The existing studies on selling strategies did not provide operational selling recommendations, did not distinguish the auction heterogeneity, or did not integrate multiple objectives of sellers into selling strategies. In this study, we incorporate seller and product heterogeneity into our analytical framework, and implement data mining analysis on four auction segments. We use classification and regression tree (CART) to identify the critical factors along with their sequences for auction success and prices. We find different determinants for auction success and prices among these four auction segments.  The classification and regress trees provide operational options for sellers to build the most effective selling strategies. We propose that, by using expected auction prices with the classification and regression trees, sellers can integrate auction success with prices in their selling strategies. Overall, this study contributes to the literature by providing an innovative prospectus for effective selling recommendations, which are important for smooth growth of the electronic marketplaces.