Minimarket idola on a daily basis there are many sales transactions, so that the data stored in the database is very large. The data can be used as much useful information for the owner of a minimarket in policy making. To explore the data that is used a lot of data mining technique. Data mining uses data analysis to discover patterns and relationships in data that may be used to make accurate predictions.In this research, data mining is used to forecast the sales of goods in Minimarket Idoal. Forecasting the future based on measuring the value of the patterns in the data collection. To perform sales forecasting in the future to use the method of time series. Forecasting time series data to predict what will happen based on past historical data.Time series methods for forecasting sales in the calculation Minimarket Idola using exponential smoothing and moving average. Of the count sought the MAD (Mean Absolute Deviation) or forecasting errors. Where MAD is the smallest value of the calculation of exponential smoothing and moving average is the result of forecasting with a small error. Forecasting results will not always be appropriate because the market demand influenced by several factors. But it does not mean that the forecast is made useless.