Mining for the truly responsive customers and prospects using true-lift modeling: Comparison of new and existing methods

True-lift modeling, also known as uplift modeling, combines predictive modeling and experimental method to enable marketers to identify the characteristics of ‘true’ treatment responders separately from the characteristics of ‘baseline’ or control responders (that is, those who would have responded anyway). By concentrating truly ‘persuadable’ treatment targets in the top deciles, true-lift models achieve the same (or more) amount of response with fewer treatments (and lower treatment costs). The identified characteristics of the ‘persuadable’ population can then guide the hypotheses of future experiments and pinpoint the most responsive recipients for the treatment in future. This article explains the concept of true-lift modeling in detail, reviews existing methods, contrasts with the traditional approach, proposes new methods that can be implemented with most standard software, and recommends metrics for model assessment and comparison in true-lift modeling. Several new and existing methods are applied to three data sets from the financial services, online merchandise and retail industries. Built on the findings from our study and prior experience, we recommend some guidelines on usage of true-lift modeling methods.

[1]  S Greenland,et al.  Concepts of interaction. , 1980, American journal of epidemiology.

[2]  David H. Wolpert,et al.  The Relationship Between PAC, the Statistical Physics Framework, the Bayesian Framework, and the VC Framework , 1995 .

[3]  Yu Xie,et al.  Who Benefits Most from College? , 2010, American sociological review.

[4]  Dominique Haughton,et al.  Direct marketing modeling with CART and CHAID , 1997 .

[5]  Dominique Haughton,et al.  Application of multiple adaptive regression splines (MARS) in direct response modeling , 2002 .

[6]  Eric Siegel,et al.  Predictive analytics: The power to predict who will click, buy, lie, or die , 2013, Journal of Marketing Analytics.

[7]  L. Lai.  INFLUENTIAL MARKETING: A NEW DIRECT MARKETING STRATEGY ADDRESSING THE EXISTENCE OF VOLUNTARY BUYERS , 2006 .

[8]  Paul D. Berger,et al.  Direct Marketing Management , 1989 .

[9]  Roman Kubiak.  Net Lift Model for Effective Direct Marketing Campaigns at 1800flowers.com , 2012 .

[10]  Patrick D. Surry,et al.  Real-World Uplift Modelling with Significance-Based Uplift Trees , 2012 .

[11]  J. Louviere,et al.  The Role of the Scale Parameter in the Estimation and Comparison of Multinomial Logit Models , 1993 .

[12]  Paul Wang,et al.  Strategic database marketing , 1994 .

[13]  L. Tian,et al.  Analysis of randomized comparative clinical trial data for personalized treatment selections. , 2011, Biostatistics.

[14]  Nicholas Radcliffe,et al.  Using control groups to target on predicted lift: Building and assessing uplift model , 2007 .

[15]  Victor Lo.  New Opportunities in Marketing Data Mining , 2009, Encyclopedia of Data Warehousing and Mining.

[16]  Victor S. Y. Lo.  The true lift model: a novel data mining approach to response modeling in database marketing , 2002, SKDD.

[17]  Paul Burton Brown,et al.  Sexy Little Numbers: How to Grow Your Business Using the Data You Already Have , 2012 .

[18]  Scott A. Neslin,et al.  Next-product-to-buy models for cross-selling applications , 2002 .

[19]  Neil D. Lawrence,et al.  Dataset Shift in Machine Learning , 2009 .

[20]  D. Wolpert.  The Supervised Learning No-Free-Lunch Theorems , 2002 .

[21]  R J Glynn,et al.  Assessing the Comparative Effectiveness of Newly Marketed Medications: Methodological Challenges and Implications for Drug Development , 2011, Clinical pharmacology and therapeutics.

[22]  Edward C. Malthouse,et al.  Improving predictive scoring models through model aggregation , 2008 .