TIMELINE
Feb - May 2024
12 weeks
ROLE
Design Consultant
TEAM
2 Project Managers
6 Design Consultants
SKILLS
UX Research, Market Research, Product Thinking, Branding, Figma
WHAT IS ?
A personal fashion discovery engine for consumers and retail sales associates.
Founded in 2022, LOOKS aims to match users to products based on fit and style to help consumers shop more efficiently and generate additional revenue for sales associates in the digital space.
However, the current MVP fails to address this goal...
It consists of a simple posting, search, and shopping feature with limited ways of actually discovering fashion.
Therefore, we came up with a new project vision:
Shared Wardrobes
Within a community, users can create shared wardrobes by adding individual clothing pieces (upload image or AI recommended) into a collection fostering engagement and inspiration.
REFLECTIONS
As my first client-based project, this experience offered valuable takeaways.
While working with a client, our goals may not align.
The client requested skipping research to focus on designing app features. As a human-centered design team, we compromised by conducting research (interviews, competitive analysis, personas) while delivering initial designs based on the MVP for feedback. This reinforced the importance of balancing client needs and creating designs that are user-backed.
Transparency is key.
When the client made that request, we were initially taken aback. In our next team meeting, our PMs facilitated an open discussion, gathering input from everyone. I appreciated their transparent communication and aim to encourage the same in future leadership roles.
Rebrand = many iterations.
Rebranding the MVP required redefining its core market value, developing a new design system, and integrating client requests like AI features. The expanding project scope made connecting features logistically challenging, but it reinforced the importance of iteration in the design process.
Now, let’s take a look at how we got here ;)
...
How might we empower LOOKS users to more effectively explore and discover personalized clothing and fashion styles through AI-driven community building?
...
"Hi team! As a starting point, I’d like you to leverage AI in your designs for enhanced matching and fashion recommendations."
CLIENT REQUEST
created with stray kids discography,
peach iced tea, & big dreams
©
2025
| evelyn chaewon kim
CONTEXT
As a Design Consultant in Design Consulting at Cornell, a student-run human-centered design consultancy, we worked on a contracted project with LOOKS, a fashion discovery startup to improve their current MVP. I led market research on AI fashion matching algorithms and visual designs of the communities and shared wardrobes. One key constraint was prioritizing the client’s emphasis on creativity and innovation over technical feasibility.
INTRODUCING A GLIMPSE OF LOOKS’ REBRAND
Fashion Communities
Users can explore different fashion style communities through the map and list view to offer flexibility of choice between more organized vs. explorative views.
Messaging and FashionAI Recs
Users can ask for opinions and recommendations on fashion through direct messages with other users, within the community group chat, or by asking their FashionAI.
INITIAL IDEATION
How might we empower LOOKS users to more effectively explore and discover personalized clothing and fashion styles with AI?
We began by conducting market research on AI fashion matching, e-commerce fashion platforms, and community engagement with fashion to understand the current fashion landscape and existing AI systems in the industry to identify potential integrations for LOOKS.
AI Models
3D virtual fitting (style.me)
Fashion recommendations
Interactive search
AI Limitations
Training
Comprehension
Implementation
E-Commerce Trends
Sustainability
Personalization
“Social commerce”
At the same time, in order to meet the client’s request to brainstorm initial designs, we focused on improving existing MVP features...
Search
AI-driven user matching system based on similar fashion styles
Commentary
Comment option synced with personalized AI style analytics
Filtering
Filters based on size, style, brand, and price
❌ Lacks active fashion discovery
❌ Lacks building meaningful connections
✅ Optimizes existing features
✅ Introduces AI-driven features
"Hmm...actually can you create a more unique solution, diverging from the MVP and minimizing the social media platform approach?"
CLIENT FEEDBACK
“Yes, we’re on it!”
OUR RESPONSE
A NEW DIRECTION
Redefining Our Product
To inform this process, we evaluated the characteristics of key competitors and conducted 17 potential user interviews consisting of students interested in fashion or those studying fashion design or management. We identified 4 key insights to shape LOOKS’ rebrand.
Trendy
Users are curious about different aesthetics that has developed within Gen-Z based communities on Instagram and TikTok.
Community-Based
Users would like a tight-bond community where ideas and questions related to fashion can be exchanged freely.
Exploration
Users enjoy the process of discovering niche and unique brands and styling.
Personalization
Users want more detailed feedback and comments about their outfits and experience personally tailored outfit recommendations.
REDEFINING THE PROBLEM STATEMENT
...
How might we empower LOOKS users to more effectively explore and discover personalized clothing and fashion styles through AI-driven community building?
EXPLORATIONS + ITERATIONS
Fashion Communities
Discover and join new communities (different fashion aesthetics).
✅ Compact, efficient space use
❌ No preview of wardrobes
❌ Cumbersome scrolling
✅ Thumbnails of wardrobe items
✅ Clear information organization
❌ Cumbersome scrolling
✅ Interactive and engaging
✅ Proximity on style similarity
✅ Overview of community avatars
❌ Less structured layout
Shared Wardrobes
Inside a community, you can interactively add items (upload image or AI recommended) into a wardrobe together to build outfits and aesthetics that match the community.
Outfit Visualization
Manually add individual clothing items onto your avatar to visualize how they look on your body type. View comments and suggestions from others in your community.
Messaging + FashionAI Recommendations
Message and share items with others in your community through individual or community group chats. Ask FashionAI for outfit recommendations (text input or voice).
Onboarding
Questionnaire after account creation to gauge fashion styles for community matching.
A profile page where outfits can be saved to be accessible when creating shared wardrobes and used by FashionAI for personalized recommendations.
FINAL TOUCHES
Design System
We rebranded LOOKS with a refreshed design system to evoke a sleek, modern, and editorial aesthetic aligned with the fashion industry. The palette features complementary colors — orange channeling creativity and confident energy balanced by green’s harmonizing nature. We selected these unique colors to distinguish LOOKS from other fashion platforms. Additionally, the darker background enhances their contrast, complemented by the clean, sans-serif SF Pro Display font.
KEY REVISIONS
Integrated wardrobes into the community page
Avatar outfit ‘try-on’ instead of manual placement
FashionAI made accessible anywhere through one-click
...
INTRODUCING LOOKS’ FULL REBRAND
01 | Onboarding
Users can match with communities based on shared styles and personalize their experience by inputting details like height and weight for a tailored avatar.
03 | Shared Wardrobes
Within a community, users can create shared wardrobes by adding individual clothing pieces (upload image or AI recommended) into a collection fostering engagement and inspiration.
05 | Messaging and FashionAI Recs
Users can ask for opinions and recommendations on fashion through direct messages with other users, within the community group chat, or by asking their FashionAI.
02 | Fashion Communities
Users can explore different fashion style communities through the map and list view to offer flexibility of choice between more organized vs. explorative views.
04 | 3D Outfit Visualization on Avatars
Users can auto try-on shared wardrobe items on their avatar to visualize how it would look on their body type, view outfit feedback through the comments, and link clothing directly in the app.
06 | Profile
Users can view their own outfits (on avatar), view their saved looks (other users’ outfits they saved), and collections (build their own outfits from individual clothing pieces).
NEXT STEPS
Where to take LOOKS next?
After presenting our final prototypes to the client, we proposed key considerations for future development.
BACK TO TOP
Technical Research
How feasible is it for LOOKS to currently integrate AI-driven 3D outfit visualization technology on their platform?
Shopping Features
How do we turn LOOKS into a profitable e-commerce platform? How can businesses use it to reduce inventory?
Sustainability
What features can we build to encourage more sustainable fashion through online trading and/or resale?