Predicting customer churn is a critical component of any business. As a growing enterprise company, preventing churn is as important as growing our customer base here at Okta. However, because churn events are less common in the enterprise world, and patterns across customers vary greatly, predicting customer behavior is a complicated task. In this talk, we discuss how we tackle the task of churn prediction at Okta, the issues we encountered, and areas of improvement in our churn model.
The Challenge of Predicting Churn in the Enterprise World
Monday, October 23, 2017 - 12:00 pm
Sharon Lin is the staff data scientist at Okta, a cloud security and identity management company based in San Francisco. At Okta, Sharon works on building data analytics for products and leads the efforts in developing predictive analytics solutions. She holds a BA in statistics and economics, and a Master of Information and Data Science from the UC Berkeley School of Information.