Isabelle, vice-president of customer loyalty and insight at a big bank, has led the development of a package of new products/services for clients, and a five-minute presentation to explain the offering. In a pilot test, where client managers randomly select walk-in customers and offer to go through the presentation, some agree to listen but others don’t have the time. Several months later, when data about client profitability is available, she notices that average profit from clients who listened to the presentation is lower than those who did not. Disappointed by the outcome and at a loss to understand why, she pulls the customer-profile data hoping that data analysis will explain the decrease in profitability.
The overarching objectives are: (i) “Story-telling with data” – Does the presentation work? If so, how can Isabelle tell the story so that it works despite the negative profitability? (ii) Illustrating the role of analytics in the design of experiments and the analyses of results (iii) Understanding how different kinds of analytics (descriptive–predictive–prescriptive) can be used in a business context. The specific analytical tasks required are: (i) To run and interpret a single-variable linear regression model (ii) To run and interpret a multi-variable linear regression model (iii) To build visualizations to explain the differences between the findings of the two models.
- Story-telling with data
- Analytics
- descriptive analytics
- predictive analytics
- regression
- design of experiment
- analyses of experiments
- A/B testing
- visualization
- tableau
- female
- bank
- new product development
- digital
- Q12021