The conversion funnel is a model describing the stages consumers go through in their journey towards a purchase. A large body of literature has focused on using observable search patterns to identify consumers’ hidden purchasing stages and to estimate their likelihood of conversion. We add to this literature by using the diversity of the searches that a customer engages in while browsing an e-commerce website as a means of estimating conversion likelihood. We develop novel measures of diversity and show their usefulness in predicting conversion in the context of a medium-size tourism website used for comparing and booking flights. We
find that incorporating diversity measures improves model recall and F-score by factors of 2.61 and 2.46, respectively, as compared with a baseline model. We show that as search-diversity decreases, conversion likelihood increases, indicating that within-visit search-diversity may serve as a proxy for a consumer’s stage in the funnel.
Goldstein, Anat and Oestreicher-Singer, Gal and Barzilay, Ohad, Deep into the Funnel? Predicting Online Conversion Using Search Diversity (May 4, 2017). Available at SSRN: https://ssrn.com/abstract=2962960