About this course

Course code TPBNET
Duration 2 Days

The true effectiveness of a marketing campaign is not the response rate; it is the incremental impact. That is, true effectiveness is additional revenue, directly attributable to the campaign, that would not otherwise have been generated. The problem is that targeting strategies often are not designed to maximize the incremental impact. Typical targeting models are successful at finding clients who are interested in the product, but too often these clients would have bought the product regardless of whether they received a promotion. In such cases, the incremental impact is insignificant, and marketing dollars could have been spent elsewhere. Incremental lift models are designed to maximize incremental impact (that is, the incremental lift over the control group) by targeting the undecided clients who can be motivated by marketing.


Before attending this course, you should

  • have experience using SAS/STAT software to build statistical models.
  • have experience using SAS/STAT to build predictive models. You can gain this experience by taking the Predictive Modeling Using Logistic Regression course.
  • have experience with linear regression and logistic regression. You can gain this experience by taking the Statistics 1: Introduction to ANOVA, Regression, and Logistic Regression course.

Who should attend

  • Statisticians, business analysts, and market researchers who build predictive models for marketing and retention campaigns

Delegates will learn how to

  • build incremental lift models that maximize the difference in response rates between the clients who receive the offer and their control group
  • identify good incremental lift predictive variables
  • build incremental lift models using a variety of techniques, with special focus on nonlinear additive modeling based on Generalized Naive Bayes Classifiers
  • evaluate and deploy incremental models.


What Are Incremental Lift Models and Why Do We Need Them?

  • incremental impact
  • incremental lift models versus propensity models

The Green Card Marketing Campaign

  • campaign details
  • the INCREMENTAL SAS macro
  • example of a simple incremental lift model

The Weight of Evidence and the Information Value

  • introduction
  • the weight of evidence (WOE) and the information value (IV)
  • estimating WOE and IV
  • penalizing WOE tables
  • the INFORMATION SAS macro

The Generalized Naive Bayes Classifier

  • introduction
  • background: generalized additive models
  • the Naive Bayes classifier
  • the Generalized Naive Bayes model
  • the GNBCREG SAS macro
  • goodness of fit statistics for logistic regression models

Net Weight of Evidence and Net Information Value

  • introduction
  • the net weight of evidence (NWOE) and the net information value (NIV)
  • Building Incremental Lift Models
  • introduction
  • nonlinearity in incremental lift models
  • evaluation incremental lift models
  • regression-based incremental lift models
  • non-regression methods
  • the net Naive Bayes and net semi-Naive Bayes classifiers
  • comparison of incremental lift modeling approaches

This course addresses SAS/STAT software.

2 Days


This is a QA approved partner course

Delivery Method

Delivery method


Face-to-face learning in the comfort of our quality nationwide centres, with free refreshments and Wi-Fi.

Find dates and prices

Online booking is currently not available for this course, to find out more please call us on 0113 220 7150 or email us at info@qa.com to discuss how we can help.

The world's leading organisations trust QA with their learning and development.

Find out why

Trusted, awarded and accredited

Fully accredited to ensure we provide the highest possible standards in learning

All third party trademark rights acknowledged.