Sleep scientists on bedtime routines: where personalised stories fit

The name that keeps coming up is Jodi Mindell, a clinical psychologist at Children’s Hospital of Philadelphia who has spent decades studying how young children fall asleep. Her 2009 study in the journal Sleep put 405 mothers of infants and toddlers through a plain experiment: one group kept doing whatever they normally did at bedtime, the other followed a fixed three-step routine — bath, massage, then a quiet activity like a lullaby or cuddling — every night for three weeks. The routine group fell asleep faster, woke up less overnight, and their mothers rated their own mood as measurably better by the end of it. Nothing about the routine was exotic. It was the same three things, in the same order, every night.

Six years later, Mindell ran a much bigger version of the same question with an international team that included sleep researcher Avi Sadeh, surveying over ten thousand families across thirteen countries. The 2015 result is the one worth remembering: the benefit was dose-dependent. Kids whose parents ran a bedtime routine every night had fewer sleep problems than kids whose parents ran one occasionally, and occasional beat none at all. That’s a more useful finding than “routines help,” because it points at the actual mechanism. A child’s nervous system is learning a pattern, not enjoying a specific bath temperature.

The American Academy of Pediatrics distills this into something a tired parent can act on without reading a journal: brush teeth, read a book, go to bed, in that order, every night. Their own language on why it works is worth quoting directly:

The AAP is also specific about timing in a way a lot of routine advice skips: start winding down before your toddler is rubbing their eyes, because by then they’re already overtired, and putting them down even fifteen or twenty minutes earlier than feels necessary tends to help more than anything else in the routine itself.

Strip the research down and a bedtime routine for toddlers isn’t complicated. It’s three or four short steps, done in the same order, every night, starting before the kid is exhausted:

Here’s the part I have to be straight about: Moonmoth is a phone or tablet screen, open at exactly the hour the Sleep Foundation’s own guidance tells parents to dim the lights and turn screens off. None of the studies above tested a screen at all — Mindell’s routine was a bath and a lullaby, not an app. If you want to follow the research to the letter, the honest answer is a physical book, a dim hallway light, and the phone left in another room.

So why build Moonmoth at all. Two reasons, and neither one pretends the research doesn’t apply to us. First, it’s meant to sit inside a routine you already run, not replace it — a stand-in for the nights a parent is solo with two kids, traveling, or has read the same four board books until the words stopped meaning anything. Every design choice we made is aimed at behaving like the quiet-activity step of a routine rather than a typical screen: it’s audio-first, softly narrated over one ambient soundscape, the pages auto-advance so there’s nothing to tap through, the on-screen controls fade out three seconds after you press play, and the whole interface is dark. No notifications, no streaks, nothing asking for attention back once the story starts.

Second, the personalization is aimed at the same mechanism Mindell’s data points to — familiarity, not novelty. A story about your own kid, set in a vibe they already recognize (we wrote about the six we ship and why each one is built to slow a child down), is closer to “the same steps every night” than it is to a new cartoon. Same character, same soft narration, same five to fifteen minutes, a different plot. Repetition with just enough variation that a five-year-old doesn’t get bored of it by Thursday.

None of that erases the screen problem, and I’m not going to pretend it does. If a toddler’s bedtime routine is working fine with a physical book and a dim hallway light, the research says stick with it — it’s already doing what it needs to do. Where we think Moonmoth earns a place is on the nights the physical-book version breaks down and the alternative is no routine at all, which is the comparison Mindell’s data is actually making. A five-minute personalized story a kid finds calming and familiar beats a chaotic, routine-less bedtime. It doesn’t beat a parent, a book, and a lamp.

If you’re building a bedtime routine for a toddler from zero, the research is almost boringly clear: pick three or four steps, run them in the same order every night, and start before your kid is already tired. A story — personalized or not — is one option for the quiet-activity slot, not a replacement for the rest of it. If you want to try ours, the landing page has a free story and the six vibes we built around this exact research.