Grocery stores often face problems in stock management, such as shortages or excess stock. This can lead to financial losses and customer disappointment. This study aims to develop a website-based stock ordering management prediction system using the Apriori algorithm. This algorithm will analyze historical sales data to identify customer purchasing patterns and predict items that are often purchased together. Thus, stores can anticipate market demand and optimize stock ordering. This system is expected to help grocery stores improve stock management efficiency, reduce the risk of shortages or excess stock, and increase customer satisfaction. Efficient stock management is the key to the success of a grocery store. An accurate prediction system can help stores manage stock better, avoiding losses due to shortages or excess stock. In the apriori study, 4 rules were produced. if a consumer buys Royco chicken, the consumer also buys rice, if a consumer buys Ladaku, the consumer also buys rice, if a consumer buys Royco chicken, the consumer also buys Ladaku, and if a consumer buys Ladaku, the consumer also buys Royco chicken. Testing was carried out on the application system using blackbox testing, from 10 scenarios tested all were successful as desired. By using the apriori algorithm, stock management can be done using the results of the confidence value of the rule obtained, there are 3 items with the highest sales, namely rice, Ladaku and Royco chicken.