The effect of competing product network for sale performance in internet storefronts

A glut of sponsored information from the Internet makes consumers hard to make a purchase decision. It enables to share their opinions and experience in electronic commerce. Moreover, recommendation network which not only include co-view network, co-purchase network but also provide similar product recommendation network to help customers identify goods in which they might have an interest. Through such recommendation mechanisms, product pages have hyperlinks to other product pages, which can use to create product networks. In this paper, we empirically investigate the impact of the similar product network, co-view product network and the consumer feedbacks on product sales based on data about cell phones which is gathered from Amazon.com. Furthermore, we differentiate the product sale impact of two different recommendation mechanisms. Our results indicate that the number of reviews and customer rating are positively associated with product sales. On the other hand, in similar product network, the degree centralities of the products are positively related to product sales. However, the higher of eigenvector centrality of the competing product has a negative association with product sales. In co-view product network, interestingly and surprisingly, none of the node centrality variables are significant. Moreover, the direction of degree is considered. Typically, we are interested in in-degree, since in-links are given by other nodes in the network. The result suggests an expected positive relation with the focal products sales. Comparing with co-view product network, similar product network has a more influential relation with product sale. Therefore, our study provides theoretical contributions in term of the economic impact of product recommendation network on product sales. Our paper shows an evidence quantifying the role of the product network position in electronic commerce sites. Additionally, we also provide worthy implications for product marketing and recommendation network design.Maste