B2C Shopping
AI Mode
UX design
Design Testing
Team
Google Shopping (AI Mode)
Role
User Experience Design Intern
Much of my work is under NDA
I'm sharing what I can. For full details, feel free to contact me at design.hitarthi@gmail.com or @LinkedIn
TL;DR: About My Projects
Proposed and tested redesign with 800+ users, achieving 37% preference
Led design exploration across 20+ iterations, balancing AI capabilities with user needs
Supported high-priority initiative with 30+ design variations in a tight deadline
Problems in the Current Experience
Design Exploration
I explored 6 directions, testing different approaches to information architecture, visual hierarchy, and content density.
Option 1 & 2 Learnings
Minimal lists maximized speed but lacked decision-making context. Adding visual elements (brand logos, product images) improved recognition, but the non-interactive images created confusion about next steps. Both options showed only 2 cards at a time, requiring excessive scrolling.
Option 2 & 3 Learnings
Categorization helped users navigate different price ranges, but the carousel limited comparison. Expandable cards gave control but hid key information behind interactions.
Exploration outcome
Direction 6 balanced all user needs: scannability, context, and efficiency without requiring interaction. This direction advanced to testing with 800+ users.
Testing & Results
A/B tested with 800+ users across 20+ combinations to refine layout, information density, and visual treatments.
37%
Context helped decision-making
Better comparison
Key Learnings
Small details matter
A misinterpreted rating icon significantly affected user decisions, showing how minor design choices can impact behavior
Balance is critical
Finding the right amount of information: not too much, not too little was key to helping users make confident decisions
Every iteration has value
Even "losing" variations revealed valuable insights about user expectations and mental models that shaped future designs
Making responses more Actionable
After exploring different ways to represent brand lists, I focused on ensuring users had a clear next step that felt natural and intuitive: guiding them through their shopping journey.
Product Imagery Research
The Challenge
My Contribution
Example: Watches
Impact
5+ Categories tested
Guidelines created
Project 3: Contributing to a P0
Created in 1 week
50+ mocks
300+ tests
Project 4: Figma Plugin Side Project
In my final month, I took the initiative to collaborate with a Staff designer, learn Figma plugin development, and identify an opportunity to solve a repetitive design task.
Within three weeks, I designed, developed, and launched a plugin that automated repetitive task, making it available to all designers at Google.
From idea to launch
3 weeks
Team-wide launch
Get early feedback
Test at scale
Always ask why
Iterate with team
