PENENTUAN PRIORITAS REKOMENDASI STRATEGI UNTUK MENINGKATKAN LOYALITAS PELANGGAN MENGGUNAKAN DATA MINING

• Main Author: Muhammad Naufal, Alif
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/71511
• Institution: Institut Teknologi Bandung

The development of the Muslim fashion industry in Indonesia presents a good opportunity for sellers of Muslim fashion products, including small and medium-sized enterprises (SMEs). At the same time, the rapid development of technology and the adoption of e-commerce in Indonesia have also increased, so SMEs need to improve their competitiveness to compete in the market. Noore, which is the object of this research, has a problem of declining income due to a marketing strategy that has not yet adjusted to customer characteristics. Therefore, this research aims to determine the priority of recommendations for strategies that are appropriate for customer characteristics in order to increase customer loyalty and improve Noore's competitiveness in the market. To determine the priority of recommendations for strategies that are appropriate for customer characteristics, customer segments will be grouped using data mining techniques with the kmeans clustering method based on the variables recency, frequency, and monetary (RFM). Then, the customer lifetime value (CLV) of each customer segment is calculated to determine the loyalty value of customers in each. CLV is calculated by weighting RFM variables using the analytical hierarchy process (AHP) with the Expert Choice software. After the loyalty value and characteristics of customers are known, a customer development strategy that is prioritized will be designed using one of the multi-criteria decision making (MCDM) methods, complex proportional assessment (COPRAS). Two customer clusters are formed, with the calculation results of CLV for cluster 1 being 0.036 and cluster 2 being 0.092, which shows that cluster 2 is a more profitable group of customers. Three strategies are prioritized based on the characteristics of each cluster that is formed with the goal of allowing the company to focus on some strategies that are more effective according to customer characteristics. The loyalty point, merchandise, and free gift strategies are prioritized for cluster 1, while for cluster 2 the recommended priority strategies are free gifts, product guarantees, and up-selling strategies.