This research analyses the effect of numerous factors that allow E-commerce platforms to obtain previous knowledge about customer purchase tendencies. This research improves customer behaviour, product discoverability, and warehouse maintenance. Demanding product categories assist maintain the warehouse and generate sales ideas by offering an appropriate discount to attract consumers and promote same-day delivery. The system uses an e-commerce company database to store consumer purchase information. Our model analyses the data to classify customers and commodities. Our study includes descriptive, predictive, and prescriptive analytics to anticipate e-commerce sales. The descriptive study includes data cleaning, preprocessing, and visualization to analyze gender, city tier, spending money, age, city, marital status, financial position, brand cost, and product categories these are the attributes of dataset. To advance to predictive analytics, machine learning techniques such as naive Bayes classification, support vector classification, logistical regression, decision tree, KNN classification, and random forest classification are applied to the dataset to anticipate product sales. This proposed research reduces e-waste and promotes sustainable development by anticipating product sales