
Wow: Building a Design System Across a Multi-Product Platform
How I contributed to Prezi's first shared design foundation - tokens, components, and cross-product consistency across three products and millions of users.
Role
Contributing Senior Product Designer
Timeline
2021 – 2022
Scope
Token system · Component library · Cross-product consistency
Team
VP of Design · DS Manager · Engineering · Brand
Key results
Handoff time
Products unified
Three products, zero shared foundation
Before Wow existed, Prezi's three products - Present, Video, and Design - each had their own visual patterns and components.
The cost was felt everywhere: designers duplicated work across products, engineers implemented inconsistent UI, and new features took longer because there was no shared starting point.
Wow started as a system for Prezi Video. New platform gave us the opportunity to rewrite the components using new codebase and scale it to the entire product range. What began as a Video-specific foundation was eventually expanded into a shared backbone for Present, Video, and Design, covering the full end-to-end platform experience.
Tokens and components, simultaneously
Token layer
My contribution spanned two layers of the system simultaneously.
On the token side, I worked with the design team on the foundational decisions - colour, typography, spacing, and elevation - that needed to hold up across three products with different visual contexts.
Colors
Grey
Red
Orange
Green
Teal
Blue
Purple
Icons
Property (Selected examples)
Content (Selected examples)
Action (Selected examples)
Component library
On the component side, I contributed to building and documenting components from the core library: buttons, inputs, cards, navigation patterns, with production-ready handoff specifications. The design-to-engineering gap was one of the most expensive problems in the old system.
Components existed in Figma but weren't being implemented consistently in code because the specs weren't precise enough to remove ambiguity. I treated every component as a joined effort between design and engineering.
Legacy adoption
Getting existing products to adopt new components was the real challenge.
The old codebase had deeply nested, product-specific code with no token layer - values were baked directly into components. Swapping in new tokens wasn't a find-and-replace operation. Adoption was slower than the system's quality warranted - not because teams didn't want to use it, but because the technical migration cost was real and had to be sequenced carefully alongside live product work.
I was tasked with making the new components precise enough that engineers could trust them without negotiation. The goal was to make adopting Wow easier than continuing to maintain the legacy system.
Faster launches, fewer back-and-forths

40%+ faster handoff
Design-to-dev handoff dropped by more than 40% compared to the pre-Wow baseline. Engineers had a shared reference they could trust, which meant fewer rounds of back-and-forth clarification before build could start.
Shorter review cycles
Design review cycles shortened. With a shared component language in place, reviews shifted from resolving inconsistencies to evaluating decisions - a fundamentally different conversation that moved faster.
Higher designer velocity
Designer velocity improved across the board. Fewer decisions per screen, less time spent recreating solved problems, and faster onboarding for anyone new to the product because the system carried the baseline.
Full platform migration
All three products - Present, Video, and Design - migrated to Wow. What started as a Video-specific foundation became the shared language for Prezi's entire platform.
Systems thinking over component craft
Adoption matters more than architecture.
A design system is only as good as its adoption rate. The most architecturally elegant token structure means nothing if engineers can't implement it or designers route around it. The hardest design problem in system work isn't the components - it's the migration path from whatever existed before.