HGVR Brand is a creative industry engaged in the production and distribution of ready-to-wear clothing established in 2015, which has a reseller network in several major cities in Java. This study aims to analyze the prediction of HGVR Store product sales levels using data mining methods, specifically the C4.5 and Naïve Bayes algorithms, so that it can assist the company in determining marketing strategies and inventory management. The data used in this study consists of 500 sales data collected in June 2019 through observation, interviews, and internal company documentation. The input variables used include the number of orders (PO), quantity, price, and sales status, while the target variable is the classification of sales into "high" and "low" categories. The analysis process is carried out through the stages of data cleaning, transformation, and validation using the split validation technique (70% training data and 30% testing data). The C4.5 algorithm is used to build a decision tree model, while the Naïve Bayes algorithm is used to calculate the classification probability. The test results show that the C4.5 algorithm has a 100% accuracy rate with an excellent classification category based on the ROC curve (AUC = 1.00). Meanwhile, the Naïve Bayes algorithm also produced good classification results, although its accuracy was lower than that of C4.5. The conclusion of this study is that the C4.5 algorithm is more optimal than Naïve Bayes in predicting sales levels at the HGVR Store. These findings are expected to inform decision-making for the HGVR Brand in formulating business strategies.