Three essays on the effects of classification strategies for online advertising, waste management, and market transactions

Abstract

This dissertation investigates how appropriate classification of phenomena or circumstances enhances the comprehension of human decision-making process and behavior across various domains by employing diverse methodologies. In the first study, individuals’ ad clicks in the online advertising context are classified into two categories depending on their post-click actions, followed by an analysis of differential impacts on purchase. Furthermore, we delve into the determinants for the occurrence of click categories and validate the superiority of the customer targeting strategy based on proposed click classification criteria compared to the conventional click-based strategy. The second study investigates the optimal waste classification level for public recycling bins in order to enhance recycling rates. Empirical validation from field experiments verifies that intermediate level of waste classification induces proper recycling most effectively, while also highlighting the significant effects of installation location and visual guides. Moreover, we introduce a concept of entropy to estimate the economic values generated by recycling bins, incorporating it into the calculation of separation cost. In the third study, we develop a comprehensive model for decision-making processes for customer purchases within a multichannel environment. The study scrutinizes channel usage characteristics and empirically examines the priority between channels and brands during product purchases. We delve into the alterations of aforementioned elements in times of exceptional situations such as COVID-19, and deliberate on strategic orientation for businesses.