A Bibliometric literature review on telecommunication customer churn

This article aims to identify the most impactful research on customer churn in the telecommunications industry and assess the intellectual development, the characteristics of authors, and research trends in this industry. A bibliometric analysis was performed on 282 publications from 1999 to 2024, sourced from the Web of Science (WoS) database. Applied bibliometric techniques include co-word analysis, co-citations, bibliographic coupling, and co-authorship networks. The bibliometric analysis was conducted using the Biblioshiny package in R. The results indicated that the terms "customer churn," "churn management," and "churn prediction" were commonly used by the majority of churn researchers. The University of Southampton was identified as the affiliation with the highest number of published articles over time. China, India, and the USA emerged as the leading countries in terms of corresponding authors and the volume of published articles. Churn studies' most frequently used models include logistic regression, random forests, and various machine-learning techniques.