Is uplift modeling (i.e., net lift, incremental lift, or true-lift modeling) on your radar? If you have a high response rate, what about those who would have purchased anyway without expending the cost of contact? If your targeted retention is saving customers, what about the "casualties," i.e., the customers who actually respond negatively to contact — they would have stayed but now you triggered them to leave!?
By scoring each customer as to the incremental benefit of contact (yeah, you need a second training set, a control set, to do this), targeted campaign ROI has potential to go where no ROI has gone before!
And here's the description of a session on this topic from those at Fidelity, coming to Predictive Analytics World San Francisco, March 14-15, 2011:
Mining for the Most Truly Responsive Customers and Prospects
Stop spending direct marketing dollars on customers who would purchase anyway!
True-lift modeling can identify:
– which customers will purchase without receiving a marketing contact
– which customers need a direct marketing nudge to make a purchase
– which customers have a negative reaction to marketing (and purchase less if contacted)
This discussion will describe:
– the basic requirements needed to succeed with true-lift modeling
– scenarios where this modeling method is most applicable
– the pros and cons of various approaches to true-lift modeling
Kathleen Kane, Principal Decision Scientist, Fidelity Investments
Jane Zheng, Principal Decision Scientist, Fidelity Investments