A 30-DAY PIVOT THAT
SAVED THE BUSINESS
MY ROLE
End-to-end Product Designer
TIMELINE
1 month
STACK
Figma, FigJam, Material Design 3, Confluence
IMPACT
Beta delivered on Day 30 — validated by fashion professionals, B2C model proven viable, company operations preserved
OVERVIEW
The Doris App emerged from a survival pivot. With the widget product failing to scale, the company faced closure. I identified the opportunity to shift toward a B2C mobile experience — a "style wizard" delivering personalized fashion recommendations and creating a new revenue stream through brand partnerships. I designed the entire product from scratch, in 4 weeks.
THE CHALLENGE
Prove a new business model was viable before funding ran out — with no existing product, no visual identity, and no room for slow iteration.
THE PROBLEM
The widget model hit a ceiling. Three gaps made the pivot inevitable:
No personalization layer: The widget showed clothes but didn't guide users. No styling logic, no preference data, no reason to return.
No B2C revenue path: No mechanism to monetize end users or attract brand partnerships.
No product-market fit signal: Without a real user-facing product, there was nothing to validate with the market.
"The tech existed. The product didn't."
THE SOLUTION
Shifted from a visualization tool to a data engine based on the insight that users seek guidance over discovery.
Logic: Onboarding captures high-intent data through preference elimination.
Framework: Grounded in Seven Universal Styles for personalized blending
Mapping the New Flow
KEY DECISIONS
Inclusive onboarding: Research showed fashion preferences are increasingly non-binary. I replaced the binary gender selector with a style-first flow — users mix categories based on identity, not labels.
Geolocation precision: In cities like São Paulo, temperatures vary significantly by neighborhood. I sourced weather data at neighborhood level to make outfit suggestions contextually accurate, not just regionally approximate.
Closet as data layer: Scoped to individual item logging to hit the deadline — but designed with a larger purpose. The closet captures wardrobe preferences, building a scalable intelligence layer with real B2B monetization potential.
Material 3 as foundation, editorial as output: To move fast, I aligned with stakeholders on Material Design 3 as the base. I then customized it into a premium visual language — making AI suggestions feel like a high-end fashion magazine, not a utility app.
Component reuse: I leveraged the widget system's component architecture for the outfit preview feature, reducing lead time while maintaining consistency across the platform.
RESULTS
The redesign transformed the Doris Mirror from a "confusing screen" into a high-performing sales tool.
85% Approval
from fashion professionals during testing.
VIABLE B2C
model proven, company operations preserved.
DAY 30
delivered on schedule.
