STRATEGY TO INCREASE ARPU OF PREPAID CUSTOMERS HIGH VALUE TELKOMSEL (CASE OF PREPAID CUSTOMERS IN THE JABOTABEK AREA OF JABAR) Purnama Adhiputra(a*), Dr Maya Ariyanti(b)
Faculty of Economic and Business, Telkom University, Jawa Barat
Abstract
In 2019, Telkomsel Jabotabek Jabar listed itself as the operator with the highest number of customers compared to its competitors with a contribution of 35.9%. Telkomsel Jabotabek Jabar^s biggest revenue came from the use of the Internet with a contribution of 70.3%, followed by the use of Voice & SMS at 22.1%, and Digital 7.6%. Based on the customer category, Telkomsel Jabotabek Jabar categorize customers based on subscription period (Length of Stay / LOS) and average usage for 3 months (Average Revenue Per User / ARPU) where customers with LOS > 6 months and ARPU > IDR 100,000 or known as a High Value customer, contributes 63.5% of total revenue based on March 2020 period data.
This research was made to examine how to increase revenue from High Value customers where segmentation will be created based on customer behavior, then from the segmentation results will be determined what marketing tactics will be used based on the concept of E-Marketing mix by utilizing the resources of Telkomsel Jabotabek Jabar. As for the target of this research was the Pre-Paid customers that were included in the High Value category due to the contribution of the largest number of Pre-Paid customers and its revenue.
The data collection method will use the Pre-paid customer population data along with variables that have been determined based on internal data (Big Data), then create a cluster using the K-Means Clustering method. From the specified number of clusters, characteristics of customers will be obtained which will become the basis for determining the marketing strategy that will be personalized to each customer. The theoretical approaches used in this study are Market Segmentation, Customer Behavior, Marketing Mix, and Personalized Marketing. Data analysis and processing was performed using Hadoop software and Phyton.
Keywords: Segmentation, High Value, Personalized, Big Data, dan Clustering
