Designed a beauty tracker app that helps users understand what's in their products,not just find more of them. A self-initiated project grounded in real research and personal experience.
Users mentioned beauty waste unprompted in every interview. The real problem wasn't discovery, it was ownership.
Where this came from
I've worked as a beauty consultant. I've helped customers figure out what to buy, watched them struggle to understand ingredient labels, and seen the pile of half-used products that accumulates when people buy things that don't actually work for them.
The beauty industry sells discovery as the experience. But for most people, the experience is waste, of money, of products, and of time spent trying to figure out what went wrong.
That frustration became this project.
Understanding the problem space
The U.S. beauty and personal care industry generates over $100 billion in annual revenue. A meaningful portion is driven by impulse purchases, social media recommendations, and the anxiety of not knowing what actually works. A 2023 LendingTree study found that 33% of consumers regret overspending on beauty products, rising to 52% for Gen Z.
The problem isn't that people don't care about what they buy. It's that they have no reliable system for tracking what they own, understanding what's in it, or knowing whether it's working.
How Might We?
→ How might we give people the clarity to shop intentionally and use what they have?
→ How might we make ingredient information comprehensible without being alarming?
→ How might we replace influencer-driven discovery with something users can actually trust?
The finding that reframed everything
I came into this project with assumptions,that the main issue was product discovery, that people wanted better recommendations. The interviews challenged that.
I spoke with three participants aged 21–24 across structured interviews covering shopping habits, product usage, ingredient literacy, and organizational systems. Small sample, but the patterns were sharp.
Every participant mentioned beauty waste unprompted. Not overspending in the abstract,actual physical waste. Products purchased, used twice, and forgotten. Duplicates bought because they couldn't remember what they already had. This wasn't a discovery problem. It was an ownership problem.
Three insights shaped the design:
→ Organization comes before discovery. People don't need more product recommendations. They need to know what they already have.
→ Ingredients matter, but the barrier to understanding them is high. Users wanted to know what was in their products but found ingredient labels incomprehensible.
→ Social proof is broken. 76% of users said influencer recommendations feel fake. Real peer reviews and objective ingredient data were what people actually trusted.
Interview Findings + Affinity Mapping
Top: Key points from user interviews · Bottom: Affinity mapping to surface patterns
User Personas
Karissa and Greg, the user personas kept in mind while designing Glo
From pain points to possibilities
Mind Mapping
Taking pain points and brainstorming all potential ways Glo could solve them
Sketching
Sketching the main screens and key features of the app
Low-fidelity Wireframes
Low-fidelity wireframes laying out major components and features
Making ownership feel light
The design challenge was making product management feel light and rewarding rather than like homework. The onboarding quiz answered this,rather than asking users to do work upfront, it created a personalized experience immediately, shaping which products Glo surfaced and which ingredient warnings were most relevant.
The Ingredient Scanner
Users can scan any product and see what's in it, flagged in plain language. Not a list of chemical names,an interpretation. Is this fragrance likely to irritate sensitive skin? Is this preservative one to avoid? My early designs erred toward too much information,ingredient lists with long explanations that overwhelmed users in testing. I pulled back to a simple traffic-light system (safe, worth knowing, flagged) with expandable detail for users who want to go deeper.
High-fidelity Prototypes
Onboarding · Routines · Ingredient scanner
What testing revealed
→ "I wish there was a dupe/recommendations feature, and I'd like to see where I can buy the product"
→ Hue/color wheels needed to pick eye and skin tone more accurately
→ The safety scale was initially alarming with chemical names, simplified to traffic-light system
Improvements after testing
Features added after user testing: dupe finder, where to buy, refined ingredient display
Final Prototype
Final version of Glo, incorporating all user feedback iterations