About

This project analyzes Zerve’s event-level user activity data to understand which behaviors and workflows are most predictive of long-term success. I define success using retention and engagement-based metrics and transform raw event logs into user-level features such as activity consistency, feature breadth, and early lifecycle engagement. I compare successful and unsuccessful users, visualize key behavioral differences, and train a machine learning model to identify the strongest predictors of success. The results show that early engagement, consistent usage across weeks, and broader feature adoption are the main drivers of long-term success. These insights can help improve onboarding, activation, and product guidance strategies on Zerve.