Transparency

Built on real research

Our predictions are based on WHO guidelines, CDC recommendations, and peer-reviewed studies. Here's exactly what research backs each feature — and where the evidence runs out.

Most baby apps call their predictions “smart” or “magical.” We publish the math, cite the papers, and tell you when the science is unsettled.

Important Disclaimer

LilSense predictions are designed to help you anticipate your baby's needs, not replace your parental judgment.

Every baby is unique. Our predictions combine:

  • Evidence-based population data from WHO, CDC, and peer-reviewed research
  • Your baby's individual patterns learned from your logged events

Predictions become more personalized as you log more events. After 2 weeks of consistent logging, predictions are primarily based on YOUR baby's actual patterns. Always consult your pediatrician for medical advice.

Evidence by Category

Total Sleep Duration

Strong Evidence

How many hours babies typically sleep in a 24-hour period, including naps.

Sources: WHO 2019, Galland 2012, AASM 2016

Night Sleep Stretches

Moderate Evidence

How long babies typically sleep at night before waking.

Sources: WHO 2019, Galland 2012

Feeding Intervals

Strong Evidence

How often babies typically need to eat based on age.

Sources: CDC 2024, Johns Hopkins, WHO 2023

Wake Windows

No Evidence

How long babies can stay awake between sleep periods.

No peer-reviewed evidence available. Predictions based solely on your baby's individual patterns.

About Wake Windows

Important: “Wake windows” (how long babies can stay awake between sleeps) have no peer-reviewed scientific evidence. A PubMed search returns zero studies on this term.

LilSense does NOT use population averages for wake window predictions. Instead, we learn your baby's individual patterns from your logged events. This means wake window predictions require more logged data to become accurate.

Source: Emily Oster, ParentData (2023) - “Are Newborn Wake Windows Real?”Read more →

How we predict — the math

No black box. No proprietary “AI.” Just a statistical model we're happy to explain in public.

1

We weight recent data more heavily

Your baby's behavior from this week matters more than from a month ago — but last month isn't discarded. We use exponential smoothing with a factor of α = 0.25, meaning each new event pulls the prediction 25% toward the most recent value.

smoothed = α × new + (1 − α) × previous

2

We tell you when we're not confident

Every prediction has a confidence score. We track variance using Welford's online algorithm and compute intervals with a t-distribution. If your baby's patterns are inconsistent — or if we just don't have enough data yet — we lower the confidence instead of pretending we know.

Confidence = 60% sample size + 40% consistency.

3

Predictions get more personal as you log more

We start with population data from WHO, CDC, and peer-reviewed studies, and shift toward your baby's own patterns as we learn more about them.

Cold start

Fewer than 3 events

100% population data (WHO / CDC / peer-reviewed)

Learning

3 to 14 events

Linear blend — population + your baby

Personalized

15+ events

100% your baby's own patterns

What this is not

This is not machine learning. It is not AI. No large language model ever sees your baby's logs. No neural network is trained on your family's data.

It is a well-understood statistical method — the same class of math used in weather forecasting and inventory planning — applied to peer-reviewed pediatric sleep and feeding data. It works because the underlying research is good, not because the algorithm is clever.

Full Source List

Click any source to view the original publication

Guidelines on Physical Activity, Sedentary Behaviour and Sleep for Children Under 5 Years of Age

World Health Organization

World Health Organization (2019)

WHO guidelines establishing evidence-based recommendations for sleep duration in infants and young children. Recommends 14-17 hours for 0-3 months, 12-16 hours for 4-11 months, and 11-14 hours for 1-2 years.

Sleep duration predictionsNight sleep stretch estimates

View source

Normal Sleep Patterns in Infants and Children: A Systematic Review of Observational Studies

Galland BC, Taylor BJ, Elder DE, Herbison P

Sleep Medicine Reviews (2012)

Systematic review of 34 studies establishing reference values for infant sleep. Found mean sleep duration of 12.8 hours (range 9.7-15.9) for infants 0-12 months, with 0-3.4 night wakings for newborns.

Sleep duration varianceNap duration estimatesNight waking patterns

View source

Recommended Amount of Sleep for Pediatric Populations: A Consensus Statement

Paruthi S, Brooks LJ, D'Ambrosio C, et al.

Journal of Clinical Sleep Medicine (2016)

American Academy of Sleep Medicine consensus statement based on systematic review. Recommends 12-16 hours for infants 4-12 months and 11-14 hours for children 1-2 years, including naps.

Sleep duration predictionsTotal daily sleep estimates

View source

How Much and How Often to Feed Infant Formula

Centers for Disease Control and Prevention

CDC Infant and Toddler Nutrition (2024)

CDC guidelines for infant feeding frequency. Recommends feeding every 2-3 hours for 0-3 months, every 3-4 hours for 3-6 months, and 4-5 feedings per day by 6 months.

Feed interval predictionsFeeding frequency estimates

View source

Feeding Guide for the First Year

Johns Hopkins Medicine

Johns Hopkins Health Library (2024)

Clinical feeding guidelines from Johns Hopkins describing typical feeding patterns, amounts, and schedules from birth through 12 months.

Feed interval predictionsFeeding amount context

View source

Guideline for Complementary Feeding of Infants and Young Children 6-23 Months of Age

World Health Organization

World Health Organization (2023)

WHO guidelines on introducing solid foods and transitioning feeding patterns for infants 6-23 months.

Feed interval for older infantsMeal timing context

View source

Ready to try evidence-based predictions?

Free to download. No credit card required.

Download Free on iOS