Predictive models or 'Scorecards' helps marketers plan and manage the future - be it in building strategies for identifying Credit Card Customers most likely to default on their next payment, or in identifying prospects most likely to respond to a direct mail offer.

Click here for a Case Study on how a Consumer Finance company lowered its portfolio default rates by predicting for Prospects most likely to default on their first payment, and thereby declining their loan application.

If we take the example of how predictive modeling can be used for profitably acquiring new customers, it involves:

- Building ‘Response’ or ‘Targeting’ scorecards that will help you identify prospects that you want to acquire – either those that are most likely to respond to a particular product offering or those that will engage at a pre-determined level once they accept a given product.

- Uplift or 'Incremental' Models - very often models are built that target Prospects/Customers that would have responded in any case. Uplift models save precious resources by 'only' targeting those prospects that will respond given an offer.

- Adding a layer of complexity, the concept of ‘Lifetime Customer Value’ can be developed on a predictive basis and overlayed on ‘Response’ dimension leading to a holistic, profit-based customer acquisition strategy.

To see how this works in a real business scenario, please check out the following Case Study: Building a Profitable HELOC portfolio