Daniel Caraway

Daniel Caraway

I put the fun in functions.

MSDS MAR 653: Financial Analytics

4 minute read

Marketing Analytics

MSDS (MBA) - Q3: MAR653

Description:

This course will focus on developing marketing strategies and resource allocation decisions driven by quantitative analysis. Topics covered include market segmentation, market response models, customer profitability, product recommendation systems, churn predictions, media attribution models, and resource allocation. The course will draw on and extend students’ understanding of issues related to integrated marketing communications, pricing, digital marketing, and quantitative analysis. The course will use a combination of cases, lectures, and a hands-on project to develop these skills.

Deliverables:

Class Outline:

You can use any of the following marketing research techniques:

  • Cluster analysis
  • Regression analysis
  • Perceptual maps
  • Logistic regression
  • Collaborative filtering
  • Text analytics
  • Ordinal logit
  • Conjoint analysis

Assessments

  • Class Participation
  • Group Homework Assignment 1
  • Group Homework Assignment 2
  • Project Idea and Team Composition
  • Project Plan and Research Design
  • Final Report and In-Class Presentation
  • Peer Evaluation
  • Final Exam Quiz

Unit 1 I Marketing Resource Allocation

  • 1.1 Weekly Readings
  • 1.2 What’s the Opportunity with Marketing Analytics?
  • 1.3 Moneyball Example
  • 1.4 Samsung Example
  • 1.5 Resource Allocation Process
  • 1.6 Resource Allocation Process: Pfizer Example
  • 1.7 Optimizing Sales Force at IBM
  • 1.8 Marketing ROI
  • 1.9 What to Expect with the Case Study Approach
  • 1.10 Setup for the Case
  • 1.11 Weekly Wall

Unit 2 I Cluster Analysis

  • 2.1 Weekly Readings
  • 2.2 What is Segmentation?
  • 2.3 Identifying Segments
  • 2.4 K-means Clustering
  • 2.5 Implementing K-means in Excel
  • 2.6 How Many Ks in K-Means?
  • 2.7 Profiling Segments
  • 2.8 Segmentations in Practice
  • 2.9 Using Prizm
  • 2.10 Weekly Wall

Unit 3 | Sticks Kebob Case Study

  • 3.1 Weekly Readings
  • 3.2 Introduction to Sticks Kebob Case
  • 3.3 How to do K-means in Excel Stat
  • 3.4 Profiling Segments
  • 3.5 Setup for Case Study Assignment
  • 3.6 Weekly Wall

Unit 4 | Regression Basics

  • 4.1 Weekly Readings
  • 4.2 Introduction to Regression Basics
  • 4.3 Understanding Regression Output
  • 4.4 Omitted Variable Bias
  • 4.5 Industry Perspective
  • 4.6 What is Log?
  • 4.7 Price Elasticity
  • 4.8 Statistical and Economic Significance
  • 4.9 Setup for the Case
  • 4.10 Weekly Wall

Unit 5 | Customer Lifetime Value

  • 5.1 Weekly Readings
  • 5.2 The Netflix Example
  • 5.3 Base CLV
  • 5.4 Digging Deeper into the Base CLV Model
  • 5.5 CLV Calculations: Netflix Rebooted
  • 5.6 CLV - Horses for Courses
  • 5.7 CLV - Cohort and Incubate
  • 5.8 The Power of CLV: IBM Example
  • 5.9 CLV Strategic Implications
  • 5.10 Setup for the Case
  • 5.11 Weekly Wall

Unit 6 | Propensity Models

  • 6.1 Weekly Readings
  • 6.2 Propensity Models
  • 6.3 Understanding the S-Shaped Curve
  • 6.4 Best Buy Example
  • 6.5 Interpreting Coefficients of Logistic Regression
  • 6.6 Market Share Predictions
  • 6.7 Hitrate: What’s It All About?
  • 6.8 Beers and Diapers
  • 6.9 Weekly Wall
###Unit 7 Predicting Churn
  • 7.1 Weekly Readings
  • 7.2 Introduction to Retail Relay
  • 7.3 How to Do Logistics Regression
  • 7.4 Measuring Hit Rate Demonstration
  • 7.5 Predicting Retention for Non-contractual Settings
  • 7.6 Weekly Wall
###Unit 8 Multinomial Logit Regression and Conjoint Analysis
  • 8.1 Weekly Readings
  • 8.2 Introduction to Choice Models
  • 8.3 Utility Theory and Model Formulation
  • 8.4 Business Application 1 - Conjoint
  • 8.5 Business Application 1 - Conjoint, Continued
  • 8.6 Business Application 1 - Conjoint, Pt. III
  • 8.7 Business Applications of Multinomial Cross Selling
  • 8.8 Conclusion
  • 8.9 Weekly Wall

Unit 9 | Marketing Experiments

  • 9.1 Weekly Readings
  • 9.2 Beers and Diapers Redux
  • 9.3 A Basic Experiment Design
  • 9.4 Advanced Experiment Designs
  • 9.5 Ohio Art Company
  • 9.6 Etch-a-Sketch Campaign Finance
  • 9.7 Weekly Wall

Unit 10 | Collaborative Filtering - Netflix

  • 10.1 Weekly Readings
  • 10.2 Collaborative Filtering
  • 10.3 Simple Slope One
  • 10.4 Ordinal Logit Deep Dive
  • 10.5 Value of Collaborate Filtering
  • 10.6 Weekly Wall

Class Project

Start with a directed research question and then ensure that you have data available to be able to answer that research question. I have provided some datasets that you may use for the project. Alternatively, you may wish to use data that you collect, have access to or source on your own from websites such as Kaggle. Kaggle datasets can be accessed at https://www.kaggle.com/datasets. Some academic journals such Journal of Marketing Research and Journal of Marketing have accompanying datasets with their published studies that you can access. Note that you can access journal articles and their datasets for no cost if you access them via https://library.syr.edu/help/work-off- campus.php

Your project idea should address a business problem for an existing brand or product using marketing analytics. Your final project report should provide marketing strategy guidelines for the business problem you addressed using analysis of marketing data. Some sample project ideas are given below (note that not all of these sample issues have data available):

  • What are customer perceptions of hybrid cars? How does a Toyota Prius compare to other hybrid cars in the market? You might use customer surveys and social media conversations to obtain a perceptual map.
  • Through a conjoint experiment, evaluate how consumers trade off between prices and the various attributes and features of a vacation cruise line.
  • Identify drivers of customer revenue and retention in an online grocery store such as Relay Foods. How can Relay Foods use this information to customize its product and marketing communications?
  • What are the different customer segments for Harris Teeter? Do the segment sizes differ across stores? How can Harris Teeter use this information to design its product assortment? The above are merely examples. You have a wide latitude in terms of what you can explore and discover.

You may also enjoy