Identifying target audience on enterprise social network
| DOI | https://doi.org/10.1108/IMDS-01-2018-0007 |
| Published date | 04 February 2019 |
| Date | 04 February 2019 |
| Pages | 111-128 |
| Author | Jianhong Luo,Xuwei Pan,Shixiong Wang,Yujing Huang |
| Subject Matter | Information & knowledge management,Information systems,Data management systems,Knowledge management,Knowledge sharing,Management science & operations,Supply chain management,Supply chain information systems,Logistics,Quality management/systems |
Identifying target audience on
enterprise social network
Jianhong Luo, Xuwei Pan, Shixiong Wang and Yujing Huang
Department of Management Science and Engineering,
Zhejiang Sci-Tech University, Hangzhou, China
Abstract
Purpose –Delivering messages and information to potentially interested users is one of the distinguishing
applications of online enterprise social network (ESN). The purpose of this paper is to provide insights to
better understand the repost preferences of users and provide personalized information service in enterprise
social media marketing.
Design/methodology/approach –It is accomplished by constructing a target audience identification
framework. Repost preference latent Dirichlet allocation (RPLDA) topic model topic model is proposed to
understand the mass user online repost preferences toward different contents. A topic-oriented preference
metric is proposed to measure the preference degree of individual users. And the function of reposting
forecasting is formulated to identify target audience.
Findings –The empirical research shows the following: a total of 20 percent of the repost users in ESN
represent the key active users who are particularly interested in the latent topic of messagesin ESN and fits
Pareto distribution; and the target audienceidentification framework can successfully identify different target
key users for messages with different latent topics.
Practical implications –The findingsshould motivate marketingmanagers to improve enterprisebrand by
identifyingkey target audience inESN and marketing in a way that truthfullyreflects personalizedpreferences.
Originality/value –This study runs counter to most current business practices, which tend to use simple
popularity to seek important users. Adaptively and dynamically identifying target audience appears to have
considerable potential, especially in the rapidly growing area of enterprise social media information service.
Keywords Social network, Marketing, Enterprise, Target audience, Repost preference, Topic model
Paper type Research paper
1. Introduction
Online social networks (OSNs) have become popular platforms for discovering real-time
information on theweb. An increasing number of enterpriseshave utilized social media, such
as Sina Weibo, Twitter and Facebook, as their communication tool with consumers, thereby
forming their enterprise social network (ESN). ESNs are online platforms used in a business
context, which facilitate light-weight communication via short messages and provide
functions to find, connect and interact with users (Aoun and Vatanasakdakul, 2012). Social
media has become an important area of exploration for researchers interested in online
technologies and their impacts on consumers and businesses value creation (Ketonen-Oksi
et al., 2016). Currently, there are manypopular research topics in OSNs, such as modeling the
networking structures, dynamic evolution and formed human relationships. The research
problems mainly focused on characterizing the information diffusion, which depends on the
network structures and the scale of nodes.However, these studies were mainly carried out on
typical large social networks and only a few focused on ESN.
In ESN, the content of messages is arguably a main influence on a user’s judgment of
repost or retweet. When a user views the micro-blog message, the content is a critical
determinant of whether the user will repost to share or neglect the micro-blog message.
The analysis of Zhang et al. (2015) shows that marketing micro-blogs with a high percentage
of retweets are directly retweeted from the original micro-blog, that is, the direct retweets are
dominant in the retweet network. Therefore, more users must be encouraged to repost a
message to improve the dissemination process and to reach more potential customers in the
social network (Luo et al., 2017). Appropriate messages must be posted to attract more
Industrial Management & Data
Systems
Vol. 119 No. 1, 2019
pp. 111-128
© Emerald PublishingLimited
0263-5577
DOI 10.1108/IMDS-01-2018-0007
Received 5 January 2018
Revised 28 April 2018
Accepted 6 May 2018
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0263-5577.htm
111
Enterprise
social network
followers to repost to reach more direct and indirect followers, because the early repost
reaction of users depends on their interest in the content of the messages. However, evidence
shows that information overload is present: users tend to search through hundreds of
messages to find those that are interesting to reply to or retweet (Comarela et al., 2012).
Prior studies have shown that the capability to deliver content and information to
potentially interested users is one of the distinguishing applications of OSN (Newman and Park,
2003). If one user is only concerned with a product of the enterprise, and a few micro-blogs
posted to the user have contents with jokes or short stories that disturbed his purchase
intention, then the user may likely continue to search for other information. The proliferation of
customer data has made it increasingly possible for companies to personalize their offerings
(Rust and Huang, 2014). Customers prefer offerings that are a better fit to their needs, and
repeatedly adapting to the observed behavior of customers improves personalization
performance (Chung et al., 2016). Therefore, studying the reposting behavior of users is valuable
to improve personalizing services on ESN.
We consider mining user repost preference through micro-blog text analysis to explore the
reposting behavior of users. The textual contents of an enterprise’s micro-blog texts on social
media undeniably contain rich information. However, analyzing the textual content of the
micro-blog texts is challenging because they are written in natural language. The latent Dirichlet
allocation (LDA) (Blei et al., 2003) topic model is a popular, unsupervised statistical model used to
automatically discover topics from a corpus of text documents. Using the LDA topic model, we
can learn the latent subtopics in an enterprise’s micro-blog texts. Based on the interpretable
meaning of the latent subtopics and the repost behavior, the topic-oriented preference metric was
proposed to calculate the degree of the user’s potential interest toward the content.
Topic-oriented preference measure in marketing is pursued to bring relevant content
tailored to the interests of the right user. We termed the right user as target audience that
would repost the interested micro-blog message brought by the enterprise. By reposting, the
micro-blog message will be rapidly diffused on ESN. Through such process, the diffusion
effect will increase, and the enterprise will succeed more in marketing. The time, energy and
money should be better invested in the target audience. Social media data comprise a host of
information that might serve as predictors (Schoen et al., 2013). Therefore, this paper studies
a target audience identification model for targeting the exact user that can help an
enterprise to extend its marketing on ESN.
The objective of this research is to better understand the repost preferences of users
and identify target audience on ESN. Therefore, we developed an approach that flexibly
captured different preferences of the user for different contents to adaptively and
dynamically identify target audience. We show that the target audience can be adaptively
and dynamically identified on ESN using an experimental study. These studies also suggest
possible marketing strategies to bring the preferable micro-blog content to the target
audience. Our results also provide a set of useful summaries on theoretical and managerial
implications for ESN marketing. Accordingly, the results of this research are highly relevant
for internet marketing practitioners.
The remainder of this paper is organized as follows. First, we outline previous works
related to this paper. Second, we describe the target audience identification framework.
Third, we present the case study results and analysis. Fourth, we discuss the results and
implications. Finally, we offer concluding remarks and avenues for future work.
2. Related work
This section is related to the growing literature focusing on understanding user repost
behavior and its marketing implications. We focus on the three relevant aspects to our
research: internal and external use of ESN; information diffusion in OSN; and influence of
information diffusion for marketing.
112
IMDS
119,1
To continue reading
Request your trial