Fork It: Dinner Decider — from PRD to the App Store
Live on the App Store — validation in progress.
The problem
Dinner decisions happen at the worst possible moment — end of day, lowest mental energy. The blocker isn’t a lack of recipes; it’s the abundance of options. Existing apps require you to already know what you want before you can search. Fork It solves the decision moment before the recipe moment.
It started at home: I built it for my partner, who has always found cooking daunting.
Most “dinner decider” apps route you to a restaurant. Fork It routes you to your own kitchen.
The approach
The scoping philosophy was to strip the decision down to its minimum viable shape:
- One high-impact input: protein. That single choice narrows the space enough to make everything after it easy.
- Three curated options, not an endless scroll.
- Refresh replaces rather than accumulates — you never build a backlog of candidates to compare.
Just as deliberate were the non-goals: no meal planning, no AI chat, no giant recipe library, no delivery integrations. The guiding constraint for every scope call: a confident decision in under one minute.
What shipped
A React Native (Expo) app backed by a Node/Express API and Postgres (Neon), hosted on Railway. It shipped with auth, favourites, and 90 curated recipes with grocery lists — and it’s live on the App Store.
What the data said
TODO — validation in progress. The PRD defined the success metrics up front: time to decision, percentage of sessions ending in a selected meal, refreshes per session, and repeat usage. This section will be filled in as real usage data arrives.
What changed
TODO — this section depends on the data above. Predictions were made before launch; the honest comparison against real usage comes next.
Outcome / status
Live on the App Store — validation in progress. See it on the App Store.