Picky-eater tracker for parents

Consumer1 monthMobile

StackReact Native / ExpoSQLite

Market read pendingWe are still researching this idea.
How scoring works

The idea

Around a quarter to a third of toddlers go through significant picky-eating phases, and the standard clinical advice — repeated low-pressure exposure, sometimes 10–15 times per food — is exactly the kind of thing an app can track and a stressed parent can’t. Existing baby apps focus on milk feeds and weaning schedules for infants; the 2–6-year-old picky-eater niche is oddly empty.

Who pays

Parents, at the peak-anxiety moment when their pediatrician says “he’s fine, just keep offering foods.” One-time purchase beats subscription here — parents are subscription-fatigued and the problem is temporary. $12.99 once, no account, privacy-first (it’s data about a child — make local-only a headline feature, not a limitation).

MVP scope

  • 5-second meal logging with the three-outcome model
  • Food explorer grid + exposure counts
  • Rule-based pattern callouts and weekly suggestion (no server, no LLM)
  • One-time unlock via RevenueCat
  • Skip for v1: multi-child, caregiver sync, photo logging, recipes
// Build prompt — paste into Cursor / Claude Code / Lovable / Bolt
Build Bitesized: a mobile app for parents of picky eaters, using Expo
(React Native + TypeScript) with expo-sqlite for fully local storage.

Core loop — logging must take under 5 seconds: from the home screen, tap
a meal (breakfast/lunch/dinner/snack), tap foods from a recent-foods grid
(plus search over a built-in food list with emoji icons), then for each
food tap one of three outcomes: ate it / tried it / refused. That's the
whole log flow.

Insights screen: a "food explorer" grid showing every food ever offered,
colored by current status (accepted / warming up / refused / not offered
recently), a streak of "new foods tried this month", and pattern callouts
computed locally with simple rules — e.g. "Crunchy foods get accepted 3x
more than soft ones" (tag foods with texture/food-group metadata in the
built-in list) or "Broccoli has been refused 4x — research suggests 10-15
exposures; try again in a low-pressure way next week."

Include a weekly "try this" suggestion: pick one refused-but-not-recently-
offered food adjacent (same texture or flavor family) to something
accepted.

Design: warm, calm, zero guilt. Never scores the child; frames everything
as exposure counts, because repeated exposure is the evidence-based
mechanism. All data on-device; a one-time $12.99 unlock (RevenueCat) after
a 14-day full-featured trial. No accounts, no ads, no tracking — state
this prominently.

// More Consumer ideas