UX Research
·
Product Design
·
Jan–May 2025
BudgetCart.
An AI-powered grocery app that helps low-income users shop smarter, compare prices, and stay within budget.
The Problem
Grocery shopping for budget-constrained users is stressful, confusing, and unsupported by existing apps.
bro groceries at Target are SO expensive lately 😭
ikr i literally budgeted $80 and spent $140
and they never tell u if something is SNAP eligible until the register
there HAS to be a better way to do this
No budget tracking built into shopping
Diet/allergy info hidden until too late
SNAP/WIC eligibility unclear before checkout
Too many brands, ads, cognitive overload
“How might we design a grocery app that makes affordable, healthy, and accessible food more visible and manageable?”
87%
of users completed checkout within budget
4/5
users said it felt “easy to trust”
3-step
receipt scanning flow
The Process
01
Paper Prototype
Category-First, No Brand, No Price. Reduced cognitive load but users struggled to locate items and lacked confidence without price info.
02
Low-Fi Prototype
Item-First, Still No Price. Navigation improved but users still felt unsure without price reference.
03
Hi-Fi Prototype
Item-First + Lowest-Price Display. Showed lowest available price per item. Reduced uncertainty, increased trust.
Budget Ring Dashboard
A real-time spending visualization that shows budget vs. actual spend at a glance. Connected to a calendar for tracking frequency.
AI Receipt Scanner
Snap a receipt → AI analyzes → verify items → done. Transforms manual data entry into a 3-step frictionless workflow.
Price Transparency Alerts
When a substitution is triggered, the app shows exactly how much more it costs before the user confirms. Ethical design that prioritizes financial health over quick sales.
How We Got There
Heuristic Evaluation
Assessed usability against Nielsen's 10 heuristics to identify structural pain points early.
Usability Testing
Task-based sessions with 5 participants across 3 prototype iterations to validate navigation and clarity.
User Interviews
Semi-structured interviews with SNAP/WIC users to surface real emotional and cognitive friction points.
Onboarding
The entry flow acts as a setup wizard — users configure dietary restrictions and preferred stores so every feature downstream is tailored to them.
Reflection
What worked
• Price transparency reduced uncertainty
• Budget ring created emotional clarity
• Receipt scanner removed highest friction point
What’s next
• Live API connections to store inventories
• SNAP/WIC database integration
• Edge case error handling for OCR failures
next case study