Dynamic Modeling of Customer Data

  • 7-weeks
  • 1.5 credits
  • Prerequisite: K507, X504

This course is designed to meet the increasing demands from the industry and recruiters for the application of quantitative and analytical skills to support sophisticated marketing decision making. The content of the course is based on cutting-edge research in optimization and interactive marketing to study customer relationship management. The goal is to apply optimization tools to derive analytical solutions to more customized and proactive marketing decisions, such as relationship pricing, promotion and inventory management, cross-selling campaigns and service allocation. The students learn how to make state-of-the-art marketing strategies and relevant analytical techniques that can be used to support these decisions. We aim to package the hard core strength of our students (quantitative techniques) in appealing and attractive business applications (relationship marketing). This course meets the imperative demands from the industry and enhances the comparative advantages of our MBA students.

Suggested courses to take before or with this course: Pricing (M574), Applied Marketing Research (M503) and Marketing Intelligence Management (M549).

This course is based on four case studies designed to give students the opportunity to analyze real-world marketing problems. The course provides students with the opportunity to practice problem formulation, develop problem solving and communication skills and to enhance their understanding of institutional knowledge of some emerging industries. The focus on real-world case studies gives students a better appreciation of the value of the analytical marketing tools in solving day-to-day marketing decisions. This course introduces the students to recent marketing practices such as:

  • Customer relationship management
  • Customer-centric marketing
  • Proactive marketing and customer lifetime value analysis
  • One-on-one interactive and dynamic marketing intervention
  • Firm learning and analytical decision making

This course will be a combination of lectures and case consulting projects. We invite students to work as consulting teams and work out analytical solutions for a variety of marketing problems. We believe that this format will highly motivate and involve the students. This setup also allows us to help students improve their institutional knowledge as well as professional presentation skill.

The marketing decisions addressed by our case projects are as follows.

  • Optimal design of two-parts pricing (which is increasingly adopted by software,  service, and subscription industries)
  • Win-back strategies
  • Cross-selling campaign management
  • Optimal allocation of service channels

These examples are used to illustrate some basic optimization primitives used in marketing including the following.

  • Using regression analysis to predict consumption.
  • Using discrete choice models based on the logistic function and estimating its parameters using the Maximum Likelihood method.
  • Setting up and solving the basic optimization decision problem for a firm.
  • Employing the latent class approach to segmentation in marketing.
  • Incorporating learning effects into optimization models using observed parameters such as consumption and consumer reactions to firm actions.
  • Adaptive learning of latent customer segments using a Bayesian update framework.
  • Use of dynamic programming on a time-discounted utility model reflecting customer lifetime values to solve the optimization model for the firm.

The case projects we assign are real world marketing problems, for which the industries have turned to the instructor for business solutions. We created slightly simplified version of these real business problems for students to work as projects. In this way, we have created a consulting experience for our students.

Kelley School of Business

Faculty & Research