IA / Systems
Discovery
Navigation Redesign
Rebuilt the product's information architecture after workshop research revealed a fundamental user mental model mismatch.
Problem
Low feature adoption traced back to a broken user mental model — users experienced the product as one undifferentiated blob.
Outcome
+12% teampage visits (87.4% confidence, live A/B). 7/10 users completed tasks faster post-rollout.
Workflow
Strategy
Retrospective Prep Flow
Redesigned the retro preparation flow. Identified scope boundaries, shipped a validated solution, and advocated for the deferred problem in subsequent cycles.
Problem
Users weren't completing their first retrospective — the prep flow was cognitively overwhelming with no mental model of what they were preparing for.
Outcome
29% → 38% first retro completion (A/B, n=571). Retro survey adoption 3% → 15%.
Growth
Implementation
Signup & Activation
Two consecutive iterations on top-of-funnel and M1 activation. Implemented directly in the production codebase.
Problem
Signup and M1 activation were both underperforming and completely untested — no structured experiment had ever been run on either.
Outcome
+192% M1 activation 1.49%→4.36% (99.14%, n=1,974). +49.6% signup conversion (95.23%, n=3,500+).
Design System
Consistency
Retro Component Architecture
Replaced six inconsistent retro surfaces with a single context-aware component — designed, implemented in Angular, and shipped end to end into the live design system.
Problem
The retro feature appeared across six product surfaces that visually felt like different products — because most had no shared component at all.
Outcome
6 contexts unified under one component. 0→1 shared foundation. Signed off by devs, design & PM. A/B test live.