About
Churn is typically defined as
<MATH> \text{churn} = \frac{ \text{the number of churned customers} }{ \text{total customers}} </MATH>
Churn can be deceiving especially if your growth is accelerating (it will look lower than it actually is).
A customer churn:
- from a online store: When a customer stops buying from a business
- for a service company: rupture of contract
When a customer churns, you are impacting your growth negatively.
Churn is typically rare but quite costly
Model
In a churn model:
- The highest customers are the reliable customers.
- The lowest customers are unlikely to come back.
- The customers around 50% probability are at risk of churning,
Usage
- marketing campaigns (targeted to gain back the customers with a churn risk)
- order fulfillment prioritization,
Documentation / Reference
- Churn Analysis as outlier detection (e.g. Mamitsuka and Abe ’00).