PERSONALIZED CONTENT RECOMMENDATIONS AND USER SATISFACTION 45
Journal of Management Information Systems / Winter 2006–7, Vol. 23, No. 3, pp. 45–70.
© 2007 M.E. Sharpe, Inc.
0742–1222 / 2007 $9.50 + 0.00.
DOI 10.2753/MIS0742-1222230303
Personalized Content Recommendation
and User Satisfaction: Theoretical
Synthesis and Empirical Findings
TING-PENG LIANG, HUNG-JEN LAI, AND YI-CHENG KU
TING-PENG LIANG is National Chair Professor in Information Systems and Director of
the Electronic Commerce Research Center of National Sun Yat-sen University in Tai-
wan. He received his Ph.D. from the Wharton School of the University of Pennsylva-
nia. Prior to joining his current university in 1993, he taught at the University of Illinois
and Purdue University. At National Sun Yat-sen University, he has served as the Dean
of the College of Management (1994–97) and University Provost (1999–2002). His
research has been published in Journal of Management Information Systems, MIS
Quarterly, Management Science, Operations Research, Decision Support Systems,
Information and Management, and many other academic journals. Professor Liang’s
primary research interests include electronic commerce, knowledge management, in-
telligent systems, and strategic applications of information systems. He has served on
the editorial boards of several journals, including Decision Support Systems, Interna-
tional Journal of Electronic Commerce, Industrial Management and Data Systems,
and Electronic Commerce Research and Applications, among others. He is a fellow of
the AIS and a member of ACM, INFORMS, and IEEE Computer Society.
HUNG-JEN LAI is an Assistant Professor in the Department of Information Manage-
ment at the Naval Academy, R.O.C., Kaohsiung, Taiwan. He received his Ph.D. from
National Sun Yat-sen University. He is a recipient of the Outstanding Teacher Award
from the Naval Academy, R.O.C. His master’s thesis received the Long-Term Thesis
Award from Acer Company. Professor Lai’s research interests include electronic com-
merce, knowledge management, intelligent systems, and medical information systems.
His works have been published in several proceedings such as Hawaii International
Conference on System Sciences and journals such as Information & Management.
YI-CHENG KU is an Assistant Professor in the Department of Computer Science and
Information Management, Providence University, Taiwan. He received his Ph.D. in
Information Management from National Sun Yat-sen University. His research inter-
ests include recommendation systems, innovation adoption and diffusion, and knowl-
edge management. His papers have been published in Decision Support Systems,
Electronic Commerce Research, and various conference proceedings.
ABSTRACT: Personalized services are increasingly popular in the Internet world. This
study identifies theories related to the use of personalized content services and their
effect on user satisfaction. Three major theories have been identified—information
overload, uses and gratifications, and user involvement. The information overload
theory implies that user satisfaction increases when the recommended content fits
user interests (i.e., the recommendation accuracy increases). The uses and gratifica-

46 LIANG, LAI, AND KU
tions theory indicates that motivations for information access affect user satisfac-
tion. The user involvement theory implies that users prefer content recommended by
a process in which they have explicit involvement. In this research, a research model
was proposed to integrate these theories and two experiments were conducted to
examine the theoretical relationships. Our findings indicate that information over-
load and uses and gratifications are two major theories for explaining user satisfac-
tion with personalized services. Personalized services can reduce information overload
and, hence, increase user satisfaction, but their effects may be moderated by the
motivation for information access. The effect is stronger for users whose motivation
is in searching for a specific target. This implies that content recommendation would
be more useful for knowledge management systems, where users are often looking
for specific knowledge, rather than for general purpose Web sites, whose customers
often come for scanning. Explicit user involvement in the personalization process
may affect a user’s perception of customization, but has no significant effect on
overall satisfaction.
KEY WORDS AND PHRASES: content recommendation, personalization, recommenda-
tion systems, user satisfaction.
THE RAPID PROPAGATION OF THE INTERNET, along with the evolution of information
technologies (IT), has changed the way firms are adapting to changing customer needs.
For physical products (e.g., computers and televisions), mass customization and fast
response to dynamic market needs have become critical to remaining competitive.
For digital products (e.g., news services and other Internet content providers [ICPs]),
personalized services that provide tailored content to different clients, based on their
interests, become feasible and necessary. The large amount of transactional data, col-
lected from the use of Internet-enabled information systems, allows a company to
understand customer needs and integrate the discovered knowledge into its product
design and marketing plans. Existing literature has proved that customized sellers can
charge more for customized products (e.g., [13]).
The Internet is an excellent platform for content providers to tailor their products
based on customer preference. This is particularly true for online news services and
knowledge management. For Internet news Web sites, most readers are only inter-
ested in certain types of news among the large number of reports. Some may be inter-
ested in political news, while others are interested in stock market movements.
Therefore, providing news reports that meet a reader’s interests can save time and
effort. As a result, personalized services have been adopted by many news Web sites,
including crayon.net and Google News. Similarly, it would be useful if a personalized
recommendation system could find relevant documents in the knowledge repository
for users when they use a knowledge management system to solve a specific problem.
Although it is intuitive that personalization could add value to content providers,
existing literature has not provided adequate theoretical and empirical evidence to
tell whether the user really likes personalized services. Therefore, it would be useful