Declining number of clients as well as the number of services used by clients.
A portion of the clients that do not have a claimed termination are using services less frequently.
Clients exhibit different behaviours, making it difficult to accurately predict churn.
How can we Identify customers at risk of churning to take corrective action?
Build a model to predict the probability of churning for clients on a monthly basis.
Create KPIs to capture each client’s service frequency and churn risk over time.
Divide customers into three churn segments.
Identified high-risk clients allowed the client to put retention policies in place.
Customized interactions with different churn segments depending on the risk.