•  
  •  
 

Abstract

Many e-stores offer product recommendation services to increase sales and customers’ satisfaction. However, previous studies indicate that the performance of recommendation systems may be influenced by consumers’ perception of the recommended products, which may be affected by how the product was recommended. The purpose of this study is to investigate the effects of product type and recommendation approach on consumers’ intention to purchase recommended products. A laboratory experiment was conducted to collect empirical data and ANCOVA was adopted to test our research hypotheses. We found significant interaction effects between recommendation approach and product type on consumers’ intention to buy the recommended products and their disconfirmation of the recommended products. Before experiencing the products, consumers’ purchase intention was significantly higher for search goods when they were recommended by the Top-N approach than when they were recommended by the collaborative filtering approach. After experiencing the products, however, consumers’ disconfirmation was significantly higher for experience goods when they were recommended by the Top-N approach than when they were recommended by the collaborative-filtering approach. Furthermore, the disconfirmation of experience goods was significantly higher than that of search goods when they were recommended by the Top-N approach. Our findings provide valuable insights into choosing product recommendation approaches for different types of products.

Share

COinS