Optimizing Marketing Strategy for PT PLN Through Customer-Based Clustering Using Machine Learning

Abstract

PT PLN (Persero) operates across Indonesia, with branch units managing diverse customer categories and volumes, including industrial, business, and residential customers. This variation in customer profiles presents challenges in optimizing marketing strategies and determining appropriate marketer allocations for each unit. To address this, this study applies machine learning, specifically the K-Means clustering algorithm, to group PLN units based on the number and type of customers managed. The Elbow Method identifies the optimal number of clusters, ensuring that the grouping reflects meaningful distinctions among units. Through clustering, PLN aims to enhance marketing strategies by tailoring resource allocation and optimizing marketer distribution. This study demonstrates that customer-based clustering can provide actionable insights for targeted marketing and resource optimization within PLN’s operational framework.