Implementation of K-Means Algorithm to Determine Marketing Strategy

In the ever-evolving digital era, the use of artificial intelligence (AI) technology is becoming increasingly important for companies in optimizing their marketing strategies. Through this study, applying K-Means analysis to product sales data at ABC Soap Stores, especially for several soap brands, results were obtained from the last three months of sales data, namely October to December 2023. The K-Means algorithm was used to uncover hidden patterns in the data, allowing customization of marketing approaches by ABC Soap Shop. The analysis process showed consistency in data grouping in Iteration I and Iteration II, with random cluster center selection in Iteration I involving data 1 and data 3. The results of Iteration I are used in Iteration II, where the same cluster center is found. Analysis of soap product sales data, it was identified that Cluster 1 consisted of nuvo, dove, Dettol, Lifebouy and shinzui soap products. Products in Cluster 1 show a consistent buying pattern, so it is advisable to add more stock because these products have a high level of interest from consumers, which is reflected in the increase in purchases every month. Meanwhile, Cluster 2 consisting of Lux Soap showed lower interest, and therefore, it was not recommended to increase the stock. Products in Cluster 2 still have sufficient stock, and there is no significant increase in purchases from consumers In cluster I the product can be concluded to have increased by 70% while cluster II nuvo soap decreased by 30%. The results of the implementation of the K-Means Algorithm contribute to the understanding of consumer purchasing patterns towards certain products in addition to optimizing marketing strategies and inventory management.