From group-chat chaos to cooking together
FamCook · Co-cooking product concept
- Role
- UX Researcher / UX Designer / UI Designer
- Period
- 3 months (self-initiated / course project)
- My Contributions
- End-to-end concept design · User research · Visual design & style guide
- Platforms
- Mobile app (concept)
- Collaborators
- Solo project — structured critique from 8 design mentors via ADPList
- Methods
- User interviews, affinity diagramming, competitive analysis, Crazy 8s, lo-fi & hi-fi prototyping, usability testing, mentor critique
50%
improvement in task success
30%
increase in user satisfaction
Discovery
Picture ten friends over for a housewarming cook-together: coordinating in a group chat, discovering everyone has different dietary needs, and realizing cooking skills vary widely across the group. That's the scenario FamCook started from. I ran 5 user interviews to understand how people actually organize group meals today, pulled out patterns with affinity diagramming, and ran a competitive analysis of what already existed. People cook together in roughly three settings — daily with family, day-to-day with housemates, or one-off gatherings with friends — and most of the friction sat before anyone touched a pan: tracking dietary needs, agreeing on dishes everyone would eat, communicating plans and getting feedback, and making sure the right ingredients were actually in the kitchen.
Define
There was more here than one project could reasonably solve. I mapped four possible directions — a shareable dietary profile, a recipe generator, a group communication platform, and a co-cooking instruction app — and weighed each against viability, privacy, and how directly it addressed the friction I'd actually heard about. The recipe generator and the communication app were the two directions that held up: together they covered the planning-and-preparation problem without needing a dietary "foodie card" nobody asked for. A co-cooking instruction app stayed on the table as an exploratory direction for later — worth revisiting, but not part of this scope. Accessibility was a concern from the start too — cooking means occupied hands and divided attention, so the interface needed a clear hierarchy that didn't depend on careful scrolling.
Develop
I used Crazy 8s to explore layouts fast, built a lo-fi prototype, and tested it with 4 participants. Their feedback reshaped the structure directly: a hamburger menu became widgets, an information card was added to explain ingredients, the calendar came out of the meal plan to cut clutter, and the cooking interface moved from a split screen to a single one, since people didn't actually want to compare two things with their hands full. I then took the design to 8 mentors through ADPList, and the recurring note was blunt — the design was functional, but it didn't hold up against platform conventions and lacked consistency.
The cooking interface moved from a split screen to a single, focused one — with hands full and a pan to watch, comparing two panels wasn't realistic.
Deliver
That feedback led to real changes, not cosmetic ones. Onboarding swapped a budget question for one about cooking skill, since budget data had no clear use later in the product, with an option to skip and complete it later. The meal plan and to-do list moved to sit pinned above the conversation itself, so a decision didn't get lost three messages later, with quick actions to accept or reject suggestions. And inspiration for what to cook came from three entry points — snapping a photo, picking a mood, or naming an occasion — rather than asking people to browse a generic recipe list.
Onboarding asks about cooking confidence instead of budget — with an option to skip and come back to it later.
The meal plan and to-do list sit pinned above the conversation, with quick actions to accept or reject suggestions, so a decision doesn't get lost three messages later.
The finished home screen — search, mood/occasion/snap entry points, and the group's meal plan, all one tap away.
Outcome
Task success improved 50% and satisfaction 30% across rounds of usability testing. This was also my first project as a UX researcher/designer, and the biggest lesson wasn't a specific screen — it was learning to prioritize. There were more real problems in this space than one product could fix, and choosing one mattered more than trying to solve all of them at once.
Takeaway
Design for the context, not just the screen — planning dinner on the sofa and cooking with wet hands are not the same interface problem. And structured critique from people who owed me nothing did more to improve the design than another round of internal iteration would have.