On the role of information in digital marketing

This dissertation is an empirical examination of the role that information available online plays in transforming the field of digital marketing. The first chapter examines purchase behavior in the context of cashback shopping - a relatively young but increasingly popular form of sales promotion online. In general, the literature in marketing recognizes two reasons to delay paying a discount: consumers often fail to claim the money promised to them, and tying the refund to a future purchase event. However, in cashback shopping payments are automatic and unconditional, which brings into question the usefulness of the practice. An analysis of panel data from a large cashback company reveals that, aside from the predictable positive effect of cashback offers on demand, cashback payments induce and increase further spending through its website. In the second chapter, I explore decision making in crowdfunding - an innovative and increasingly popular model to raise funds on the internet. A major marketing challenge faced by crowdfunding websites is to decide which projects to display on their websites to which potential donors so as to increase the likelihood of success of crowdfunding projects. I develop a structural model of donors' contributions to cause-based crowdfunding projects, accounting for their uncertainty about whether a project will reach its target and the effect of the donor's uncertainty over project completion on the amount that they contribute. I estimate my model on data obtained from the US educational crowdfunding website Donors Choose. My results demonstrate that donors' evaluation of uncertainty of project completion is an important driver of whether and how much a donor contributes to a project. I demonstrate that customizing the projects displayed to potential donors based on their preferences leads to better crowdfunding outcomes. In the third chapter, I focus on product reviews which are a powerful source of information for consumers. In this research, I investigate how the position of a review on a product webpage affects its effect on consumers' purchase likelihood. Utilizing the variation in review positions generated as newer reviews are added on top of older ones, I specify an empirical strategy to quantify the drop in effectiveness of reviews as their position drops. My results from analysis of a panel data of reviews shows that reviews further down on the product page receive are less effective in influencing purchase decision compared to reviews at the top of the pag