How GameChanger used LTV modeling to predict and prevent churn

This post details how GameChanger used LTV modeling to predict and prevent churn. It explains the process of redefining lifetime value (LTV) by calculating 'Lifetime' (using survival regression for time-to-churn) and 'Value' (using ARPU based on subscription plans). It also covers the extension of LTV to free users by modeling advertising potential and likelihood of subscription. The post highlights the use of Hex notebooks for data preparation and exploration, Snowflake for data storage, and dbt for data modeling. It emphasizes how LTV insights clarified churn drivers and enabled revenue forecasting, ultimately leading to a more predictable business.