Predicting a Person's Churn

You can detect churn and be correct, but still not be able to prevent it. The event that ties to the churn isn't always something that you have data about. And so what you're really trying to do is get your predictions as close to that event as possible. You want a high level of accuracy to predict a person churning. But in order to actually be able to recommend an intervention, you have to understand a little bit more.