Baby Sleep Schedule Generator — And Exactly How It Works

Start with two numbers, because the third one is a trap

The tool asks for age in months and this morning's wake time. It does not ask for weight, birth order, feeding method, temperament, or how last night went, and the reason is worth stating plainly: none of those improve the prediction enough to justify asking, and every extra field makes a tool feel more authoritative than its underlying evidence supports.

Age does real work. It determines the wake window range, the expected number of naps, and the daytime sleep budget — three quantities that shift substantially across the first eighteen months and barely at all across a fortnight. If your baby is between age bands, use the lower band for a week or two before switching up; babies grow into longer awake stretches rather than arriving at them on a birthday.

Morning wake time does the other half. Every subsequent time in the day cascades from it. This is why a tool that asks only for age and hands back fixed clock times — nap at 9:00, nap at 1:00, bed at 7:00 — is close to useless for the family whose baby woke at 5:40am. Their 9:00am nap is arriving after three hours and twenty minutes of wakefulness in a baby who can manage two and a half.

One genuine limitation of the two-input design: it assumes this morning was roughly normal. If your baby woke at 4:50am after a terrible night and went back down until 7:15am, feed in the later time, not the first stirring. The generator wants the start of the day, not the last night waking. That distinction sounds pedantic and it changes the entire output by two hours.

A note on corrected age, since it comes up often. For a baby born preterm, use corrected age — weeks since the due date, not since birth — until at least twelve months, and expect the fit to be looser than for a term baby. That is standard practice in developmental guidance rather than a quirk of this tool, and it matters more here than almost any other adjustment.

• Age in months — drives wake windows, nap count and the day-sleep budget
• This morning's wake time — every downstream time cascades from it
• Between age bands? Use the lower band for a week or two first
• Preterm baby? Use corrected age until at least 12 months
• Woke at 4:50am then resettled? Enter the later, real start of the day

Watch the arithmetic run on a real morning

Take a concrete case: a seven-month-old who woke at 6:45am.

Step one, the wake window. At seven months, typical awake stretches run roughly two and a half to three and a half hours, and they lengthen across the day — the first window of the morning is almost always the shortest and the final one before bed the longest. The generator applies a graduated set rather than one number, so it uses something like 2h45m, then 3h, then 3h15m, instead of three flat hours three times. Flat wake windows are the single most common flaw in hand-built schedules, and they reliably produce a first nap that arrives too late and a bedtime that arrives too early.

Step two, nap count and day-sleep budget. At seven months the expected pattern is two to three naps, carrying roughly three to three and a half hours of total daytime sleep. That budget gets allocated unevenly, because the first two naps generally hold most of it and any third nap is a short bridge to bedtime rather than a real sleep.

Step three, lay it out. 6:45am wake plus 2h45m gives a first nap at 9:30am. Seventy-five minutes of sleep ends it at 10:45am. Add three hours and the second nap starts at 1:45pm, running ninety minutes to 3:15pm. Add a final window of about 3h15m and bedtime lands at 6:30pm — which for a 6:45am riser is exactly right, producing roughly twelve hours in bed.

Step four, the adjustment that actually matters. If that arithmetic had put bedtime at 8:15pm, the tool shortens the day rather than accepting it, because a late bedtime for a baby who rose early compounds into an even earlier rise tomorrow. If it had produced 5:30pm, it lengthens the final window instead. Bedtime is anchored inside a sensible band and the last wake window flexes around it.

That is the entire model, and it is deliberately simple enough to run yourself on the back of a receipt when the naps go sideways — which is most days. The reference ranges it draws on are below.

The evidence half and the convention half

The parts of this tool that sit on published consensus and the parts that sit on parenting-community convention are not the same thing, and almost every schedule generator online blurs them. Separating them is the most useful thing this page can do for you.

The total-sleep figures are the well-supported half. The American Academy of Sleep Medicine's pediatric consensus statement recommends 12 to 16 hours of sleep per 24 hours, including naps, for infants aged 4 to 12 months, and 11 to 14 hours for children aged 1 to 2 years. Those are the ranges the generator sanity-checks its output against. Two features of them deserve attention. They are extremely wide — four hours and three hours respectively — which is the consensus panel's way of saying that healthy variation is enormous. And there is no AASM recommendation below four months at all, because the panel judged the evidence and the variation insufficient to support one. Any tool that hands a newborn a confident sleep target has gone past what the consensus was willing to say. This one deliberately does not.

The wake window figures are the weaker half, and it is only honest to say so. "Wake windows" are a parenting framework rather than a clinical measurement. You will not find a table of them in AAP guidance or in the AASM consensus. The underlying idea is sound — sleep pressure builds with time awake, and infants tolerate far less of it than adults — but the specific minute ranges circulating online are conventional, derived from aggregated practice, and vary noticeably from source to source.

The normative research supports caution about precision here rather than confidence in it. Galland and colleagues' systematic review of observational studies found wide variation around every sleep measure at every age studied. That is the finding to carry into any schedule tool: an average describes a population, not a target your baby has failed to meet.

So read the output roughly the way you read a clothing size. It gets close, most of the time, for most people. It is a starting point to adjust from.

Adjust the moment the day stops matching the plan

A generated schedule is a plan for a day that behaves. Days rarely behave, so the more important skill is recalculating, and it is simpler than it looks: take what actually happened, add the appropriate wake window, and that is the next sleep. The original clock times are now fiction and holding them is the most common way a good schedule turns into a bad one.

The rule of thumb for a short nap is compensatory. A nap that ran twenty-five minutes instead of ninety leaves your baby with far less relief from sleep pressure than the plan assumed, so the following wake window should be shortened by roughly a third — not held, and certainly not lengthened to get back on schedule. Trying to claw back to the printed times by keeping an under-slept baby awake produces an overtired baby, and overtired babies sleep worse, not better. The plan then breaks for the rest of the day and often for the night.

A long nap works the same way in reverse. If the second nap runs two hours instead of ninety minutes and ends at 4:00pm, the final window stretches and bedtime moves later. Cap that drift at about forty-five minutes past your usual, and if it keeps happening, shorten that nap tomorrow rather than pushing bedtime into the evening indefinitely.

A failed nap — twenty minutes of protest and no sleep — should be closed out rather than extended. Give it a defined attempt, twenty to thirty minutes, then get your baby up and shorten the next window by about fifteen minutes. Letting one failed attempt swallow an hour damages the rest of the day more than the missing nap does.

And on a day where two of three naps collapse, the compensation is bedtime, not a rescue nap after 5:00pm. Bringing bedtime forward thirty to forty-five minutes is the correct response, and it does not cause an early rise — chronic overtiredness reliably does, while an early bedtime after a bad day generally does not.

The practical version of all this is to rerun the tool mid-day using the real end time of the nap that went differently. Everything downstream recalculates, including bedtime. That is the intended way to use it, more than the morning run.

• Short nap → shorten the next wake window by about a third
• Long nap → let bedtime drift, capped at ~45 minutes past usual
• Failed nap → close it out after 20–30 minutes, trim the next window by 15
• Two collapsed naps → bring bedtime forward 30–45 minutes, skip the late rescue nap
• Any of the above → rerun from the real time, not the planned one

Ignore the schedule entirely in these situations

There are circumstances where following the generated day is actively the wrong move, and recognising them is worth more than the schedule itself.

Under four months, treat the output as informational only. Newborn sleep is not yet organised by a circadian rhythm — the day–night pattern establishes over roughly the first two to three months — so there is no underlying structure for any schedule to describe. Feeding on demand and following your baby's cues is the standard guidance for this period, and no arithmetic improves on it. The AASM's refusal to issue a duration recommendation below four months is the clearest possible signal on this point.

Mid-transition, the generator describes the pattern you are moving toward or away from, never the messy middle you are actually standing in. Those weeks need day-by-day flexibility no age-based lookup can supply — alternating nap counts, a bedtime floating anywhere from 5:45pm to 7:15pm, capped emergency catnaps. Our 3 to 2 nap transition and 2 to 1 nap transition pages cover the process properly.

When your baby is telling you something different. If the tool says the next nap is in forty minutes and your baby is rubbing their eyes, staring through you and losing coordination now, go now. Cue-reading beats clock-reading consistently, and this tool exists to inform your judgement rather than to overrule it.

During illness, teething, travel, or the week a new motor skill lands. Sleep needs rise during illness and schedules should loosen accordingly. A baby who has just learned to pull to stand will rehearse it at 2am for several nights, and no schedule adjustment prevents that.

And finally, if following the schedule is making things worse. A schedule that produces a better-rested, happier baby is a good schedule. One that produces daily battles over naps your baby does not want at times they are not tired is not, however well-sourced its numbers. That judgement is yours, and it is better information than any lookup table can offer.

Separately from all of the above: persistent poor sleep that does not respond to any schedule change is worth raising with your pediatrician rather than solving with arithmetic. Snoring, mouth-breathing, observed pauses in breathing, poor growth, or feeding difficulty are medical questions.

If you want the wider picture rather than one day's times, the homepage carries the schedule-by-age overview, and the wake window calculator handles the single-number version of this question.

• Under 4 months — informational only; feed on demand and follow cues
• Mid nap-transition — the tool shows the destination, not the journey
• Clear tired cues arriving early — go with the cues, every time
• Illness, teething, travel, or a new motor skill — loosen everything
• Daily conflict over naps — the schedule is wrong, not the baby
• Snoring, breathing pauses, poor growth — pediatrician, not a schedule tool