Predicting

Customer Response

Cornell Course

The success of any targeting strategy is dependent on the validity of the models used to select customers, which requires that value distinctions are optimized within the constraints of a marketing budget. While there are a variety of methods available to identify and assess likely drivers of customer response, their ability to differentiate and approximate relative influence on purchase behavior can be limited in scope.

In this course, you will explore logistic regression as a means to enhance the predictive specificity and granularity of response likelihood, estimating and iterating on logistic models to maximize expected profitability. You will go from identifying and leveraging categorical response data common to real-world business interactions to evaluating the probabilistic relevance of associated predictor variables to optimize customer selection for targeting. Along the way, you will compare the relative efficacy of a variety of approaches in their ability to improve return on investment, recognizing the potential implications of those differences with regard to marketing success.

The following courses are required to be completed before taking this course:

• Leveraging Customers for Growth
• A/B Testing and Analytics
• Customer Behavior Segmentation Analysis

Note: While familiarity with R/RStudio is not a prerequisite for this course, you will be asked to use R/RStudio for a variety of analytical activities. Solution code provided for practice opportunities can be copied, pasted, and adjusted for graded assessments, but some manipulation of that code will still be required.