Home // BUSTECH 2018, The Eighth International Conference on Business Intelligence and Technology // View article

Revenue Optimization of Telecom Marketing Campaigns for Prepaid Customers

Authors:
Maurus Riedweg
Pavol Svaba
Gwendolin Wilke

Keywords: telecom; churn prediction; predictive analytics; Naive Bayes; Nomogram

Abstract:
The design and optimization of marketing campaigns today usually still includes a high level of manual expert involvement. This applies particularly to the prepaid mobile phone sector of the highly competitive telecommunication industry. Since prepaid telecom customers are characterized by highly volatile and sparse usage data their future behavior is hard to predict, and marketers often rely mainly on experience and gut feeling when designing marketing campaigns, using only simple data analysis tools. The project developed a methodology and software prototype that helps marketers in this area to exploit the full potential of real-time big data-driven analytics for microtargeting, allowing them to make fact-based and informed decisions. Specifically, it provides an interactive solution for the semi-automated visual support of the design and optimization of single-channel marketing campaigns. The developed solutions bring a huge step towards the automation of the whole process of optimizing marketing campaigns in the telecommunication business, keeping the possibility of interactive interventions of marketers to implement strategic management decisions or use their expert knowledge. The system provides enough information for marketer to comprehend the reasons for the decision and by retracing it provides precious insights for the design of new campaigns. The solution uses machine learning closed loop and intuitive visualization based on nomograms and was prototypically implemented on Apache Spark big data stack and evaluated on sample data from two real-world prepaid telecom use cases.

Pages: 45 to 50

Copyright: Copyright (c) IARIA, 2018

Publication date: February 18, 2018

Published in: conference

ISSN: 2308-4391

ISBN: 978-1-61208-614-9

Location: Barcelona, Spain

Dates: from February 18, 2018 to February 22, 2018