Why people churn with causal inference

A high churn rate is especially a problem in marketing as it negates all the good marketing income that was generated from previous marketing campaigns. It is critical to have an in-depth understanding of why people churn and what factors need to be optimized to reduce the customer churn rate. As we have seen in previous sections, regression analysis and decision tree analysis are great at identifying linear relationships between the potential factors and the outcome and the inter-relationships between various factors and the outcome variable. However, as noted before, these identified relationships or correlations do not necessarily mean causations.

Identifying the causes of certain outcomes (for example, causes of customer churn) is often a difficult and complex task to achieve. This is where causal analysis comes in. If regression analysis was used to identify the relationships among the variables and decision tree analysis was used to...