Marketing has become much more quantitative and data intensive in recent years. Strategies like interactive marketing, customer relationship management, and database marketing push companies to utilize the information they collect about their customers in order to make better marketing decisions. Marketing transaction data-which is a common type of Big Data-often forms the core set of information used for making marketing decisions. This course focuses on how analytic techniques from data mining, machine learning and statistical modeling can be applied to solve marketing problems.
The approach in this course is to complete a series of data intensive case studies which gives us a hands-on approach to learning marketing analytics. Each case exposes students to the data, the marketing problem, and a compatible analytical technique. Specifically, the case studies considered include pricing decision support systems using retail transaction data, understanding customer churn in the cell-phone market, upgrading freemium customers to paying customers, and lifetime cycles in direct marketing. On the analytical side we use regression, mixed models, logistic regression, decision trees, cluster analysis, and topic models.
This course is an intermediate level one, and is meant for those that have successfully completed the regression analysis course. Students are asked to complete a considerable amount of analysis and computer work, specifically using R statistical software. Although no prior knowledge about R is assumed.
1. Understand how to solve marketing problems involving pricing, promotion, and targeting using data mining techniques.
2. Hands-on analysis of marketing databases using data intensive case studies.
3. Learn how to apply and interpret the following techniques: regression models, mixed models, logistic regression, decision trees, and topic models.
4. Experience both the strengths and limitations of Data Science for business decision making.