Perfect Fit
Mobile iOS App
TL;DR
Perfect Fit is an AI-powered shopping app that solves online clothing fit issues through personalized size, style, and color recommendations. By combining user measurements, style preferences, and color analysis, the app reduces returns and builds shopping confidence. As project lead, I directed user research and design strategy while coordinating a 5-person UX team to validate a measurement-first approach to fashion personalization.
Project Overview
ROLE
Project Lead, User Research
UI Design, Information Architecture, Prototyping & Testing
Sep 2023 - Nov 2023
THE SETTING
Working on a tight 3-month timeline, our team faced the challenge of tackling multiple complex problems simultaneously - sizing inconsistencies across brands, personalized style recommendations, and color matching. With 5 designers bringing different experience levels and perspectives, I needed to maintain research rigor while ensuring team alignment and preventing scope creep in a solution that could easily become overwhelming for users.
The Big Idea
Through our research, I discovered four key insights that fundamentally shifted how we approached the shopping personalization problem:
The sizing problem is emotional, not just technical - 89% of users had bad fit experiences, but what really deterred them was the confidence loss and disappointment, not just the inconvenience of returns.
Users want body-type guidance, not just size matching - Beyond getting the right "Medium," people struggled with finding clothes that actually flattered their body shape, revealing an unmet need for styling intelligence.
Onboarding friction is worth it when value is clear - Despite initial concerns about asking for measurements upfront, users were willing to invest time in setup if they understood how it would improve their experience.
Navigation clarity directly impacts feature adoption - Our tree testing revealed that unclear labeling (like "PerfectFit Passport") killed user confidence in trying advanced features like AR try-on.
User Interviews & Surveys
→ revealed the emotional impact of poor fit
The deliverable: Conducted 15+ user interviews and surveys reaching 100+ participants across different demographics and shopping behaviors.
What I learned: Users' language around fit failures was surprisingly emotional rather than practical. Phrases like "I felt terrible about myself" and "I'll never shop there again" revealed that poor fit wasn't just an inconvenience - it was damaging their confidence and relationship with brands.
How I used it: This insight shifted our entire messaging strategy. Instead of positioning Perfect Fit as an efficiency tool ("save time on returns"), we reframed it as a confidence-building tool ("shop with certainty"). Our UI copy changed from functional language to empowering language.
Interview Insights
Poor fit creates emotional impact
Users recalled many bad experiences with shopping online and having their item not fit them well. However, their language was surprisingly emotional rather than practical, revealing that poor fit was damaging their confidence and relationship with brands.
1.
Users described experiences with phrases like "I felt terrible about myself" and "I'll never shop there again”
2.
Trouble matching sizes across brands
Users may know their size for one brand but it may not be the same for another brand. This creates anxiety and forces workaround behaviors that reduce shopping confidence.
Many users shop in-store so that they can assure that the sizing is accurate.
Want body-type guidance beyond basic sizing
Our surveys and interviews revealed that users struggled with finding clothes that flattered their body shape, even when the technical size was correct.
3.
Users are open to the adoption of softwares that can help them shop for their body types.
Literature Review + Competitive Analysis
→ validated market gap opportunity
The deliverable: Comprehensive competitive analysis of existing fashion apps (Stitch Fix, Trunk Club, etc.) and academic research on sizing standardization and user behavior in online shopping.
What I learned: Existing solutions focused heavily on style curation but largely ignored the technical sizing problem. Most apps relied on user self-reported sizes, which perpetuated the cross-brand sizing inconsistency issue. There was a clear gap for measurement-based personalization.
How I used it: This research convinced our team to prioritize the measurement feature over pure style recommendations. It also helped us identify white space in the market and shaped our value proposition around technical accuracy rather than competing directly with established styling services.
User Personas + Affinity Mapping
→ Identified three distinct user archetypes
The deliverable: Created comprehensive personas (The Realistic, The Trend-Setter, The Researcher) through affinity mapping of interview data and behavioral patterns.
What I learned: Each archetype had completely different triggers for abandoning online purchases. The Realistic gave up when sizing seemed complicated, The Trend-Setter when style felt generic, The Researcher when information seemed insufficient.
How I used it: These personas guided our feature prioritization and onboarding flow. We designed a branching onboarding experience - quick setup for The Realistic, style-focused questions for The Trend-Setter, detailed measurement options for The Researcher. This prevented us from creating a one-size-fits-all solution that satisfied no one.
1
2
Card Sorts
→ Streamlined information architecture
The deliverable: Conducted 3 card sorts (2 hybrid, 1 closed) with 30+ participants, focusing on navigation expectations around account settings, measurements, and product features.
What I learned: Users had strong mental models about where privacy settings belonged versus style preferences. They expected measurement data to live separately from social features, and wanted quick access to "try-on" features from multiple entry points.
How I used it: Restructured our app navigation to match user expectations rather than our internal feature groupings. Moved privacy controls to a dedicated section, created multiple pathways to AR try-on, and separated personal data management from style preferences. This reduced cognitive load during onboarding by 40%.
Iterative Tree Testing
→ Improved navigation success rates from 26% to 53%
The deliverable: Two rounds of tree testing with 30+ participants each, specifically targeting problematic user flows like AR try-on access and saved item purchasing.
What I learned: Small language choices had massive impact on user confidence. "PerfectFit Passport" confused 74% of users, while "Avatar Try-On" was clearer than "Virtual Try-On." Users also needed clearer question wording - vague prompts killed task completion rates.
How I used it: Made targeted copy changes and added missing navigation paths. Renamed "PerfectFit Passport" to "My Measurements," clarified unclear feature labels, and added direct shortcuts to high-traffic features. These seemingly small changes doubled our navigation success rates and gave users confidence to explore advanced features.
