Shu Ying Seng was a member of the first graduating class of the new Master in Analytics program of the National University of Singapore (NUS). In contrast to many of her classmates who had little prior work experience, Shu Ying had worked for the last 7 years for S-Mobile, a leading cellphone carrier in Singapore. Because she had loved working there and S-Mobile had helped her pay for her degree, she had returned to the company half a year ago.
Shu Ying’s last job at S-Mobile before she went to NUS was to manage the retention desk at the company’s call center. Her team’s task was to persuade customers who called to leave S-Mobile to stay with the carrier. While Shu Ying’s team could prove high “save” rates, she had always felt uneasy about this form of “reactive” churn management --- she felt that it trained customers to threaten to leave to get discounts.
One of the key moments during her master’s program was when Shu Ying discovered that data analytics provided a compelling alternative to reactive churn management: instead of waiting until a customer tried to leave the company, the company could be proactive and predict for each customer how much at risk they were of churning before they ever called and threatened to leave. This seemed like a much better approach because it allowed a company to act before customers were dissatisfied enough to want to leave, and retention offers had a much better chance of delighting customers. After all, how good could a retention offer look if it was given in response to a customer’s threat to quit?
Given Shu Ying’s experience in customer relations and her new data analytics skills, she now managed an analytics team in charge of improving customer churn. She realized that this was a big task. In the future she might have to change what data needed to be collected, how customers were treated, what plans were available, etc.
The first step, however, was to take the data that existed and to see whether they could be used to identify whether some customers were more likely to churn than others. And if so, what marketing actions and offers could be used to reduce churn.
Shu Ying asked her team to pull data on a random sample of customers in order to build a predictive churn model. The dependent or target variable was whether a customer had churned in the last 60 days. The predictor variables or features described customer characteristics and behavior over the 4 months preceding the last 60 days. The idea was to see whether these data could be used to predict whether a customer was likely to churn (see the data description at the end of this document).
The data consisted of two data sets (both contained in smobile_churn.RData):
The model was going to be used to “score,” i.e., generate churn predictions for 8,012 customers for whom Shu Ying had been authorized to test the proactive churn management program.
The task
As Shu Ying briefed her team, she laid out what they would have to accomplish:
using the rollout data set
rollout data set
CS 340 Milestone One Guidelines and Rubric Overview: For this assignment, you will implement the fundamental operations of create, read, update,
Retail Transaction Programming Project Project Requirements: Develop a program to emulate a purchase transaction at a retail store. This
7COM1028 Secure Systems Programming Referral Coursework: Secure
Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip
CS 340 Final Project Guidelines and Rubric Overview The final project will encompass developing a web service using a software stack and impleme
