TIME Best Invention 2023
25% Conversion Boost
20% Faster Design Iterations
500+ In-Person Sessions
The Challenge
One in four people can't find properly fitting shoes. Mass manufacturing can't accommodate natural foot variation. Custom footwear previously required expensive in-person fittings.
How might we enable accurate foot measurement using only a smartphone camera?
My Role
Lead UX Designer & Researcher (May 2020–November 2024)
- End-to-end product design across web, iOS, Android
- Built and managed research program with 250+ participants
- Collaborated with CEO, CTO, engineers, branding partners
- Guided company pivot from B2B to D2C strategy
- Built and managed research program with 250+ participants
- Collaborated with CEO, CTO, engineers, branding partners
- Guided company pivot from B2B to D2C strategy
Discovery & Research
Initial Challenge: No Precedent
- No existing consumer foot-scanning solutions
- Users had zero mental model for the task
- Required precision for custom manufacturing
- Users had zero mental model for the task
- Required precision for custom manufacturing
Research Scale
- Led recruitment for AI training data collection
- 500+ bicoastal in-person usability sessions
- 200+ consumer and customer discovery interviews
- 1000+ survey respondents
- 250-member ongoing research community
- 500+ bicoastal in-person usability sessions
- 200+ consumer and customer discovery interviews
- 1000+ survey respondents
- 250-member ongoing research community
Key Insight
Users needed progressive guidance during scanning, not lengthy upfront instructions. Real-time feedback > pre-task tutorials.
Research Infrastructure
IAMBISPHERE Research Community
- Built and managed 250-member consumer research panel
- Directed recruitment, screening, and all participant communications
- Directed recruitment, screening, and all participant communications
- Enabled continuous testing cycles (surveys, interviews, usability studies, product wear tests)
- Insights directly informed product enhancements and reduced journey friction
- Insights directly informed product enhancements and reduced journey friction
This permanent research panel allowed rapid iteration without starting recruitment from scratch through a third-party platform each cycle.
Design Process
Methodology: NSF I-Corps Venture Design
Participated in National Science Foundation's Beat the Odds and I-Corps Virtual Lean Bootcamps, applying their systematic Venture Design approach:
- Rapid hypothesis validation through customer interviews
- Evidence-based pivots (B2B → D2C shift)
- Real-time feedback integration into design iterations
- Stakeholder alignment around validated insights
- Rapid hypothesis validation through customer interviews
- Evidence-based pivots (B2B → D2C shift)
- Real-time feedback integration into design iterations
- Stakeholder alignment around validated insights
This structured methodology enabled us to test assumptions quickly and make confident strategic decisions.
From B2B Tool → Consumer Experience
Phase 1: B2B Fit Prediction Tool
Piloted with 3 brands while directing consumer and market research backed by the U.S. National Science Foundation
Insight: Traditional brands couldn't solve fit problems inherent with a mass manufacturing model that can't accommodate 1 in 4 foot shapes, revealing an opportunity for D2C
Phase 2: Research App
Designed mobile web, iOS, Android data collection apps
Purpose: Train AI to accurately generate measurements from just 3 photos of each foot, understand user behavior around scanning
Insight: Identified which aspects of foot scanning users understood intuitively vs. needed help
Phase 3: Client Portal
Consumer-facing mobile scanning experience
Challenge: Make unprecedented task feel simple and trustworthy
The Solution
Simplified User Flow
✓ Split into independent steps: Fit Profile (preferences) + Scans (measurements)
✓ Reduced cognitive load, allowed flexible completion
✓ Reduced cognitive load, allowed flexible completion
Instructional Clarity
✓ Video tutorials showing exact technique
✓ Real-time guidance during scanning
✓ Clear error messaging and retry options
✓ Real-time guidance during scanning
✓ Clear error messaging and retry options
User Control Features
✓ Rescan ability (mistakes happen)
✓ Profile editing (preferences change)
✓ Progress saving (complete at own pace)
✓ Profile editing (preferences change)
✓ Progress saving (complete at own pace)
Preparation Guide
Based on customer feedback and backend performance metrics, I redesigned the onboarding experience to onboard users with clear expectations and provide step-by-step guidance before being directed to the Client Portal.
Impact: These improvements increased scan completion rates by 25%, meaning 1 in 4 additional users who started scanning successfully finished the process. For a startup, this conversion improvement directly impacted revenue and validated our AI-powered approach.
Customer Journey Optimization
Beyond the scanning interface, I oversaw all customer touchpoints to build trust throughout the end-to-end experience and custom footwear process.
Pre-Purchase Education
- Social media and email campaigns explaining value proposition
- Built confidence in remote fitting process
- Addressed common concerns about custom footwear
- Social media and email campaigns explaining value proposition
- Built confidence in remote fitting process
- Addressed common concerns about custom footwear
Streamlined Order Flow
- Refined checkout experience for clarity
- Optimized transparency in order status
- Determined ideal frequency for updates (not too many, not too few)
- Refined checkout experience for clarity
- Optimized transparency in order status
- Determined ideal frequency for updates (not too many, not too few)
Post-Purchase Communication
- Developed follow-up sequences tracking order progress
- Gauged customer satisfaction and fit adjustments
- Encouraged repeat engagement and referrals
- Developed follow-up sequences tracking order progress
- Gauged customer satisfaction and fit adjustments
- Encouraged repeat engagement and referrals
Research-Driven Refinement: All communications tested and iterated based on customer feedback and response rates.
Cross-Functional Collaboration
Product design doesn't happen in isolation. I worked closely with:
Engineering Team
- Translated AI, physical, development, and other constraints into design requirements
- Ensured scan accuracy balanced with user ease
- Prototyped solutions developers could implement upon hand-off
- Ensured scan accuracy balanced with user ease
- Prototyped solutions developers could implement upon hand-off
Branding & Creative Collaboration
- Creative Director: Aligned website, social, email, and app touchpoints to execute unified creative vision
- Coordinated website updates, identified technical issues, managed iterative improvements
- Visual refinement of UI and promotional assets; facilitated tutorial shoot based on my scripts and storyboards
- Coordinated website updates, identified technical issues, managed iterative improvements
- Visual refinement of UI and promotional assets; facilitated tutorial shoot based on my scripts and storyboards
Leadership (CEO/CTO)
- Presented research insights to guide strategy
- Influenced pivot from B2B to D2C model - Aligned product vision with business goals
- Influenced pivot from B2B to D2C model - Aligned product vision with business goals
Manufacturing Partners (Portugal)
- Understood shoe-making constraints
- Ensured scan data translated to physical product - Validated measurement accuracy requirements
- Ensured scan data translated to physical product - Validated measurement accuracy requirements
Testing & Iteration
Validation Methods
- Moderated sessions (watch users struggle/succeed)
- Unmoderated remote testing (scale insights)
- A/B testing tutorial approaches
- Manufacturing verification of scan accuracy
- Unmoderated remote testing (scale insights)
- A/B testing tutorial approaches
- Manufacturing verification of scan accuracy
Major Iteration: Simplified from 360° video requirement to 3 photos per foot after research showed users couldn't execute complex videos but could handle photos with proper guidance.
Results & Impact
📈 25% increase in scan completion rates—translating to customers receiving custom sneakers more quickly and reduced support burden from abandoned attempts
🏆 TIME Best Invention 2023—validating the innovation and execution
🔄 Enabled B2B → D2C pivot—research insights demonstrated market opportunity traditional brands couldn't address
👥 250+ member research community established
📊 20% faster design iteration cycles—established research infrastructure enabled rapid testing
💰 Revenue Impact—Higher conversion directly translated to more orders and validated business model feasibility
User Feedback"Way easier than I expected"
"The videos made it so clear"
"Can't believe this actually worked"
Business Impact
Research insights directly informed company strategy, guiding shift to direct-to-consumer model and securing additional funding.
Research insights directly informed company strategy, guiding shift to direct-to-consumer model and securing additional funding.
Reflection & Learnings
What Worked
- Observational research revealed insights interviews missed
- Progressive disclosure reduced cognitive overload
- Real-time feedback built user confidence
- Observational research revealed insights interviews missed
- Progressive disclosure reduced cognitive overload
- Real-time feedback built user confidence
Challenges Overcome
- No precedent for consumer foot scanning
- Balancing ease of use with technical precision
- Building trust in unfamiliar AI technology
- No precedent for consumer foot scanning
- Balancing ease of use with technical precision
- Building trust in unfamiliar AI technology
Skills Developed
- Cross-platform mobile design (iOS, Android, web)
- AI/ML product design considerations
- Research program building from scratch
- Stakeholder management (CEO, CTO, developer, data scientist, branding, manufacturers)
- Cross-platform mobile design (iOS, Android, web)
- AI/ML product design considerations
- Research program building from scratch
- Stakeholder management (CEO, CTO, developer, data scientist, branding, manufacturers)
If I Could Do It Again
- Earlier international user testing (different floor types, lighting)
- More diverse technical comfort levels in early research
- AR exploration for real-time guidance
- Earlier international user testing (different floor types, lighting)
- More diverse technical comfort levels in early research
- AR exploration for real-time guidance
Additional Details
For a closer look at these projects and the processes behind them, visit the following case studies:
Fit Prediction Tool (Coming Soon)
Research App (Coming Soon)
Customer Journey (Coming Soon)
