Here is the thing almost nobody tells new parents: your baby waking in the night is not a malfunction.
All babies surface out of sleep several times a night, at every age — including the ones whose parents swear they sleep through. When researchers put cameras in nurseries and watched babies overnight across the first year, they found that the number of times a baby actually wakes barely drops. What changes is whether the baby calls out. Some babies wake, look around, and drift back off without a sound, and their parents genuinely believe they slept all night. Your baby is not doing something other babies aren’t doing. Yours is telling you about it.
Why waking is built in
A baby’s sleep runs in cycles of roughly fifty to sixty minutes — about half the length of an adult’s. At the end of every cycle everybody, adults included, surfaces towards waking. You roll over and have no memory of it in the morning. A baby does it twice as often as you do, and in the early months a very large share of their sleep is the light, active kind, where they twitch, grimace, grunt and make noise while still fast asleep.
So the waking itself isn’t the thing we’re trying to eliminate. Nobody has eliminated it, and no approach has been shown to reduce the number of times a baby wakes. What behavioural approaches reliably change is what happens next: whether your baby needs you there to get back down, and how quickly everyone is asleep again.
The invisible half
This has a practical consequence for how you read your own nights. Parent-reported sleep runs systematically high: in the largest pooled review of normal infant sleep, parents over-estimated sleep duration by roughly fifteen to fifty minutes compared with movement monitors, and the gap was widest in infants. The reason is not dishonesty. You cannot see a waking you slept through.
In one trial of a behavioural sleep programme, parents reported that night wakings more than halved — but actigraphy showed that around ninety-seven percent of babies in both groups were still waking twice or more a night. The programme had worked exactly as intended: the babies were resettling on their own, and their parents had stopped noticing.
What this means at 3am
Three things, in order of how much weight they take off.
First, the number of wakings is not a scorecard. If your baby wakes three times and your friend’s “sleeps through”, the difference is very likely in the signalling, not the sleeping.
Second, a run of good nights followed by a run of bad ones is the expected shape of infant sleep, not a regression and not something you undid. When researchers tracked six-month-olds over thirteen consecutive nights, only one baby in forty-four managed eight hours on every single night, and half never did at all.
Third, the thing worth working on is not waking but resettling — and the best-supported first step there is dull and reliable: a short, consistent bedtime routine, the same two or three calm things in the same order every night. In a randomised trial it improved how fast children fell asleep and how much they woke, and most of the benefit appeared within three nights.
When waking is worth a call
None of this means every waking is nothing. Snoring three or more nights a week, mouth breathing in sleep, pauses or gasping, working hard to breathe, poor weight gain, or clusters of brief jerks on waking are all things to raise with your pediatrician — not with a sleep coach. If any of those are happening, start there.
For everything else: your baby is on the curve, and so are you.