How NapCast learns your baby and predicts the next nap
If you've ever stood over a crib doing mental math — "she woke at 6:40, she's been up about two hours, so is now too early or already too late?" — that's the exact moment NapCast was built for. It's the small feature inside Cradlo that looks at how your baby has actually been sleeping and gives you a gentle heads-up about when the next nap is likely to land. Not a generic by-age chart. Your baby, this week. Here's how it works, and just as importantly, what it deliberately doesn't try to do.
It reads your real logs, not a chart
Most nap guidance online is a table: at this age, expect this much awake time. That's a fine starting point, but it describes an average baby who doesn't exist. Your baby has their own rhythm, and it drifts as they grow.
NapCast works from the sleep you and your other caregivers have already logged — when recent naps started, when they ended, and how long your baby stayed awake in between. From that it estimates two simple things:
- Roughly when the next nap is likely to begin
- Roughly how long the current wake window has left
The key word is your. If your baby consistently runs a little longer awake than the charts suggest, NapCast learns that and leans with it instead of fighting it. The more you log, the more it has to go on, and the more it sounds like your actual day rather than a textbook.
Why we made it a suggestion, not a command
This was a real design decision, and we went back and forth on it. We could have made NapCast loud — a countdown, an alarm, a red "NAP NOW" banner. We chose the opposite on purpose.
A prediction is a starting hypothesis, not a verdict. Babies have off days. Teeth come in, a growth spurt rearranges everything, a big day at the park burns more energy than usual. On any given afternoon, the live baby in front of you is a better source of truth than any estimate. So NapCast offers a window and then gets out of your way. You're meant to pair it with what you're actually seeing — the yawns, the glazed stare, the sudden fussiness — and trust those when they disagree.
What the prediction is genuinely good at is removing the cold start. When the babysitter arrives at 2pm, or grandma takes the afternoon shift, they don't have to reconstruct the morning from a foggy text thread. They open Cradlo and see roughly when the next nap is due, because the morning caregiver already logged it. That hand-off is where NapCast earns its keep.
What it can and can't promise
We try hard not to overclaim, so let's be plain.
NapCast can: give you a personalized, moving estimate that gets sharper as you log; smooth out caregiver hand-offs so everyone's working from the same picture; take a little of the daily guesswork off your plate.
NapCast can't: see the future, override your baby, or know about the thing you haven't logged yet. If today is wildly different from the last few days, the estimate will be off — and that's expected, not a malfunction. It's reading the recent past to make a reasonable guess about the near future, which is exactly what an experienced caregiver does in their head. We just wrote it down so the whole family can lean on it.
One more honest note: it works the same on web, iPhone, and Android, and it works whether or not the next caregiver was in the room this morning. That's the entire point of a shared log. The prediction is only as good as what's been written down — so the real magic isn't the algorithm, it's the habit of everyone logging into one place. NapCast is just the quiet payoff for keeping that single, shared record.