Predicting Churn

American Express began looking for indicators that could really predict loyalty and developed sophisticated predictive models to analyze historical transactions and 115 variables to forecast potential churn. 

Amex wanted to shift their traditional approach to business intelligence of hindsight reporting and trailing indicators to a prediction model that would indicate how loyal their customer base was, now and in the future. By developing a new approach, looking in depth at spend analysis, our sophisticated predictive model analyses trillions of historical transactions, using over 100 variable factors to forecast potential churn. In the Australian market alone, they now believe they can identify 24% of accounts that will close within four months.