Information technology unplugged : mitigating churn prediction in the mobile industry / P. S. Rajeswari., Vinoth Balaji V., R. Seetharaman.

Format:

Book

Author/Creator:

Rajeswari, P. S., author.

Balaji V., Vinoth, author.

Seetharaman, R., author.

Series:

SAGE business cases.

SAGE business cases

Language:

English

Subjects (All):

Customer loyalty--Case studies.

Customer loyalty.

Information technology--Case studies.

Information technology.

Physical Description:

1 online resource.

Place of Publication:

London : SAGE Publications: SAGE Business Cases Originals, 2026.

Summary:

Predicting customer churn is vital for mobile service providers globally and is particularly pressing in India. Customer churn, the rate at which customers leave a company, is costly as acquiring new customers is more expensive than retaining existing ones. Factors such as price sensitivity, poor service, billing issues, and better offers from competitors drive churn. Service providers are heavily investing in customer campaigns to position their market. The absence of customer churn prediction leads to huge losses for service providers. Predictive models that use machine learning can identify customers at high risk of churning. However, without strong data infrastructure, advanced analytics, and effective retention strategies, service providers risk losing both customers and profitability. Recognizing these pain points is crucial for mitigating churn and enhancing customer satisfaction. Common techniques include logistic regression, decision trees, random forests, gradient boosting, and neural networks. These models assign a churn probability, which highlights high-risk customers. The main problem involved in this case is that although marketers have various predictive churn models, selection of the best model is crucial to predict and control the customer churn. Predictive models are subjected to data quality concerns. Moreover, selecting appropriate performance metrics also plays a vital role. Thus, this case study focuses on how to select, train, and validate predictive models using different techniques and how to build a robust predictive model that minimizes customer churns. Students will be asked to select an appropriate predictive model using suitable examples. Students will also be asked to examine a predictive model and retention strategies in a real-world setting to control customer churn.

Notes:

Description based on XML content.

ISBN:

1-0719-8976-6

9781071989760

OCLC:

1569208519

Publisher Number:

T294978