When a forward forecast appears

Before 8 weeks corrected age, Cocoon shows elapsed time, caregiver-observed cues and the typical awake range for the age — never a predicted time.

Later, a simple feed estimate needs 4 recorded milk-feed intervals across 3 days; the detailed model needs 8 eligible intervals. Sleep times from your baby’s own pattern need 3 awake intervals across 5 days or your explicit setting; until then, from 8 weeks to 12 months, they are typical for the age (published wake-window ranges blended with the sleeps logged so far) and labelled so. After 12 months no age-based time is substituted. Unchecked logs remain a rough guide.

What the calculation uses

Feed timing uses up to 14 days of recent milk-feed intervals, weighted toward recent observations, with clear outliers removed. It combines breast and bottle feeds so combination feeding does not create artificial gaps, then compares daytime or overnight intervals. Sleep plans use recent wake gaps and nap slots. Invalid, future, overlapping sleep and exact duplicate records are left out. Days you mark as missing logs, illness or travel are excluded from learning; a disruption today withholds precise windows.

feed usual window =
  weighted 10th–90th percentile
  of comparable recent intervals

centre = weighted median

minimum width keeps a consistent history
from becoming an exact-time promise

Why it is a range

Babies vary from day to day, so Cocoon says “usual feed window” or “usual sleep window,” never “hungry,” “late,” or “due.” Feeding cues, tired cues, safe-sleep guidance, and a clinician's plan always take precedence.

Shared guidance, private feedback

Home, the day plan, Sleep and widgets use the same care-record calculations. Private timing feedback does not shift one caregiver's forecast away from another's. Times use the device's local timezone, so check the day plan when travelling.

The app keeps displayed personal forecasts, model versions and matched outcomes for up to 90 days on this device. Edits and deletions are reconciled. This evaluation history and your timing feedback are never uploaded. A matching logged event is not proof that a forecast improved care.

Full-day schedules

Full-day projections need at least 3 checked completed days plus recorded nap lengths. Otherwise only the next sleep is estimated. Nap counts and awake intervals come from records, not age tables. Short naps do not automatically move bedtime earlier. Later ranges widen as uncertainty compounds.

Day reviews and useful observations

Day reviews are optional, visible Journal notes shared with your family through normal care sync. They distinguish unreviewed logs from a day you know is complete. Adding, editing or deleting a care record invalidates its earlier completeness check. Last-nap observations compare three recent checked days with the three before, within the same usual or childcare context. Small or contradictory changes are not called out. These comparisons do not establish a cause.

Milk packing estimates require three checked days with expressed bottles during comparable hours. Nursing, mixed and unlabelled feeds are not converted to milk volumes. Stock counts exclude used, discarded, expired and unknown-expiry batches. The estimate assumes the recorded storage conditions continue; thawing and transport may shorten the use-by time. Trying a scenario never changes your logs or inventory.

Evidence and review status

Cocoon's predictive accuracy has not yet been established in a prospective study with real families. Software tests check how the calculations behave; they do not establish clinical accuracy. Population guidance supports total sleep ranges, safe sleep, and responsive feeding; it does not establish an exact timetable for an individual baby. The sources below were read and recorded by the Cocoon product team (October 2026). That is a review of published guidance, not a clinical endorsement of the forecast, and no clinician has signed off on an individual timetable — which is why every forecast is a range, shows its working, and defers to your baby's cues and your clinician's plan.

Limitations

Logging gaps, a new routine, illness, travel, and rapid developmental change can make recent history a less reliable guide. Until there is more of your baby's own history, Cocoon says so and keeps elapsed-time information available. Safe-sleep advice remains separate from forecasts.

Sources the team reviewed