This session shares how we evolved Prezi from a traditional SaaS product into an AI-first, agent-native experience, moving through several stages: AI features, AI onboarding, native AI editing, and fully agentic workflows.

For each stage, we will connect product choices to the metrics they most directly influence, including activation, time-to-value, retention, and conversion, and where measurement can get misleading.

We will also dig into the data challenge of chat-based UX: when users express intent in language instead of clicks, tracking becomes messier, attribution gets fuzzy, and “what happened” often lives in multi-turn context.