, look at the nap before you look at the routine. Karen Thorpe and colleagues reviewed 26 studies of napping in children aged 0 to 5 for Archives of Disease in Childhood in 2015. The most consistent finding across all of them was that napping goes with a later sleep onset and shorter, poorer night sleep, and the evidence was strongest beyond age 2.
The authors are careful, and so should we be. All 26 studies were observational, most relied on parent report, and none randomized children to nap or not, so the direction of cause is unclear. Dropping a nap is not free either: the afternoon that follows can be brutal. But if the choice is a 2pm nap or a 7:30pm bedtime, that trade is worth naming out loud.
When it falls apart
It will. Travel, holidays, illness, a sibling’s schedule. The reassurance from the dose-response data is that consistency is a batting average and not a streak: five nights a week already delivers most of the benefit, and getting back to it tomorrow beats never missing. If sleep problems persist for weeks despite a steady routine, meaning long bedtime struggles, frequent wakings, loud snoring or daytime exhaustion, that is the point to rule out a sleep disorder with a pediatrician rather than iterating on the routine. Anything involving medication, including melatonin, is a question for that pediatrician and not for a blog.
What this post does not know
The honest boundary is age. Every trial above studied children under six, and the largest study was a questionnaire. Nobody has randomized ten-year-olds or teenagers to a bedtime routine, so the age-by-age section is extrapolation from good evidence in younger children plus what is known about the adolescent body clock.
Nobody has separated the parts, either. The 2009 trial tested bath plus massage plus quiet time as one package, so which piece did the work is unmeasured. For the bigger picture on why the effort is worth it, see our companion piece on why sleep matters so much for kids.
This article is for general information only and isn't medical advice. If you have a sleep disorder or a health concern, talk to a qualified professional. Full disclaimer.