Using In-Store Analytics to Bridge the Purchase Intention and Behavior Gap at the Point of Sale
(A2023-114235)

AUTHORS

Anna Ulrichshofer, Technische Hochschule Ingolstadt; Michael Jungbluth, Technische Hochschule Ingolstadt; Christian Schlereth, WHU - Otto Beisheim School of Management

ABSTRACT

Retailers nowadays rely more and more on in-store analytics at the point of sale and have started installing technology that enables tracking customers during their shopping trips. Such data tracks the time spent in the store, which categories were visited for how long, and what had been purchased. Some retailers spend additional effort and complement shopping trip data with survey data to capture purchase intentions. With this research, we propose a hidden Markov model to condense the rich shopping trip data. We also examine the added value of the condensed data to inform unplanned purchases and purchase resignation, i.e., customers not making a purchase, even though they had purchase intentions when entering the store. Finally, we contrast its value to the added value of knowledge about customer characteristics and other behavioral information. Omitting in-store tracking variables in our field study degrades the model fit for unplanned purchases by 50.28% and purchase resignation by 14.58%. At the same time, omitting shopping motivation and customer characteristics account for just a 1.17 – 2.84% decrease in model fit.