Why Won't My Baby Sleep? A Consultant's Checklist
Before you fix anything, work out which problem you actually have, because “won’t sleep” covers four different ones with four different answers. Then check, in this order: the baby’s age against what is developmentally normal, the arithmetic of the last twenty-four hours, whether falling asleep or staying asleep is the failing part, the room, and what has changed this week. Most of the cases that reach me are timing, not method — and a decent number are not a problem at all, just an expectation set by somebody else’s baby.
I’m a working sleep consultant, which means I get paid when families decide they need help. So I want to be straight about this: this checklist is the first hour of a paid consult, and for a lot of households it is the whole consult.
First, name which “won’t sleep” you have
Write down which of these is true tonight, because families routinely try to fix all four at once and fix none of them.
- Won’t fall asleep at bedtime. Fights the crib, takes forty-five minutes, arches and protests.
- Won’t stay asleep. Goes down fine, then surfaces repeatedly through the night.
- Won’t nap. Sleeps at night, wakes after one cycle in the day.
- Won’t sleep past dawn. Everything works until roughly 5am, and then it doesn’t.
These have almost nothing in common mechanically. A bedtime fight is usually a timing or wind-down problem. Repeated night surfacing is usually about how sleep begins. Short naps are a sleep-pressure problem. Early rising is about the back end of the night. If you only take one thing from this page, take this: pick one, and leave the others alone for two weeks.
Age changes the answer more than anything else
Under about four months, a baby who wakes several times a night and sleeps in short blocks is a baby, not a problem. Nothing on this list will change that much, and I would not try. What helps is a dark room, a predictable enough rhythm, and lowering the standard you are measuring against.
From roughly four to six months, sleep consolidates for many babies and the household’s habits start to matter — this is where most of my calls come from. From six months on, the ceiling is usually not capability but consistency: what happens at 2am, and whether it is the same thing every time.
The other direction matters too. There are weeks when nobody should be starting anything — illness, a move, a brand-new daycare. I keep a full list in when not to sleep train, and it is worth reading before you commit to a plan you will abandon on night three.
The arithmetic of the last twenty-four hours
More than half the households I see are running a schedule problem and describing it as a behaviour problem. Three numbers do most of the diagnostic work.
Total sleep across the day. Not the night in isolation. A baby who napped four hours will not go down at 7pm like one who napped two.
The last wake window. The stretch between the end of the final nap and lights-out is the single most-adjusted number in my whole job. Too short and there is no pressure to fall asleep. Too long and you get the wired, inconsolable version, which I take apart separately in how to get an overtired baby to sleep.
Bedtime drift. Bedtime moves later by ten minutes a week and nobody notices for two months. Check what time lights actually went out for the last five nights, not what time you intend them to.
In consults, the pattern I see is a family convinced their baby “just doesn’t need much sleep,” whose last wake window has quietly grown by an hour since the summer.
Falling asleep is a different problem from staying asleep
Everybody surfaces briefly between sleep cycles — adults included. What decides whether a surfacing becomes a waking is whether the conditions your baby fell asleep in are still there. If sleep began while being fed, rocked or held, the baby who surfaces at midnight into a still, dark, empty crib has a genuine mismatch to solve, and calls you to solve it.
That is not a moral failing and it is not a reason to change anything. Plenty of families feed to sleep happily for a year. But it is the mechanism, and it explains why the bedtime fight and the 2am wakings often need opposite interventions. If you decide to change how sleep begins, the methods compared lays out the honest options, including the slow ones.
The room, and the week
The environment checks are cheap, one-off, and worth running once so you can stop wondering: fully dark, continuous sound rather than a timer that cuts out mid-cycle, a room on the cool side, and dressed for that room rather than for the season.
Then ask what is different about this particular week. Teething, a cold, a growth spurt, a new skill being practised at midnight, a time change, a trip, a new person doing bedtime. A baby who slept fine for three months and stopped four days ago is telling you about the last four days, not about your method.
What sends you to the pediatrician instead of to me
Some of this is not a sleep problem and no plan will touch it. Take these to your doctor rather than to a consultant, an app or a course: snoring most nights, mouth-breathing, pauses or effortful breathing, poor weight gain or feeding difficulty, a cry that reads as pain rather than protest, persistent night sweats, ear-pulling with fever, or any waking that arrived suddenly with new symptoms. That is a pediatrician conversation, full stop. When a plan is already running and failing, I go through the same rule-outs in when sleep training isn’t working.
What I’d tell a client
Pick the one complaint that costs you the most sleep and ignore the rest for two weeks. Fix the timing before you touch the method — earlier bedtime, protected last nap, honest look at the last five nights. Change one variable and hold it for a full week, because almost every “it didn’t work” I hear is a plan that was changed on night three. And if the answer to “how old is your baby” is under four months, the most useful thing I can do is tell you that you don’t have a problem to solve yet.
FAQ: why won’t my baby sleep
Why won’t my baby sleep at night but naps fine during the day?
Usually sleep pressure in the wrong place: long or late naps leave too little drive for bedtime. Cap the last nap and shorten the final wake window before you change anything else.
Is it normal for a 3-month-old to wake every two hours?
It is common, yes. Short blocks and frequent waking are normal in the first months, and are not evidence that anything has gone wrong or that you have missed a window.
Why does my baby fight sleep when they are clearly exhausted?
Because the wake window ran long, and the resulting state looks like energy rather than tiredness. The fix is counter-intuitive: bring bedtime earlier for several nights rather than later.
How long should I try something before deciding it failed?
A full week of doing the same thing every time, unless the baby is ill. Most plans that “don’t work” were three plans run over five nights.
