The Baby Bedtime Routine

A consistent bedtime routine is the only step in this whole subject that was tested on its own, with no method attached and no crying in it, and it worked. Three or four steps, always in the same order, always ending in the same place, taking about twenty minutes. In the trial, families who did that settled their children faster, had fewer and shorter night wakings and better sleep continuity — and the mothers' own mood improved. The families who carried on as usual did not change. It is the cheapest thing on this site and it is the best evidenced.

That is an unusual sentence to be able to write about baby sleep, where most of what is published is a range with a caveat. So it is worth being precise about what was tested, because the precision is where the useful part is.

The trial, and what it actually did

Two age-specific randomised studies, covering 405 mothers and their children — one set aged 7 to 18 months, the other 18 to 36 months. Half the families were asked to institute a consistent nightly bedtime routine. The other half kept doing whatever they already did.

Over three weeks, instituting the routine on its own significantly reduced problematic sleep behaviour, shortened the time taken to fall asleep, reduced the number and the duration of night wakings, and improved sleep continuity. Maternal mood improved. The control families did not change over the same three weeks, which is the part that makes it a finding rather than a description of children getting older.

Read what was manipulated: not the bedtime, not the amount of sleep, not what happened when the child woke. Only the consistency of the half hour before bed. Everything else in this subject asks you to make a decision about crying; this asks you to do the same four things in the same order, and that turned out to be enough to move the numbers.

What the routine actually is

Three or four steps, the same order every night, ending in the same place — her cot, in the room she will sleep in, with the lights already down.

The NHS describes a simple, soothing version: a bath, changing into night clothes, dimming the lights, a lullaby. The AAP's account is a predictable series of steps that help a child wind down from the day, so she can understand and learn to expect what comes next. A typical shape, and it does not matter much which shape you pick: bath or a wash, into night clothes and a fresh nappy, a feed, a book or a song, lights down, into the cot.

The steps matter far less than that they never change. It is a sentence she learns to finish rather than a treatment, and the last step is the one doing the work — arriving in the same place, the same way, every night, so that the cot is the expected end of the evening rather than a surprise at the end of it.

How long it should take

The trial did not publish a length, so nothing here is quoting one. What follows is a rule of thumb rather than a finding: somewhere around twenty to thirty minutes for a baby, half an hour to forty minutes for a toddler who wants a story with an argument in it.

Shorter than that and there is not enough of it to recognise. Much longer and it becomes an event, which is worth resisting, because an event is a thing a two-year-old can extend by negotiation. The useful discipline is to keep the last ten minutes deliberately dull — dim, quiet, undramatic, identical — so the interesting part of the evening finishes well before the settling starts.

If the day is running so late that the routine keeps getting cut in half, the problem is upstream of the routine. The sleep schedule generator stacks the awake stretches for an age into a sample day so you can see where the evening is actually going, and wake windows by age explains what those bands are and are not.

Set the bedtime, rather than waiting for eye-rubbing

The AAP is worth quoting almost directly here, because it contradicts the advice most parents are given. Bedtimes and nap times should be set and kept, rather than waiting for a child to start rubbing her eyes or yawning — by then it is probably already too late. And putting a child to bed even 15 to 20 minutes earlier can make a big difference.

That second sentence is the most useful small change on this page. Fifteen minutes sounds too trivial to matter and it is the size of intervention that most often works, partly because it is small enough that a household will actually do it every night for a fortnight.

Sleepy signals are still worth reading — they tell you whether the time you set is roughly right — but they are a check on the clock rather than the trigger for it. A baby who is showing you sleepy signals as the bath starts is a baby whose bedtime wants moving earlier, not one who needs a faster bath.

Why the bath is optional

Nightly bathing is a convention rather than a requirement, and it dries some babies' skin. A warm wash works, and so does simply starting with pyjamas.

What the trial tested was consistency, not soap. If the bath winds her up rather than down — and it does for a decent number of babies, who come out of the water livelier than they went in — move it to the morning and let the routine begin at the next step. The order being the same tomorrow is the active ingredient, not any particular thing in it.

The same goes for the other things families bolt on. Sound is a common one: if you use a machine it should be part of the same fixed sequence rather than a rescue you reach for at midnight, and volume and distance from the cot are the whole question — the noise player here starts quiet, runs on a timer and says so beside its own slider, and the guide to it covers what has and has not been measured.

Sameness, not length, is the active ingredient

A routine works by prediction. She cannot tell the time, but she can tell that the book always comes after the pyjamas and that nothing has ever come after the lullaby except the cot. That prediction is what lowers the temperature of the ten minutes before sleep, and it is why the routine can be short and still work, and long and still fail if the order keeps changing.

Which has three practical consequences. Anybody who does bedtime does the same steps in the same order, so the routine survives whoever is home. It travels — the same four steps in a hotel room are worth more than the familiar room without them, which is most of what makes travelling with a baby survivable. And it should not be abandoned during a bad fortnight; teeth, a cold or a new skill will wreck a week of nights regardless, and the routine is the one thing still telling her what is happening.

Keeping it identical is easier to intend than to do, so it helps to write the order down once — a printed day sheet on the fridge does that job, and a fortnight of it also shows whether the settling really is getting quicker or whether it only feels that way on the good nights.

Where the routine ends, and what has to be true there

Every night finishes in the same place, and the guidance about that place does not bend for any routine.

Babies go down on their back for every sleep, day and night, on a firm flat mattress that meets safety standards, in a cot clear of blankets, pillows, bumpers and soft toys, with a fitted sheet and nothing else, in your room for at least the first six months. Nothing weighted goes on or over a sleeping baby. If the last step of your routine is a feed and she falls asleep on you during it, put her down in the cot afterwards. Safe sleep basics has the full version, and it is the one page here that is not a matter of preference.

A routine also does not need to be a method, and it is worth knowing that it can be the whole plan. If you are weighing up whether to do more than this, the comparison of the methods describes what each one asks and what its evidence covers, and when to start covers the age question. Quite often the routine, held for three weeks, makes the rest of that conversation unnecessary.

When this is not a routine problem

Three weeks of a genuinely consistent routine and nothing has shifted at all is worth mentioning to somebody rather than adding a fifth step. So is a baby who cannot be settled at all, who is feeding poorly or not gaining weight, who is unusually sleepy or hard to rouse, whose cry does not sound like her usual one, or who snores or seems to struggle for breath in her sleep. And a parent who is not coping should ring somebody today rather than tighten the routine tonight. Speak to your health visitor or pediatrician. This page is general information, not medical advice, and it cannot see your baby.