Segmentasi Pelanggan Menggunakan K-Means Clustering: Menganalisis Metrik RFM untuk Strategi Pemasaran
DOI:
https://doi.org/10.35194/jmtsi.v9i1.4452Kata Kunci:
RM, customer segmentation, K-mean clustering, RFM analysis, retail industryAbstrak
Means clustering algorithm for customer segmentation in the retail sector. The background of the research is based on the need to gain a deeper understanding of customer behavior in order to develop more effective Customer Relationship Management (CRM) strategies tailored to each customer segment. The data used in this study was collected from sales transactions over a period of three years, from April 1, 2021, to August 23, 2023, totaling 62,677 transactions. The primary objective is to enhance CRM strategies by categorizing customers based on their purchasing behavior. The K-Means algorithm was employed to group customers according to their RFM values, while the Elbow Method and Silhouette Analysis were used to determine the optimal number of clusters. As a result, customers were classified into three main segments: General Development, General Maintenance, and Important Maintenance. Each segment is characterized by a unique combination of recency, frequency, and monetary values, providing insights into targeted marketing approaches. Short-term strategies include personalized promotions and targeted campaigns to encourage repeat purchases and increase the average order value. Long-term initiatives focus on developing loyalty programs and VIP services to enhance customer retention and lifetime value. These findings emphasize the effectiveness of data-driven CRM strategies in optimizing customer engagement and profitability in the competitive retail landscape.
Penelitian ini mengkaji penerapan analisis RFM (Recency, Frequency, Monetary) dan algoritma clustering K-Means dalam melakukan segmentasi pelanggan di sektor ritel. Latar belakang penelitian ini didasarkan pada pentingnya memahami perilaku pelanggan secara lebih mendalam untuk menyusun strategi Customer Relationship Management (CRM) yang lebih efektif dan disesuaikan dengan kebutuhan masing-masing segmen pelanggan. Data yang digunakan berasal dari transaksi penjualan selama periode tiga tahun, yakni dari 1 April 2021 hingga 23 Agustus 2023, dengan total sebanyak 62.677 transaksi. Tujuan utama penelitian ini adalah untuk meningkatkan strategi CRM melalui pengelompokan pelanggan berdasarkan perilaku pembelian mereka. Algoritma K-Means digunakan untuk mengklasifikasikan pelanggan berdasarkan nilai RFM, sementara Metode Elbow dan Analisis Silhouette diterapkan untuk menentukan jumlah cluster yang optimal. Berdasarkan hasil analisis, pelanggan terbagi menjadi tiga segmen utama: General Development, General Maintenance, dan Important Maintenance. Masing-masing segmen ditentukan oleh kombinasi unik dari nilai recency, frequency, dan monetary, yang memberikan wawasan dalam menentukan strategi pemasaran yang lebih tepat sasaran. Strategi jangka pendek mencakup promosi yang dipersonalisasi dan kampanye yang ditargetkan untuk mendorong pembelian berulang dan meningkatkan nilai rata-rata pesanan. Sedangkan inisiatif jangka panjang berfokus pada pengembangan program loyalitas dan layanan VIP untuk meningkatkan retensi pelanggan serta nilai seumur hidup mereka. Hasil penelitian ini menegaskan efektivitas strategi CRM berbasis data dalam mengoptimalkan keterlibatan pelanggan dan profitabilitas perusahaan di lingkungan ritel yang kompetitif.
Referensi
[2] B. Sharp, J. Dawes, and K. Victory, “The market-based assets theory of brand competition,” J. Retail. Consum. Serv., vol. 76, no. April 2023, p. 103566, 2024, doi: 10.1016/j.jretconser.2023.103566.
[3] S. Gupta, T. Justy, S. Kamboj, A. Kumar, and E. Kristoffersen, “Big data and firm marketing performance: Findings from knowledge-based view,” Technol. Forecast. Soc. Change, vol. 171, no. November 2020, 2021, doi: 10.1016/j.techfore.2021.120986.
[4] C. Rungruang, P. Riyapan, A. Intarasit, K. Chuarkham, and J. Muangprathub, “RFM model customer segmentation based on hierarchical approach using FCA[Formula presented],” Expert Syst. Appl., vol. 237, no. PB, p. 121449, 2024, doi: 10.1016/j.eswa.2023.121449.
[5] B. Ivens, K. Kasper-Brauer, A. Leischnig, and S. C. Thornton, “Implementing customer relationship management successfully: A configurational perspective,” Technol. Forecast. Soc. Change, vol. 199, no. June 2023, 2024, doi: 10.1016/j.techfore.2023.123083.
[6] F. Mohammed, R. B. Ahmad, S. B. Hassan, Y. Fazea, and A. I. Alzahrani, “An empirical evidence on the impact of social customer relationship management on the small and medium enterprises performance,” Int. J. Inf. Manag. Data Insights, vol. 4, no. 2, p. 100248, 2024, doi: 10.1016/j.jjimei.2024.100248.
[7] D. C. Li, W. L. Dai, and W. T. Tseng, “A two-stage clustering method to analyze customer characteristics to build discriminative customer management: A case of textile manufacturing business,” Expert Syst. Appl., vol. 38, no. 6, pp. 7186–7191, 2011, doi: 10.1016/j.eswa.2010.12.041.
[8] Z. Tang, Y. Jiao, and M. Yuan, “RFM user value tags and XGBoost algorithm for analyzing electricity customer demand data,” Syst. Soft Comput., vol. 6, no. April, p. 200098, 2024, doi: 10.1016/j.sasc.2024.200098.
[9] M. Alves Gomes and T. Meisen, A review on customer segmentation methods for personalized customer targeting in e-commerce use cases, vol. 21, no. 3. Springer Berlin Heidelberg, 2023. doi: 10.1007/s10257-023-00640-4.
[10] S. Singh and S. Srivastava, “Review of Clustering Techniques in Control System,” Procedia Comput. Sci., vol. 173, pp. 272–280, 2020, doi: 10.1016/j.procs.2020.06.032.
[11] M. Hosseini, N. Abdolvand, and S. R. Harandi, “Two-dimensional analysis of customer behavior in traditional and electronic banking,” Digit. Bus., vol. 2, no. 2, p. 100030, 2022, doi: 10.1016/j.digbus.2022.100030.
[12] O. Sokol and V. Holý, “The role of shopping mission in retail customer segmentation,” Int. J. Mark. Res., vol. 63, no. 4, pp. 454–470, 2021, doi: 10.1177/1470785320921011.
[13] Dedi, M. I. Dzulhaq, K. W. Sari, S. Ramdhan, R. Tullah, and Sutarman, “Customer segmentation based on RFM value using K-Means algorithm,” in 2019 fourth international conference on informatics and computing, 2019, pp. 1–7.
[14] M. Frasquet, M. Ieva, and C. Ziliani, “Online channel adoption in supermarket retailing,” J. Retail. Consum. Serv., vol. 59, no. November 2020, p. 102374, 2021, doi: 10.1016/j.jretconser.2020.102374.
[15] M. A. Rahim, M. Mushafiq, S. Khan, and Z. A. Arain, “RFM-based repurchase behavior for customer classification and segmentation,” J. Retail. Consum. Serv., vol. 61, no. June 2020, p. 102566, 2021, doi: 10.1016/j.jretconser.2021.102566.
[16] W. Qadadeh and S. Abdallah, “Customers Segmentation in the Insurance Company (TIC) Dataset,” Procedia Comput. Sci., vol. 144, pp. 277–290, 2018, doi: 10.1016/j.procs.2018.10.529.
[17] A. J. Christy, A. Umamakeswari, L. Priyatharsini, and A. Neyaa, “RFM ranking – An effective approach to customer segmentation,” J. King Saud Univ. - Comput. Inf. Sci., vol. 33, no. 10, pp. 1251–1257, 2021, doi: 10.1016/j.jksuci.2018.09.004.
[18] F. Solari, N. Lysova, M. Bocelli, A. Volpi, and R. Montanari, “Perishable Product Inventory Management In The Case Of Discount Policies And Price-Sensitive Demand: Discrete Time Simulation And Sensitivity Analysis,” Procedia Comput. Sci., vol. 232, pp. 1233–1241, 2024, doi: 10.1016/j.procs.2024.01.121.
[19] L. Hu, W. Zhang, S. Du, and X. Sun, “How to achieve targeted advertising with the e-commerce platform’s membership system?,” Omega (United Kingdom), vol. 130, no. November 2023, 2025, doi: 10.1016/j.omega.2024.103156.
[20] L. Xu and Z. Meng, “The role of membership fees in online retail market competition,” Res. Int. Bus. Financ., vol. 67, no. PA, p. 102089, 2024, doi: 10.1016/j.ribaf.2023.102089.
