Abstract

Although the value of online product recommendations is well established, the relative value of retargeted recommendations (based on individual users’ past viewed products) over crowd-based generic recommendations (based on co-views and co-purchases of the crowd) is unclear. Moreover, there is little guidance on how the relative value would vary at different stages of the consumer purchase funnel. We conduct a field experiment on the website of a mid-size retailer in the US to examine the effect of retargeted and generic product recommendations on the sales of recommended products in early versus later stages of the purchase funnel. We examine the effect of these recommendations on the two components of recommended product sales – the number of daily impressions and conversion rates (purchase conditional on impression). We find that (1) generic recommendations only increase the conversion rate in the early purchase funnel stage, but retargeted recommendations do not affect conversion rates; and (2) both retargeted and generic recommendations result in a higher number of impressions of recommended products. Overall, we find that it is beneficial to show retargeted (generic) recommendations to visitors in the late (early) purchase funnel stage. We conduct counterfactual simulations to show that the application of our findings can result in up to five percent increase in the sales of recommended products. Our findings have implications on the optimal design of product recommendation systems.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call