Combining Different Recommendation Systems: Is it Valuable to Present Substitute and Complementary Recommendations Simultaneously?

In e-commerce, different types of recommendation systems are often used simultaneously on the same site. While their individual effects on product views and sales are known, the effects of combining them are unclear. Drawing on research on consumer preference formation and information foraging theory, we propose three hypotheses to explain how the combination of substitute and complementary recommendations affects product views and sales at the early and the late stage of the purchase process. We propose a randomized between-subjects field experiment in an online fashion store to empirically test our hypotheses. Our research contributes to the literature on the effectiveness of recommendation systems in e-commerce by providing a better understanding of how the interaction of different types of recommendation systems influences product views and sales at different purchase stages. The results of our ongoing research have important implications for e-commerce platforms on how to effectively employ different recommendation systems.