Dynamic hedonic and utilitarian segmentation based on individual customer purchase patterns

Warning The system is temporarily closed to updates for reporting purpose.

Küçüksarı, Zeynep (2021) Dynamic hedonic and utilitarian segmentation based on individual customer purchase patterns. [Thesis]

[thumbnail of 10336534.pdf] PDF
10336534.pdf

Download (3MB)

Abstract

In the management and psychology literature, the consumers’ motivations to make purchases have long been studied under the dichotomic perspective of hedonic vs utilitarian decisions. This perspective is proved to be relevant either to understand why people buy and to help companies to frame their operations strategies to optimize their sales efforts and to maximize customer satisfaction. In this paper, we analyze supermarket transaction data from Brazil over the course of one years to understand and identify utilitarian versus hedonic consumer behavior in a supermarket context. While current literature studies the same notion mainly for a wider set of general shopping categories, we focus on in-supermarket purchases to understand when and how consumers are inclined to make purchases that could be considered hedonic even in a supermarket setting. We develop and propose measures to quantify depth and breadth of purchases along several dimensions including brand/no brand, purchase value, purchase quantity or amount. As the definition of hedonic vs. utilitarian may change from person to person, and in the absence of ground truth, we propose an unsupervised approach to identify outlier transactions that would likely be considered hedonic purchases. A closer examination of selected customers and their transaction sets suggests that our approach produces realistic scenarios under a variety of dimensions and scenarios considered. Our approach brings new theoretical perspectives to advance the hedonic and utilitarian literature in management/operations research.
Item Type: Thesis
Uncontrolled Keywords: Hedonic and utilitarian shopping behavior. -- Behavior Analysis. -- Clustering. -- Unsupervised Learning. -- Outlier Detection. -- Hedonik ve faydacı tüketim davranısları. -- Davranıs analizi. -- Kümeleme. -- Gözetimsiz ögrenme. -- Anormallik tespiti.
Subjects: T Technology > T Technology (General) > T055.4-60.8 Industrial engineering. Management engineering
Divisions: Faculty of Engineering and Natural Sciences > Academic programs > Industrial Engineering
Faculty of Engineering and Natural Sciences
Depositing User: IC-Cataloging
Date Deposited: 26 Oct 2021 13:02
Last Modified: 26 Apr 2022 10:39
URI: https://research.sabanciuniv.edu/id/eprint/42519

Actions (login required)

View Item
View Item