The goal of most businesses is to make profits. This is usually achieved by making more sales than expenses. Sales are the lifeblood of each, and every organization and sales forecasting play a critical role in conducting every business. Better forecasting helps to build and enhance business plans by growing awareness about the marketplace. This thesis discusses the question of projecting or forecasting big mart sales of an item on the potential demand of customers in various large mart stores across different locations and items based on the previous record. The thesis is aimed to provide a more accurate predictive model in the prediction of outlet item sales in Bigmart. This study seeks to compare the performance of the predictive models developed using machine learning and deep learning techniques. The predictive models were developed using the various algorithms, Support Vector Regression (SVR), Linear Regression (LR), Decision Tree Regression (DTr), Random Forest Regression (RFr) and the Artificial Neural Network (ANN). The results show the Mean Square Error and the Root Mean Square Error of the predictive models developed by the algorithms. The SVR, LR, DTr, RFr and the ANN had an MSE score of 1438556.03, 703286.11, 2600437.07, 1138681.59, and 691652.94, and RMSE score of 1199.40, 838.62, 1612.59, 1067.09 and 831.66 respectively. This study found the artificial neural network to performs better than the other algorithms in the prediction of outlet item sales in big marts. This thesis would aid big marts’ management on their decision as to the maximize items and outlets placement in other to provide a better customer experience which would, in turn, lead to increase in sales which are most likely to translate to increase in revenue and cause expansion of in the big mart business.