Developing a Robust Pricing Model for Second-Hand Goods in E-commerce: A Data-Driven Approach

Abstract

Most existing pricing tools for e-commerce platforms treat used goods in the same manner as new goods, failing to account for the specific factors that contribute to the resale value of those goods. For peer-to-peer marketplaces, the accuracy of the pricing of goods is important to the trust between buyers and sellers on the platform. To address these issues, we developed a pricing model for a student-run peer-to-peer marketplace that accounted for the condition of the products and the annual depreciation rate of those products. Compared to other pricing models for peer-to-peer marketplaces, our model was found to produce valuations for used goods that were more consistent and sensitive to the condition of the products. Also, usability tests of the platform with students who used it revealed that 84.8% of those users found the platform to be easy to use, even with the added complexity of the pricing model. Overall, then, the pricing model for used goods developed for this project is more flexible than existing pricing models for peer-to-peer marketplaces and can help to improve the satisfaction of both buyers and sellers on the platform. Two potential areas for future research include incorporating external market data into the model and utilizing machine learning techniques to improve the accuracy of the model's pricing predictions. Also, another area of research could involve comparing the accuracy of the pricing model to that of existing pricing models for peer-to-peer marketplaces and establishing a standard for evaluating improvements to such models.