CRISP-DM and the ARC Framework

For your initial discussion post, recall the Ordenes, Theodoulidis, Burton, Gruber, and Zaki article from this unit’s studies, and evaluate the authors’ ARC framework and its contribution to the literature versus the various methods identified in Table 1.

Ordenes, F. V., Theodoulidis, B., Burton, J., Gruber, T., & Zaki, M. (2014). Analyzing customer experience feedback using text mining: A linguistics-based approach. Journal of Service Research, 17(3), 278–295.

Don't use plagiarized sources. Get Your Custom Essay on
CRISP-DM and the ARC Framework
Just from $13/Page
Order Essay

Analyzing Customer Experience Feedback
Using Text Mining: A Linguistics-Based
Francisco Villarroel Ordenes1
, Babis Theodoulidis2
, Jamie Burton2
Thorsten Gruber3
, and Mohamed Zaki4
Complexity surrounding the holistic nature of customer experience has made measuring customer perceptions of interactive
service experiences challenging. At the same time, advances in technology and changes in methods for collecting explicit customer
feedback are generating increasing volumes of unstructured textual data, making it difficult for managers to analyze and interpret
this information. Consequently, text mining, a method enabling automatic extraction of information from textual data, is gaining in
popularity. However, this method has performed below expectations in terms of depth of analysis of customer experience feedback and accuracy. In this study, we advance linguistics-based text mining modeling to inform the process of developing an
improved framework. The proposed framework incorporates important elements of customer experience, service methodologies, and theories such as cocreation processes, interactions, and context. This more holistic approach for analyzing feedback
facilitates a deeper analysis of customer feedback experiences, by encompassing three value creation elements: activities,
resources, and context (ARC). Empirical results show that the ARC framework facilitates the development of a text mining model
for analysis of customer textual feedback that enables companies to assess the impact of interactive service processes on customer experiences. The proposed text mining model shows high accuracy levels and provides flexibility through training. As such,
it can evolve to account for changing contexts over time and be deployed across different (service) business domains; we term it
an ‘‘open learning’’ model. The ability to timely assess customer experience feedback represents a prerequisite for successful
cocreation processes in a service environment.
activities, resources, context, customer feedback, text mining, case study, value cocreation, customer experience
Collecting and analyzing customer feedback is important because
it allows organizations to learn in a continuous manner to adapt
their offerings to customer preferences (Sun and Li 2011).
Increasingly, customers use multiple communication channels
to provide feedback, making it cumbersome for organizations
to develop efficient and effective processes to collect and analyze
all the information. Companies that can manage customer feedback data regularly are, on average, 5% more productive and
6% more profitable than their competitors (McAfee and Brynjolfsson 2012). Technological advances have expanded the
choice of channels available for companies to collect customer
feedback. Structured feedback, such as quantitative surveys, is
now increasingly considered alongside unstructured feedback,
such as telephone calls, e-mails, and social media, in which customers describe their experiences more freely (Witell et al. 2011).
The latter contain information in a verbatim format and are characterized by a higher level of detail, describing the most critical
elements for customer experience (Ziegler, Skubacz, and Viermetz 2008). However, the variety of content and the time and
resources required to analyze textual feedback create barriers to
uncovering meaning in these data.
Marketing departments have become increasingly aware of
the importance of textual feedback and have used manual or
automatic approaches to analyze this information. Companies
that run analysis on a manual basis can gain a deeper understanding of customer feedback, but if their analysis lacks
1Marketing and Supply Chain Management Department, School of Business and
Economics, Maastricht University, Maastricht, Netherlands
2Manchester Business School, The University of Manchester, United Kingdom
3 School of Business and Economics, Loughborough University, Loughborough,
United Kingdom
4 Institute for Manufacturing, University of Cambridge, Cambridge, United
Corresponding Author:
Francisco Villarroel Ordenes, Marketing and Supply Chain Management
Department, School of Business and Economics, Maastricht University,
Tongersestraat 53, 6211 LM Maastricht, Netherlands.
Journal of Service Research
2014, Vol. 17(3) 278-295
ª The Author(s) 2014
Reprints and permission:
DOI: 10.1177/1094670514524625
procedural models, they tend to be inconsistent when reviewing large quantities of data (Ziegler, Skubacz, and Viermetz
2008). At the same time, companies that have adopted automated analysis of textual feedback (e.g., text mining) have
failed to realize their expectations in using this method (Fenn
and LeHong 2012). Specifically, a lack of accuracy in predicting customer sentiments (positive/negative/neutral) and the
inflexibly of methods in adapting to different business
domains represent the main causes of this disillusionment
(Fenn and LeHong 2012). The deployment of text mining
models has clear managerial implications, including the availability of accurate and timely information, for better informed
decision making.
Academic research on these problems remains scarce. Customer feedback analysis using text mining has largely focused
on developing more accurate models for automatically predicting the sentiment embedded within text (Gra¨bner et al. 2012).
The majority of these studies have emphasized how different
text mining methods (linguistic and nonlinguistic; Taboada
et al. 2011) contribute to better predicting the overall sentiment
in a customer review. Despite the importance of identifying
sentiments, more specific information is contained in textual
customer feedback. Critical elements of an organization’s
offering that trigger sentiment evaluations have largely been
ignored. For example, in the customer feedback, ‘‘Extremely
friendly and helpful staff. Disappointed that the bar wasn’t
open full time! A lovely, quiet, and relaxing place to stay!’’ the
first sentence is positive, the second is negative, the third is
positive again, and more than three different aspects of a service are covered. With sentiment analysis, the focus would rest
solely on the final sentiment output, but with multiple emotions
in the comment, this is uncertain, providing little guidance for
managerial response. However, if the focus of the automated
analysis were on the service components and related customer
interactions (e.g., with the staff, the bar, and the atmosphere),
sentiment outcomes would offer greater value for decisionmaking purposes.
Service literature has long recognized that customer evaluations of service experience are an outcome of the interactions
among companies, related systems, processes, employees, and
customers in a service context (Bitner et al. 1997). These elements, recurrent in areas such as service blueprinting, service
encounters, and service quality (Fisk, Brown, and Bitner 1993),
indicate that services consist of activities between customer and
company in a cocreation process (Payne, Storbacka, and Frow
2007). Despite their managerial and theoretical relevance, these
elements are largely absent in operationalizations of automated
models to analyze customer feedback (Taboada et al. 2011;
Ur-Rahman and Harding 2011). In particular, there is no evidence
of a framework that can aggregate relevant aspects of the service
process and operationalize them in a text mining model to help
companies analyze customer textual feedback and conversations.
At the same time, research has increasingly recognized the need to
understand the holistic nature of customer experiences in order to
design improved service systems (Patrı´cio et al. 2011; Verhoef
et al. 2009). We contend that insights into customer experiences
are more likely to be identified in rich customer comments than
through company-designed surveys, and thus the lack of text mining focusing on customer service experience is a significant oversight. In the context of these omissions, this article makes the
following contributions.
First, the study fills a gap in the text mining literature by
proposing a framework that provides a holistic approach for
analyzing customer feedback by accounting for three key components of the value cocreation process: activities, resources,
and context (ARC). Through this framework, customer textual
feedback not only can be classified as positive or negative in
terms of a specific attribute (i.e., friendliness; Feldman 2013)
but also can be mapped onto a chain of activities and resources
that describes how value is cocreated in a particular context
(Gro¨nroos and Voima 2013). The ARC framework departs
from simple output-based analysis (e.g., attribute assessment
and sentiment analysis) by offering a new processes and interactions approach that can be applied with linguistic text mining, which captures key elements of service in customer
feedback, to better represent how value is cocreated.
Second, we contribute to text mining research by expanding
the scope of automation from existing approaches to more flexible models. Implementing the ARC framework, we develop
and propose a linguistics-based text mining model to extract
detailed information about customer experiences from textual
feedback. We explain how the ARC framework guides the utilization of linguistics-based text mining features in developing
a text mining model for customer feedback analysis. This process enables the development of a model that captures customer context (personal and situational), activities and
resources (pertaining to both customer and company), and the
sentiment associated with these constructs. Thus, a more complete and holistic picture of customers’ interactions in service
encounters is automatically captured. Moreover, the text mining model can be enriched over time with evolving customer
terminology about changing service resources and activities
and can also be adapted and applied in different (service) business domains. We call this characteristic an ‘‘open learning’’
model to describe a text mining model that can be enhanced
over time and adapted.
Conceptual Framework
Development of a Modeling Framework
Service literature has considered the process nature of services,
especially the effect of customer-company interactions on customers’ evaluations of a service experience (Fisk, Brown, and
Bitner 1993). According to Bitner, Brown, and Meuter (2000),
adequate mapping of these interactions facilitates the identification of encounters, which Shostack (1985, p. 243) defines as ‘‘a
period of time during which a consumer directly interacts with
the service.’’ These encounters are critical for customers’ service
evaluations in terms of satisfaction (Bitner, Booms, and Mohr
1994; De Ruyter et al. 1997), service quality (Parasuraman,
Zeithaml, and Berry 1994), and customer loyalty (Gremler and


Calculate the price of your paper

Total price:$26
Our features

We've got everything to become your favourite writing service

Need a better grade?
We've got you covered.

Order your paper
Live Chat+1(978) 822-0999EmailWhatsApp