Traditional retail, as well as micro, small, and medium-sized firms, play an important part in the Indonesian economy. However, with the rise of progressive business competition, such as competition from modern retail, traditional retail requires a strategy to better its business and performance. The purpose of this study is to identify consumer behavior in traditional retail based on data mining using Association rules-market basket analysis (AR-MBA). Data were gathered by collecting 150 shopping transactions. Furthermore, the pre-processing stage involved data cleansing, transformation, and reduction. The study's findings revealed that several association rules were established and validated. Based on these findings, various insights were obtained, including the fact that department 3 (snacks) is the most purchased item and is associated with items in other departments; there are association rules between powdered drinks and snacks, candy and snacks, toiletries, snacks, instant noodles and snacks, cigarettes and flavored drinks, and mineral water and flavored drinks. The findings are used to improve the performance and to expand the retail industry. This study recommends product stock management by increasing the number of products that consumers frequently purchase, product marketing strategies such as discounts, product bundling, and other promotions, and layout proposals based on association rules.