This study aims to implement the Naive Bayes algorithm in predicting customer visit rates at Kyemoona Kitchen by utilizing available historical data. With the development of digital technology, data analysis has become an important aspect in supporting business decision making. However, manual analysis of complex and diverse data can be challenging. Therefore, a machine learning-based approach, specifically Naive Bayes, is used to explore patterns in big data and generate accurate predictions. In this study, the data collected includes variables such as visit time, promotion type, weather conditions, holidays, and other factors. The Naive Bayes model achieved an accuracy of 85.6%, with other evaluation metrics such as precision of 82.4%, recall of 84.2%, and F1-score of 83.3%. The results show that this algorithm can identify significant factors, such as promotions and weather conditions, that affect customer visits. This study not only provides practical insights for Kyemoona Kitchen in planning data-driven operational strategies, but also aims to inspire other small and medium-sized enterprises (SMEs) to adopt similar analytical technologies. However, this study has limitations, such as dependence on data quality, which can affect the accuracy of the model. Therefore, it is recommended that future research combine Naive Bayes with other algorithms and use larger datasets for more reliable results.