PERANCANGAN SEGMENTASI PELANGGAN BERDASARKAN VARIABEL RECENCY, FREQUENCY DAN MONETARY DENGAN PENDEKATAN CLUSTERING PADA UMKM MICHIKO SIRUP CIREBON

• Main Author: Dwi Apriyani, Intan
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/86707
• Institution: Institut Teknologi Bandung

Michiko Syrup is a business that has a product known as Michiko Syrup and Juice which focuses on Mango Lip flavored drinks and syrups which are its flagship products. Michiko Syrup is also a regional beverage producer in Cirebon that not only contributes to significant local economic growth but also plays an important role in preserving traditional products, local culture, and creating jobs for the local community. Michiko Syrup has experienced a decline in revenue since the beginning of 2022 due to the impact of the post- COVID-19 pandemic which has not been favorable and it has been found that customer repurchase or repurchase rates are classified as very volatile caused by less personalized marketing strategies. Currently, marketing strategies are only done intuitively. Data-based and targeted customer segmentation is the key to designing a marketing strategy through customer loyalty much more effectively. This research aims to be able to determine Michiko Syrup customer segmentation using clustering techniques. The limitations of the data required that this research be carried out using note-based data owned by Michiko Syrup. The methodology of this research follows the stages of the Cross-Industry Standard Process for Data Mining (CRISP-DM). The basis of customer segmentation is using the recency, frequency, and monetary (RFM) model. This model has the ability to identify customer buying behaviors and characteristics. Then the customer segmentation model was built using the k-means clustering algorithm. This research produces a customer segmentation model and marketing strategy based on RFM variables with clustering techniques. Four customer segmentation clusters with distinctive characteristics are generated between the clusters which will then be mapped out the right strategy for each segment. The resulting customer segmentation is also expected to be more accurate because it is data-based. Michiko Syrup can take advantage of these proposed strategies based on customer segmentation, to reduce fluctuations in customer purchase rates and help in the creation of more effective strategies.