Application of K-Means Clustering to Market Segmentation of Mobile Phone Users in Enugu State

Market segmentation based on usage behaviour enables telecommunications companies to design targeted service offerings that improve customer retention and revenue optimisation. This study applied K-Means clustering to segment mobile phone users in Enugu State, South East Nigeria, based on data usage, call frequency, mobile money transactions, and demographic variables. Survey data were collected from 480 smartphone users in Enugu Urban, Awgu, and Igbo-Etiti Local Government Areas using stratified random sampling. The K-Means algorithm was applied with k values ranging from two to eight, and the optimal number of clusters was determined using the elbow method and silhouette coefficient analysis. Four distinct user segments emerged: high-data social users, voice-dominant rural users, mobile money-centric traders, and low-activity elderly users. High-data social users were predominantly urban, aged 18 to 30, and primarily used data for social media. Rural voice users showed the highest sensitivity to tariff changes. Mobile money traders demonstrated the highest monthly transaction values despite moderate data usage. Cluster stability was validated using the Davies-Bouldin Index, which confirmed well-separated clusters. The study highlights significant usage heterogeneity across geographies and age groups within a single state. Recommendations include differential pricing strategies for each identified segment, targeted financial literacy programmes for mobile money users, and rural infrastructure investment to convert voice-only users into data subscribers. These findings contribute to evidence-based product design for Nigerian telecommunications operators.

Keywords: K-means clustering, market segmentation, mobile phone users, Enugu State, telecommunications analytics