Product inventory in retail companies is crucial because a product is a major component in the business and is a thing that is traded to meet the needs of consumers. In business, entrepreneurs certainly expect optimal profits that can be achieved by increasing the value of product sales. As a supporting factor in the success of a business, stock of products must always be available to meet the needs of consumers. The problems that often occur in retail companies are a product stock shortage and product stock buildup that can result in the damage of products because of being kept in the warehouse for a long time and can hinder the capital mobility. A product stock prediction system is needed to reduce the problems and can predict the amount of product stock in the future so that product availability can be controlled and the company can achieve the expected goals. This paper aims to provide a web-based product stock prediction system with a Single Exponential Smoothing and Moving Average method to determine the amount of product stock that must be available in the future. The comparison results of the two methods implemented in the prediction system show the Single Exponential Smoothing method with an alpha value of 0.5, superior to the Moving Average method with the results of the prediction error accuracy value in the Aqua 19 L product, namely MAD = -0.50, MSE = 3.02 and MAPE =11.38.