Journal of Retailing and Consumer Services 21 (2014) 889896

Contents lists available at ScienceDirect

Journal of Retailing and Consumer Services

journal homepage: www.elsevier.com/locate/jretconser

A consumer typology based on e-service quality and e-satisfaction

Grgory Bressolles a,n, Franois Durrieu a,1, Sylvain Senecal b,2
a
b

KEDGE Business SchoolBordeaux, 680 Cours de la Libration, 33405 Talence Cedex, France
HEC Montral, 3000 Chemin de la Cte-Sainte-Catherine, Montral, QC, Canada H3T 2A7

art ic l e i nf o

a b s t r a c t

Article history:
Received 15 November 2013
Received in revised form
18 July 2014
Accepted 19 July 2014
Available online 26 August 2014

The increasingly systematic use of the Internet in consumers decision-making processes, coupled with
the development of e-commerce, has led researchers and practitioners to examine issues concerning
service quality and satisfaction in an online context. This paper proposes a post-hoc predictive typology
of e-satisfaction based on four dimensions of e-service quality. A sample of 1144 French consumers
recruited on websites from three different industries (online travel, cultural goods (books, CDs, DVDs,
etc.), and electronics goods) was used to generate the proposed typology. Using a mixture modeling
regression, ve groups of consumers who differed in terms of e-satisfaction and e-service quality were
identied: the involved, the browsers, the fun seekers, the careful and the surfers. All the four
e-service quality dimensions investigated were found to inuence e-satisfaction. Theoretical and
managerial implications are discussed.
& 2014 Elsevier Ltd. All rights reserved.

Keywords:
e-Service quality
e-Satisfaction
e-Commerce
Predictive typology
Regression mixture model

1. Introduction
The Internet has become increasingly accessible and is now
used by a wider spectrum of consumers. However, as in traditional
settings, online consumers differ in their preferences (Burke,
1997). To address this heterogeneity, managers rely on market
segmentation. Prior research has proposed online consumer
typologies (Ganesh et al., 2010; for a review) to help managers
segment the online consumer market into homogeneous segments. These typologies can also help managers understand the
consumer market and eventually target segments with specic
offers.
Prior online consumer typologies have focused on general
segmentation bases such as web usage, Internet lifestyle and
shopping motivations (Brengman et al., 2005; Ganesh et al.,
2010; Gehrt et al., 2012; Swinyard and Smith, 2003). Although
useful for better understanding the overall online consumer
market, these segmentation bases may not be very relevant in
segmenting a retail website customer base. Website-specic
segmentation bases are more pertinent for this endeavor. For
instance, a segmentation based on specic website quality (e.g.,
website aesthetics) would be more useful for a website manager

Corresponding author. Tel.: 335 56 84 22 28.

E-mail addresses: gregory.bressolles@kedgebs.com (G. Bressolles),
francois.durrieu@kedgebs.com (F. Durrieu), sylvain.senecal@hec.ca (S. Senecal).
1
Tel: 335 56 84 22 54.
2
Tel.: 1 514 340 6980.
http://dx.doi.org/10.1016/j.jretconser.2014.07.004
0969-6989/& 2014 Elsevier Ltd. All rights reserved.

than a segmentation based on consumers general online shopping

motivations (e.g., Internet convenience).
This article is the rst, to our knowledge, to propose a post-hoc
predictive online consumer typology based on website-specic
segmentation bases. Using a regression mixture model (Wedel and
DeSarbo, 2002), e-service quality dimensions are used as a predictor of e-satisfaction. Although the relationship between
e-satisfaction and e-service characteristics has been supported in
past research (e.g., Massad et al., 2006), overall satisfaction can be
different from one user to another depending on the importance
given to the different characteristics of e-service quality. Overall
e-satisfaction is a relevant basis for segmentation because it is an
important determinant of ongoing performance (Gummerus et al.,
2004; Harris and Goode, 2004). E-service quality dimensions are
also relevant bases for segmentation because they are more
detailed and actionable for managers.

2. Literature review
2.1. Consumer typologies
Consumer classications are implicitly or explicitly used by
managers to cope with the heterogeneity of buyers (Myers and
Nicosia, 1968). By segmenting their market, managers can implement
marketing programs that are better attuned to customer needs.
Market segmentation can be dened as a state of demand
heterogeneity such that the total market demand can be disaggregated into segments with distinct demand functions (Dickson

890

G. Bressolles et al. / Journal of Retailing and Consumer Services 21 (2014) 889896

and Ginter, 1987, p. 5). Marketing segmentation research focuses

on segmentation bases and methods. A segmentation basis refers
to a set of variables or characteristics used to assign customers to
homogeneous groups (Wedel and Kamakura, 2000). Segmentation bases can be observable (e.g., store patronage) or unobservable (e.g., satisfaction) and general (e.g., demographics) or
product-specic (e.g., product preferences). Wedel and Kamakura
(2000) suggest that segmentation methods can be categorized as a
priori (type and number of segments determined in advance) or
post-hoc (type and number of segments determined on the basis of
data analyses) and as predictive (interdependence among bases)
or descriptive (no interdependence among bases).
2.2. Online consumer typologies
Initially, research focused on differences between online and
ofine consumers. These studies used segmentation bases such as
demographics, shopping motivations, innovativeness, risk aversion, and privacy concerns to illustrate differences between online
and ofine shoppers and buyers (e.g., Mathwick et al., 2001;
Sheehan, 2002). There have also been studies focusing on differences between online consumers that led to proposing online
consumer typologies (e.g., Christodoulides et al., 2013; Barnes
et al., 2007; Bhatnagar and Ghose, 2004; Ganesh et al., 2010;
Gehrt et al., 2012; Moe, 2003). For instance, Gehrt et al. (2012)
used the e-service quality dimensions identied by Wolnbarger
and Gilly (2003) to describe customer segments based on their
online shopping orientations. Most consumer online typology
research has used cluster analysis to segment the market (e.g.,
Gehrt et al., 2012; Hill et al., 2013). One noticeable exception is the
typology proposed by Bhatnagar and Ghose (2004), which used
latent class analysis. Furthermore, prior online consumer typologies have used general segmentation bases such as online shopping attitudes, motivations (Barnes et al., 2007; Ganesh et al.,
2010; Kau et al., 2003; Hill et al., 2013), online shopping patterns
(Kau et al., 2003; Rohm and Swaminathan, 2004), e-store attribute
importance or performance (Bhatnagar and Ghose, 2004; Ganesh
et al., 2010), psychographics (Barnes et al., 2007), and shopping
orientations (Gehrt et al., 2012). Hence, the online consumer
typologies proposed thus far have followed a post-hoc descriptive
approach, and most of them have used general segmentation
bases. Because general segmentation bases are independent of
products, services, or circumstances and because product-specic
segmentation bases are related to the customer and the product,
and/or the circumstances, the latter should be more useful to
website managers (e.g., website quality attributes). By using
product-specic segmentation bases such as e-service quality
dimensions (Bhatnagar and Ghose, 2004; Ganesh et al., 2010)
and e-satisfaction to segment the online market, managers should
be in a better position to target promising consumer segments for
their websites.
2.3. Regression mixture model
Prior research based on regression mixture models has created
classications that allow for heterogeneity in terms of price,
frequency of purchasing, and brand choice (Bucklin et al., 1998;
Bckenholt, 1993; Heilman et al., 2000; Wedel et al., 1993) and
dene the prole of each market segment. In a service context, this
method was used to measure traditional service quality with
SERVQUAL (Wedel and DeSarbo, 1994); results show the heterogeneity of customers service quality evaluations. Wedel and
DeSarbo (1995) explain customer satisfaction, using a mixture
regression model, with 5 determinants (expectations, performance,
disconrmation, attribution and inequity). DeSarbo et al. (2001)
explain customer value using a mixture regression model in an

ofine context. They used a mixture regression model to uncover

unobserved heterogeneity and explain customer value using service
quality dimensions. Concerning online shopping, mixture regressions models have been used to identify how consumers choose
assortments of brands (Papatla and Bhatnagar, 2002), how visitors
browse in relation to their past visits (Park and Fader, 2004),
how visitors search for information (Johnson et al., 2004). Finally,
Bhatnagar and Ghose (2004) successfully used a latent class analysis
approach to segment online consumers.
2.4. e-Service quality
Recent developments have led researchers and practitioners to
re-evaluate certain traditional concepts, such as service quality and
satisfaction, in the context of information technology. Online consumers actively participate in service delivery, supplying their own
effort and time, and consumers also contribute by assuming part of
the responsibility for service delivery. This aspect of online purchasing can affect perceptions of service quality and satisfaction
(Zeithaml and Bitner, 2000). Whereas traditional service quality
is dened as a consumer's judgment about an entity's overall
excellence or superiority (Parasuraman et al., 1988, p. 15), e-service
quality represents the extent to which a website facilitates efcient
and effective shopping, purchasing and delivery (Parasuraman
et al., 2005, p. 217).
The relevant literature on e-service quality indicates that a
website is not simply a utilitarian tool for nding information,
making choices, and ordering a product or service. Studies in this
area have progressively taken into account other dimensions
linked to the pleasure associated with the site visit and the
interaction between the user and the website.
Based on a review of the literature in marketing and information
systems (Barnes and Vidgen, 2001; Bauer et al., 2006; Bressolles,
2006; Collier and Bienstock, 2006; Cristobal et al., 2007; Loiacono
et al., 2007; Parasuraman et al., 2005; Wolnbarger and Gilly, 2003;
Yoo and Donthu, 2001), we identied ve main dimensions of
e-service quality based on a content analysis of the items and
dimensions of the main published scales. These dimensions are also
identied by Ladhari (2010) as common dimensions used in evaluating e-service quality, regardless of the service type:
(1) Quality and quantity of information: These elements are frequently cited as signicant reasons for online buying (e.g.,
Barnes and Vidgen, 2001) and have not been directly examined in evaluations of traditional service quality. The absence
of contact with the service provider during online buying
increases the need for clarity and precision of information.
This dimension measures the Internet user's perceptions of the
quality and quantity of commercial or technical information
on the services or goods provided by businesses.
(2) Ease-of-use: Online transactions can appear complex and
thereby intimidate some consumers. The ease-of-use of the
website is thus an important factor in e-service quality (Barnes
and Vidgen, 2001; Yoo and Donthu, 2001). Ease-of-use generally refers to usability in the Internet context (Swaminathan
et al., 1999). For a website, usability represents the ability of a
customer to nd information or enact a transaction with the
least amount of effort (Collier and Bienstock, 2006).
(3) Aesthetics: It refers to the richness of the representation of the
website's environment, linked to its formal characteristics
(Steuer, 1992) such as graphics, colors, images, animated
sequences, etc. These elements contribute to the look, feel,
and atmosphere of the website (Yoo and Donthu, 2001).
(4) Reliability and respect of commitments: As in the traditional
context, these are related to the ability of online sellers to keep
their promises, to fulll the terms of the exchange, and to

G. Bressolles et al. / Journal of Retailing and Consumer Services 21 (2014) 889896

deliver the product as ordered while meeting service requirements and respecting the delivery time (Cristobal et al., 2007).
(5) Security and privacy: Security refers to protecting the user from
the risk of fraud and nancial losses. Privacy refers to protection of personal details and the implicit or explicit agreement
not to sell or exchange the personal information gathered from
consumers during the service experience. This dimension is
specic and important to online commerce (Parasuraman
et al., 2005).

2.5. Customer satisfaction

Satisfying customers needs by delivering superior service quality
is considered to be as important online as it is ofine (Parasuraman
et al., 2005; Reichheld et al., 2000). E-satisfaction as a construct has
gained increasing importance in the marketing literature (e.g.,
Evanschitzky et al., 2004; Harris and Goode, 2004; Meuter et al.,
2000; Szymanski and Hise, 2000). The examination of satisfaction in
an online context follows the growing consensus that in Internet
retailing, as in traditional retailing, satisfaction is not only a critical
performance outcome but also a primary predictor of the Internet
retailer's durability and success (Gummerus et al., 2004; Harris and
Goode, 2004; Christodoulides and Michaelidou, 2011).
Satisfaction can be dened as an affective state arising from a
process of affective and cognitive evaluation of a specic transaction (Oliver, 1981). This judgment is relative because the evaluation is a comparative process between the subjective experience
and an initial reference or standard of comparison. In this paper,
e-satisfaction is dened as consumers judgment of their Internet
retail experience compared to their experience with other online
or traditional retail stores (Anderson and Srinivasan, 2003).
Studies investigating the relationship between perceived quality and satisfaction show that perceived quality has an inuence
on satisfaction (Cronin and Taylor, 1992; Iacobucci et al., 1994).
In addition, prior research show that e-service quality inuences
satisfaction (Wolnbarger and Gilly, 2003).
In sum, satisfaction, as well as e-service quality, is important
both for consumers and retailers. A better understanding of how
these variables vary across groups of homogeneous consumers is
important both for theory and practice.

3. Method
3.1. Sample and procedure
Three product categories represented by three French major
retail websites were selected for this research: online travel,
cultural goods (books, CDs, DVDs, etc.), and electronic goods.
These online sectors were selected because of their importance
in terms of purchase frequency and sales volume. An online
questionnaire was made available to the buyers and the visitors
of the three websites between September and November 2012.
Visitors were recruited while visiting the website (a button
inviting them to participate in our research was placed on the
home page of each website), and buyers were contacted by email
after completing their purchase on one of the website. The
respondents have to answer all the questions in order to validate
their participation. The questionnaire was completed by 1144
respondents, of whom 58% were buyers, 53% were men, and 66%
were under 45 years of age (see Appendix A for the sample
details). We were not able to calculate the response rate for the
visitors (because each visitor of the different websites could be a
potential respondent), but for the buyers, the response rate varied
from 8% to 10% depending on the website. The demographic

891

prole of respondents was similar to the French web users

population and represented the customer basis of the visitors/
buyers of the three selected websites. With the exception of the
level of education and occupation, there were no signicant
differences in demographic characteristics among respondents
across the three websites.
3.2. Measures
To assess the above mentioned e-service quality dimensions,
the NetQual measurement scale (Bressolles, 2006) was used. This
scale was developed in French and it is suggested to have better
psychometric characteristics and predictive power (Bressolles and
Nantel, 2008) than other e-service quality scales such as EtailQ
(Wolnbarger and Gilly, 2003), WebQual (Barnes and Vidgen,
2001) and Sitequal (Yoo and Donthu, 2001). It includes 18 items
distributed across 5 dimensions: Information, Ease-of-use, aesthetics, reliability, and security/privacy (Appendix B). Because the
current research recruited both consumers who did and did not
complete a transaction during their website visit (buyers and
visitors), the Reliability dimension was not included in the online
questionnaire. E-satisfaction was assessed with a measurement
scale adapted from Olivers (1997) scale. This scale measures the
cognitive, affective, and conative dimensions of satisfaction on a
commercial website (Appendix B). All items were assessed with
5-point Likert-type scales.
3.3. Mixture regression model
To better understand consumer homogeneity relative to
e-satisfaction and e-service quality, a post-hoc predictive typology
of satisfaction was proposed. Post-hoc predictive segmentation
methods identify consumer segments on the basis of the
estimated relationship between a dependent variable and a set
of predictors (Wedel and Kamakura, 2000, p. 23). Following
Wolnbarger and Gilly (2003), in the proposed typology, satisfaction was used as the dependent variable, and e-service quality
dimensions were used as the predictors. To our knowledge, this
study is the rst to propose a post-hoc predictive online consumer
typology.
The proposed typology was generated using a regression
mixture model approach (Wedel and DeSarbo, 1995). The mixture
methodology, a type of latent class analysis, is based on the
assumption that the data arise from a mixture of distributions
(e.g., normal or multinomial) (Wedel and Kamakura, 2000). In
mixture modeling, the density function (e.g., Binomial, Gamma) is
used to describe the probabilities of occurrence of the observed
value of the variable in question conditional on knowing the group
from which those values were drawn (Wedel and Kamakura,
2000). For these methods, it is not known how the observations
are mixed to form a group because the individuals could be
classied into two or more groups without knowing a priori the
probability of belonging to each group. Mixture modeling estimates the probability that objects (in our case, consumers) belong
to a group, and it differs from cluster analysis, in which objects
belong to one (non-overlapping) or multiple groups (overlapping).
Instead of estimating a regression equation for all observations
taken together, the mixture model estimates a regression for each
segment.
Fit indices such as the Bayesian Information Criterion (BIC) and
the Consistent Akaike Information Criterion (CAIC; the lowest
score being the best solution of the appropriate number of
segments) are used to determine the number of groups in the
typology. To identify a solution to the allocation problem, Entropy
(Es) is used to measure the quality of group separation and
therefore the allocation of individuals to the groups. If the result

892

G. Bressolles et al. / Journal of Retailing and Consumer Services 21 (2014) 889896

is close to 1, the groups are well separated. To trace problems of

validation of the typology, it is suggested that standard errors and
the covariance matrix be veried for each segment. Very large
standard errors and residuals, variances or covariances of the
estimates (multicollinearity) and inuential points (effect of
extremes) lead to unidentied and unvalidated models (Cooil
et al., 2007). Another way to validate the model is to analyze the
percentage of variance explained by the solution (R) and to
measure the goodness-of-t of the model with CAIC and BIC
indices, dened as a measure of the distance between the tted
model and the true distribution (Andrews and Currim, 2003).

4. Results
4.1. Measurement scales
First, a conrmatory factor analysis was performed to verify the
structure of the NetQual scale. In order to avoid problems associated
with multivariate non-normality, we applied the Robust method
(Bentler and Wu, 2002) which corrects t indices and coefcients of
the model (parameters and standard errors). We used a Jackknife
approach (Fenwick, 1979) to avoid the problem of the sample size and
the complexity of the model. This approach is a method used for
determining the sampling distribution of any parameter for the model
or any statistic as t indices. Thus, we attained more stable and more
consistent coefcients and t indices for the model. The model's
goodness of t was satisfactory with all indicators meeting generally
accepted criteria ( 120.08, p 0.000, GFI0.98 AGFI 0.97,
NFI0.97, NNFI0.98, CFI 0.99, RMSEA0.03, SRMR0.03). Thus,
the four dimensions of NetQual scale were conrmed.
Next, we examined the convergent and the discriminant
validity of the NetQual scale. The convergent validity is conrmed
if each item shares more variance with its construct than with the
error (Fornell and Larcker, 1981). Thus, convergent validity of each
dimension is veried if the average of all the (vc) are more than
0.5 and all the are more than 0.5. The discriminant validity of the
e-service quality construct has also been examined. Discriminant
validity is established if the shared variance between the dimensions is less than the shared variance between dimensions and
their items (Fornell and Larcker, 1981); which is the case for all
dimensions (Table 1). Table 1 shows the convergent and discriminant validity and the reliability () of the dimensions of the
NetQual scale (Fornell and Larcker, 1981). Moreover, the R mean
was 0.63, indicating an acceptable predictive validity of e-service
quality.
All structural, measurement, and standard error coefcients of
the model obtained by the jackknife procedure are shown in Fig. 1.
Finally, the reliability of the e-satisfaction scale was also veried
and was found to be satisfactory ( of Joreskog .84 and vc .57).
4.2. Post-hoc predictive typology
Table 2 shows the indices of the different typologies tested and
the selection of the tted solution. We determined the
Table 1
Convergent and discriminant validity of NetQual's scale.

Information
Ease of use
Aesthetics
Security/privacy
nn
n

vc**

Information

Ease-of-use

Aesthetics

0.58
0.63
0.65
0.67

0.81
0.87
0.85
0.87

1
0.50*
0.27*
0.25*

1
0.42*
0.31*

1
0.25*

vc (Must be Z 0.5).
Square correlation (must be less than each of the two vc).

corresponding optimal number of latent classes by testing solutions with 4 to 8 classes. The CAIC and BIC values were the lowest
for the 5-class solution, so this solution was retained. An Entropy
(Es) of 0.71 gives an acceptable classication of the observations
based on their probabilities of belonging to a group (DeSarbo et al.,
2001). The overall R was 0.92, which indicates that for the ve
groups, the overall explained variance was satisfactory. Thus, the
selected solution indicated the optimal typology of Internet users
in regard to their satisfaction across the e-service quality
dimensions.
All coefcients were signicant, at a probability level of 0.05
(Table 3). The standard errors of each parameter in each group
were low (standard error less than 0.03; Table 3). The error
segment variance of each group was also low (error segment
variance less than 0.02; Table 3). The independence of e-service
quality dimensions was veried for each group (variances or
covariances of the estimates were near zero). In conclusion, the
model was validated, and there were no identication problems.
The prole of each group was determined by e-service quality
dimensions and the e-satisfaction level. The ve groups were of
different sizes, ranging from 37% of the sample to 5% of the
sample. The ve groups were differentiated by their level of
satisfaction, ranging from 3.97 to 4.70 on a scale of 1 to 5. The
ve groups are described in details below.
The second column of Table 3 presents the rst group: the
involved consumers. This group was the largest group and
represented 37% of the sample. The satisfaction level of 4.29
placed these consumers in the middle, with two groups with
higher satisfaction and two groups with lower satisfaction.
Although all e-service quality dimensions had a signicant inuence on satisfaction, satisfaction was mostly inuenced by ease-ofuse (0.33) and security/privacy (0.22). This group was mostly
composed of male (57.5%; Table 4). The buyers were underrepresented compared to the percentage of buyers in the sample (58.9%
vs. 62.3%). These involved consumers were mostly coming from
the cultural goods website (53.4%; Table 4).
The most satised group was the second group, the browsers,
with a satisfaction score of 4.70. This group represented 10% of the
sample. For consumers in this group, the most important driver of
satisfaction was the website's ease-of-use. In terms of e-service
quality dimensions, this group was more similar to Groups 1 and
3 and most dissimilar to Group 5. This group was mostly
composed of males (60.2%), buyers (68.3%) and originating from
the online travel website (52%; Table 4).
The third group, the fun seekers, had the lowest satisfaction
level (3.97) and was the second largest group, with 33% of the
sample size. The most important drivers of this group's satisfaction
were ease-of-use (0.33), security/privacy (0.22), and the aesthetics
of the website (0.21). Men were underrepresented in this group
compared to the percentage of men in the sample (52.3% vs.
55.9%). This group was mostly composed of buyers (64.6%) and
from the online travel website (47.7%; Table 4). Although their
satisfaction was inuenced by various e-service quality dimensions, their lower overall satisfaction suggested that other important antecedents, apart from e-service quality dimensions, also
had an inuence on their satisfaction.
The fourth group, the careful consumers, represented only 5% of
the sample. Consumers from this group reported an average satisfaction level of 4.17. The most important drivers of their satisfaction were
security/privacy (0.25), ease-of-use (0.20), and the aesthetics of the
website (0.20). In terms of e-service quality dimensions, this group
was most similar to Group 2 and most dissimilar to Group 5. It was
mostly composed of females (56.5%). The buyers were underrepresented compared to the percentage of buyers in the whole sample
(52.2% vs. 62.3%).The consumers coming from the electronic goods
website were overrepresented in this group (26.1% vs. 10.3; Table 4).

G. Bressolles et al. / Journal of Retailing and Consumer Services 21 (2014) 889896

893

Fig. 1. CFA on the NetQual scale.

Table 2
Typology comparisons.

CAIC
BIC
Es
R-square

Class 4

Class 5

Class 6

Class 7

Class 8

208.09
181.09
0.71
0.88

98.37
64.37
0.71
0.92

109.31
68.31
0.70
0.93

157.43
109.43
0.71
0.94

182.78
127.78
0.69
0.95

The fth group, the surfers represented 16% of the sample. On

average, consumers from this group had a fairly high satisfaction
level (4.49). By far, the most important driver of satisfaction for
this group was ease-of-use (0.40). In terms of e-service quality
dimensions, this group was unique because it was the only one
focusing mainly on ease-of-use. This group was mostly composed
of male consumers (58%), buyers (61.6%), and fairly similar to the
sample average on other characteristics.
The ve groups differed in their perceptions of how the four
dimensions of e-service quality act to create e-satisfaction. Notably, the two groups with the highest levels of satisfaction (browsers and surfers) were the ones with a relatively large coefcient
(i.e., Z0.20) on only one dimension of e-service quality (i.e., easeof-use). The next group, in terms of satisfaction (involved), had
two dimensions with large coefcients (i.e., ease-of-use and
security/privacy) and it was the only group than used information
to enhance satisfaction. Finally, the two groups with the lowest

satisfaction levels (careful and fun seekers) were the ones with
large coefcients on three dimensions (ease-of-use, aesthetics, and
security/privacy). The careful used more security/privacy as a
satisfaction driver and the fun seekers use mostly aesthetics
compared to the other groups. Also, the fun seekers were not
satised without the e-service quality dimensions retained (intercept0.03 in Table 3). Ease-of-use was an important driver of
satisfaction across all groups. Additionally, the overall satisfaction
levels of consumers who had multiple important determinants of
satisfaction (e.g., ease-of-use, security/privacy, and aesthetics)
tended to be lower.

5. Discussion
This article presented ve heterogeneous groups of Internet
users who had different levels of satisfaction that were explained
by e-service quality dimensions. However, these dimensions were
not equally important across all groups.
The ease-of-use dimension was consistently important in predicting the satisfaction of the ve groups (coefcients ranging from
0.20 to 0.40). Like Bhatnagar and Ghose (2004) and Ganesh et al.
(2010), we found that ease-of-use is an important service quality
dimension in segmenting consumers. The security/privacy dimension was important to three groups out of ve (from 0.22 to 0.25).
This result is also consistent with Ganesh et al. (2010) and Bhatnagar
and Ghose (2004). For instance, Bhatnagar and Ghose (2004)

894

G. Bressolles et al. / Journal of Retailing and Consumer Services 21 (2014) 889896

Table 3
5 Group typology.
1 Involved
Coeff
Ease-of-use
Aesthetics
Security/privacy
Information
Intercept
Size (%)
Level of e-satisfaction
Variance error

0.30
0.19
0.22
0.16
0.42
37
4.29
0.02

2 Browsers
Std err.
0.01
0.01
0.01
0.01
0.02

Coeff

Std err.

0.29
0.19
0.19
0.12
0.72
10
4.70
0.00

0.01
0.01
0.01
0.01
0.01

3 Fun seekers

4 Careful

Coeff

Std err.

Coeff

Std err.

0.01
0.01
0.01
0.01
0.01

0.20
0.20
0.25
0.06
0.36
5
4.17
0.00

0.01
0.01
0.01
0.01
0.01

0.33
0.21
0.22
0.12
0.03
33
3.97
0.02

5 Surfers
Coeff
0.40
0.11
0.07
0.09
0.46
16
4.49
0.00

Std err.
0.01
0.01
0.01
0.01
0.01

Table 4
Groups characteristics.
Group

Purchase (%)

Male (%)

Online travel website (%)

Electronic goods website (%)

Cultural goods website (%)

1 Involved
2 Browsers
3 Fun seekers
4 Careful
5 Surfers
Average

58.9
68.3
64.6
52.2
61.6
62.3

57.5
60.2
52.3
43.5
58.0
55.9

39.2
52.0
47.7
34.8
39.5
43.3

7.4
11.4
10.0
26.1
12.7
10.3

53.4
36.6
42.3
39.1
47.8
46.4

demonstrated that security of sensitive information was the most

important attribute in all three of their segments. However, in our
typology, we found that security/privacy was the second most
important element, after ease-of-use, to predict satisfaction. This
may be due to an evolution in the importance of the dimensions of
e-service quality for the consumer. In 2004, when Bhatnagar and
Ghoses (2004) typology was published, security aspects were a
major concern for most buyers. Based on our ndings, it seems that
currently ease-of-use is the main concern. The aesthetics dimension was a predictor of satisfaction for only one group. Although the
information quality dimension was a signicant predictor of
satisfaction for all groups in our typology, it was the weakest
predictor for all groups. Similarly, Bhatnagar and Ghose (2004) found
that quality of information about purchase choices was not one of
the most important attributes and that websites did not perform well
on this dimension. Overall, our ndings suggest that hedonic aspects
of a website seem to be less important in predicting satisfaction than
utilitarian dimensions such as ease-of-use and security/privacy.
Interestingly, Martnez-Lpez et al. (2014) suggest that it is important
for managers of commercial websites to promote the utilitarian value
of their websites in order to achieve greater levels of satisfaction
among their users and greater volume of sales.

6. Conclusion
To our knowledge, this study is the rst to propose a post-hoc
predictive online consumer typology by using e-service quality
dimensions as predictors of e-satisfaction. Prior research on online
typologies only used descriptive typologies, not predictive typologies. Hence, this work extends the prior work of Ganesh et al.
(2010) and Bhatnagar and Ghose (2004) on online consumer
typologies. The proposed typology suggests that not all consumers
derive the same satisfaction from their interaction with a specic
website; thus, it is important to take consumer heterogeneity into
account.
Thus, our theoretical contribution is twofold. First, our work
extends prior work on online consumer typologies (Ganesh et al.,

2010; Bhatnagar and Ghose, 2004) by simultaneously using

a predictive approach to online segmentation and by using a
product-specic segmentation basis (e-service quality dimensions).
Thus, our ndings contribute to better understand how online
different segments of consumers perceive e-service quality and
how it impacts their satisfaction. In addition, we also replicate and
extend prior ndings on e-service quality in a more ecologically
valid context (i.e., intercept of actual visitors and buyers on each
website used in our study) than previous research. For instance, in
Bhatnagar and Ghoses (2004) study, participants were asked to
evaluate websites in specic product categories without ensuring
that they had actually visited websites in these categories.
Based on a large sample of web users on three different major
French websites, our ndings suggest important managerial
implications. First, our results suggest that although satisfaction
is inuenced by website attributes, different consumers will be
affected differently. We found ve distinct segments of consumers.
For four segments, ease-of-use was the most important predictor
of e-satisfaction. This result is consistent with Demangeot and
Broderick (2010) who provide corroborating evidence that website
design, in the form of usability and information availability, is a
key factor for online retailers. In addition security/privacy was an
important predictor of satisfaction for two segments. Finally, for
two other segments, in addition to ease-of-use and security/
privacy, aesthetics were a major predictor of satisfaction. We
observed also that the stronger the segments predictors were
(Z0.20), the less satised consumers were with the website. In
addition, our results suggest that not all segments are present in
the same proportions on the investigated websites (X 25.312
p .000). Thus, in order to improve satisfaction, managers need to
know which segments are present in their market in order allocate
resources efciently to focus on service quality dimensions most
important for their segments.
For instance, for a specic website, if the majority of the
consumers are from the surfers or the browsers segments,
managers need to mainly focus on how to improve the ease of use
of the website in order to increase satisfaction. They should pay
particular attention to usability aspects of their navigation

G. Bressolles et al. / Journal of Retailing and Consumer Services 21 (2014) 889896

interface by following usability guidelines such as the ones

proposed by Bastien and Scapin (1993) and industry best practices
(www.nngroup.com/topic/web-usability). For another website, if
the majority of the consumers are from the involved segment,
managers need to focus on security/privacy aspects, in addition to
their website usability, to satisfy consumers. Third party organizations such as BBBOnline and TRUSTe provide best practices to
follow to address security and privacy issues (www.truste.com/
resources/privacy-best-practices). For instance, TRUSTe suggests
that the website privacy statement should be easy to read and
understandable and that it should be aligned with the rm's
terms-of-service. For another website, if the majority of the
consumers are from the careful segment, in addition to ease of
use and security privacy aspects, managers also need to focus on
aesthetics of the website in order to improve satisfaction. With the
help of website designers, managers can benet from a welldesigned and attractive website (Robins and Holmes, 2008).
This research presents some limitations that could offer interesting avenues for future research. It suggests that the customer esatisfaction depends on a positive evaluation of the principal
dimensions of website quality and that there is heterogeneity
among consumers regarding their perception of e-service quality
and e-satisfaction. The differences among the ve identied
groups were explained to a great extent by four service quality
dimensions: ease-of-use, information, aesthetics and security/privacy. However, additional service quality determinants
could be investigated. For instance, Ganesh et al. (2010) mentioned
that interactivity and ability to offer personalized services is
another important aspect of a website that could contribute to
explaining e-satisfaction. As suggested by Moe (2003) and empirically tested by Hauser et al. (2009), website personalization in
real-time also has a positive effect on consumer responses;
therefore, personalization could play a role in determining
e-satisfaction.
The typology was developed based on the evaluation of
websites in three different industries (travel, electronics, and
cultural goods) by French consumers. Although the sample was
similar to the French Internet users population and that the three
websites used in this study are major players in France, it remains
to be seen whether the proposed typology would work in other
populations. Thus, replicating this research in different online
sectors and/or in different countries would test the validity of
the proposed typology. Moreover, based on Scarpi et al. (2014)
ndings about the relationship between shopping motivations and
shopping channels (online or ofine), future work could include
the inuence of shopping motivations and single channel vs.
multi-channel usage in future online consumer typologies. Finally,
it would be useful to test the moderating effects of personal
characteristics (e.g., culture, online expertise) and situational
variables (e.g., perceived risk, time pressure) on the relationship
between e-service quality and e-satisfaction in order have a
better understanding of consumers online shopping experience
(Michaud-Trevinal and Stenger, 2014).
Appendix A. Sample characteristics

%
Sex
What is your age?

Male
Female
Under 24
25 to 34
35 to 44
45 and over

52.80
47.20
13.59
38.91
22.68
23.79

Your profession?

When did you nish your

education?

Marital status

Annual revenue of your

household?

Buyers/Visitors

895

Non-working (students
)
Middle class
Upper class
18 years of age

39.54
26.44
34.02
26.36

2 years of further study

3 or 4 years of further
study
5 or more years of further
study
Single
Couple
Couple with children
Less than 9,000 euros

23.03
22.45

28.62
32.23
34.54
9.68

9,000 to 15,000 euros

15,000 to 30,000 euros
30,000 to 45,000 euros
45,000 euros or more
Buyers
Visitors

14.36
37.11
22.63
16.21
58.57
41.43

28.16

Appendix B. Measurement scales

Dimension
Information Info1

Item

The site gives detailed information

on the product(s) or service
(s) supplied
Info2
Information on the site is pertinent
Info3
Information on the site is accurate
Ease of use Eofu1
The site is easy to use
Eofu2
It is easy to look for information on
the site
Eofu3
It is easy to move around and nd
what I am looking for on the site
Eofu4
The organization and the layout of
the site make information research
easy
Eofu5
The site is laid out in a clear and
simple way
Aesthetics
Aesthetic1 The site looks good
Aesthetic2 The site is creative
Aesthetic3 The site is visually attractive
Security/
Secu1
Generally, I have condence in the
privacy
site's security
Secu2
The site guarantees security in
navigation
Secu3
I think my personal details are safe
on this site
Secu4
I have condence that the site will
not use my personal information for
undesirable purposes
Satisfaction Sat1
I am satised with my decision to
navigate on this Website.
Sat2
My choice to navigate on this
website was a wise one.
Sat3
I am sure it was the right thing to do
to navigate on this website.
Sat4
Would you buy on this website again
in the future?

896

G. Bressolles et al. / Journal of Retailing and Consumer Services 21 (2014) 889896

References
Anderson, R.E., Srinivasan, S.S., 2003. E-satisfaction and e-loyalty: a contingency
framework. Psychol. Marketing 20, 123138.
Andrews, R.L., Currim, I.S., 2003. A comparison of segment retention criteria for
nite mixture logit models. J. Marketing Res. 40 (2), 235243.
Barnes, S.J., Bauer, H.H., Neumann, M.M., Huber, F., 2007. Segmenting cyberspace: a
customer typology for the internet. Eur. J. Marketing 41 (1-2), 7193.
Barnes, S.J., Vidgen, R., 2001. An evaluation of cyber-bookshops: the WebQual
method. Int. J. Electron. Com. 6 (1), 1130.
Bastien, J.M.C., Scapin, D., 1993. Ergonomic criteria for the evaluation of human
computer interfaces. Institut National de recherche en informatique et en
automatique (France).
Bauer, H.H., Falk, T., Hammerschmidt, M., 2006. eTransqual: a transaction processbased approach for capturing service quality in online shopping. J. Bus. Res. 59
(7), 866875.
Bentler, P.M., Wu, E.J.C., 2002. EQS 6 for Windows: User's Guide, second ed.
Multivariate Software, Encino (CA).
Bhatnagar, A., Ghose, S., 2004. A latent class segmentation analysis of e-shoppers.
J. Bus. Res. 57 (7), 758767.
Bckenholt, U., 1993. Estimating latent distributions in recurrent choice data.
Psychometrika 58 (3), 489509.
Brengman, M., Geuens, M., Weijters, B., Smith, S.M., Swinyard, W.R., 2005.
Segmenting Internet shoppers based on their web-usage-related lifestyle: a
cross-cultural validation. J. Bus. Res. 58 (1), 7988.
Bressolles, G., 2006. Electronic service quality: NetQu@l proposition of a measurement scale to commercial websites and moderating effects. Recherche et
Applications en Marketing 21 (3), 1945.
Bressolles, G., Nantel, J., 2008. The measurement of electronic service quality:
improvements and application. Int. J. e-Bus. Res. 4 (3), 119.
Bucklin, R.E., Gupta, S., Siddarth, S., 1998. Determining segmentation in sales
response across consumer purchase behaviors. J. Marketing Res. 35 (2),
189197.
Burke, R.R., 1997. Do you see what I see? The future of virtual shopping. J. Acad.
Marketing Sci. 25 (4), 352360.
Christodoulides, G., Michaelidou, N., Theofania Siamagka, N., 2013. A typology of
Internet users based on comparative affective states: evidence from eight
countries. Eur. J. Marketing 47 (1/2), 153173.
Christodoulides, G., Michaelidou, N., 2011. Shopping motives as antecedents of
e-satisfaction and e-loyalty. J. Marketing Manage. 27 (1), 181197.
Collier, J.E., Bienstock, C.C., 2006. Measuring service quality in e-retailing. J. Serv.
Res. 8 (3), 260275.
Cooil, B., Keiningham, T.L., Aksoy, L., Hsu, M., 2007. A longitudinal analysis of
customer satisfaction and share of wallet: investigating the moderating effect
of customer characteristics. J. Marketing 71 (1), 6783.
Cristobal, E., Flavin, C., Guinalu, M., 2007. Perceived e-service quality (PeSQ):
measurement validation and effects on consumer satisfaction and web site
loyalty. Manag. Serv. Qual. 17 (3), 317340.
Cronin Jr., J.J., Taylor, S.A., 1992. Measuring service quality: a reexamination and
extension. J. Marketing 56 (3), 5568.
Demangeot, C., Broderick, A.J., 2010. Consumer perception of online shopping
environments: a Gestalt approach. Psychol. Marketing 27 (2), 117140.
DeSarbo, W.S., Jedidi, K., Sinha, I., 2001. Customer value analysis in a heterogeneous
market. Strategic Manage. J. 22 (9), 845857.
Dickson, P.R., Ginter, J.L., 1987. Market segmentation, product differentiation, and
marketing strategy. J. Marketing 51 (2), 110.
Evanschitzky, H., Gopalkrishnan, R.I., Hesse, J., Ahlert, D., 2004. e-Satisfaction: a reexamination. J. Retailing 80 (3), 239247.
Fenwick, I., 1979. Techniques in market measurement: the jackknife. J. Marketing
Res. 16, 410441.
Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with unobservable variables and measurement error. J. Marketing Res. 18 (1), 3950.
Ganesh, J., Reynolds, K.E., Luckett, M.g., Pomirleanu, N., 2010. Online shopper
motivations, and e-store attributes: an examination of online patronage
behavior and shopper typologies. J. Retailing 86 (1), 106115.
Gehrt, K.C., Rajan, M.N., Shainesh, G., Czerwinski, D., OBrien, M., 2012. Emergence
of online shopping in India: shopping orientation segments. Int. J. Retail Distrib.
Manage. 40 (10), 742758.
Gummerus, J., Liljander, V., Pura, M., van Riel, A., 2004. Customer loyalty to contentbased web sites: the case of an online health-care service. J. Serv. Marketing 18
(3), 175186.
Harris, L.C., Goode, M.M.H., 2004. The four levels of loyalty and the pivotal role of
trust: a study of online service dynamics. J. Retailing 80 (2), 139158.
Hauser, J.R., Urban, G.L., Liberali, G., Braun, M., 2009. Website morphing. Marketing
Sci. 28 (2), 202223.
Heilman, C.M., Bowman, D., Wright, G.P., 2000. The evolution of brand preferences
and choice behaviors of consumers new to a market. J. Marketing Res. 37 (2),
139155.
Hill, W.W., Beatty, S.E., Walsh, G., 2013. A segmentation of adolescent online users
and shoppers. J. Serv. Marketing 27 (5), 347360.

Iacobucci, D., Grayson, K.A., Ostrom, A.L., 1994. The calculus of service quality and
customer satisfaction: theoretical and empirical differentiation and integration.
In: Swartz, T.A., Bowen, D.E., Brown, S.W. (Eds.), Advances in Services Marketing
and Management: Research and Practice. JAI Press, Greenwich (CT, pp. 167.
Johnson, E.J., Moe, W.W., Fader, P.S., Bellman, S., Lohse, G.L., 2004. On the depth and
dynamics of online search behavior. Manage. Sci. 50 (3), 299308.
Kau, A.K., Tang, Y.E., Ghose, S., 2003. Typology of online shoppers. J. Consum.
Marketing 20 (2), 139156.
Ladhari, R., 2010. Developing e-service quality scales: a literature review. J. Retailing
Consum. Serv. 17, 464477.
Loiacono, E.T., Watson, R.T., Goodhue, D.L., 2007. WebQual: an instrument for
consumer evaluation of web sites. Int. J. Electron. Com. 11 (3), 5187.
Martnez-Lpez, F., Pla-Garca, C., Gzquez-Abad, J.C., Rodrguez-Ardura, I., 2014.
Utilitarian motivations in online consumption: dimensional structure and
scales. Electron. Commer. Res. Appl. 13 (3), 1882044.
Massad, N., Heckman, R., Crowston, K., 2006. Customer satisfaction with electronic
service encounters. Int. J. Electron. Com. 10 (4), 73104.
Mathwick, C., Malhotra, N., Rigdon, E., 2001. Experiential value: conceptualization,
measurement and application in the catalog and internet shopping environment. J. Retailing 77 (1), 3956.
Meuter, M.L., Ostrom, A.L., Roundtree, R.I., Bitner, M.J., 2000. Self-service technologies: understanding customer satisfaction with technology-based service
encounters. J. Marketing 64 (3), 5064.
Michaud-Trevinal, A., Stenger, T., 2014. Toward a conceptualization of the online
shopping experience. J. Retailing Consum. Serv. 21 (3), 314326.
Moe, W.W., 2003. Buying, searching, or browsing: differentiating between online
shoppers using in-store navigational clickstream. J. Consum. Psychol. 13 (1-2),
2939.
Myers, J.G., Nicosia, F.M., 1968. On the study of consumer typologies. J. Marketing
Res. 5 (2), 182193.
Oliver, R.L., 1981. Measurement and evaluation of satisfaction process in retail
setting. J. Retailing 57 (3), 2548.
Oliver, R.L., 1997. Satisfaction: A Behavioral Perspective on the Consumer. McGrawHill, New York (NY).
Papatla, P., Bhatnagar, A., 2002. Shopping style segmentation of consumers.
Marketing Lett. 13 (2), 91106.
Parasuraman, A., Zeithaml, V.A., Berry, L.L., 1988. SERVQUAL: a multiple-item scale
for measuring consumer perceptions of service quality. J. Retailing 64 (1),
1240.
Parasuraman, A., Zeithaml, V.A., Malhotra, A., 2005. E-S-QUAL: a multiple-item
scale for assessing electronic service quality. J. Serv. Res. 7 (3), 213233.
Park, Y.H., Fader, P.S., 2004. Modeling browsing behavior at multiple websites.
Marketing Sci. 23 (3), 280303.
Reichheld, F.F., Markey Jr., R.G., Hopton, C., 2000. e-Customer loyalty: applying the
traditional rules of business for online success. Eur. Bus. J. 12 (4), 173179.
Robins, D., Holmes, J., 2008. Aesthetics and credibility in web site design. Inf.
Process. Manage. 44 (1), 386399.
Rohm, A.J., Swaminathan, V., 2004. A typology of online shoppers based on
shopping motivations. J. Bus. Res. 57 (7), 748757.
Scarpi, D., Pizzi, G., Visentin, M., 2014. Shopping for fun or shopping to buy: is it
different online and ofine? J. Retailing Consum. Serv. 21 (3), 258267.
Sheehan, K.B., 2002. Toward a typology of internet users and online privacy
concerns. Inf. Soc. 18 (1), 2132.
Steuer, J., 1992. Dening virtual reality: dimensions determining telepresence.
J. Commun. 42 (4), 7393.
Swaminathan, V., Lepkowska-White, E., Rao, B.P., 1999. Browsers or buyers in
cyberspace? An investigation of factors inuencing electronic exchange.
J. Comput.-Med. Commun. 5 (2) (unpaged).
Swinyard, W.R., Smith, S.M., 2003. Why people (don't) shop online: a lifestyle study
of the internet consumer. Psychol. Marketing 20 (7), 567597.
Szymanski, D.M., Hise, R.T., 2000. e-Satisfaction: an initial investigation. J. Retailing
76 (3), 209222.
Wedel, M., DeSarbo, W.S., 1994. A review of class regression models and their
applications. In: Baggozzi, R.P. (Ed.), Advanced Methods for Marketing Research.
Balckwell, Cambridge, pp. 353388.
Wedel, M., DeSarbo, W.S., 1995. A mixture likelihood approach for generalized
linear models. J. Classif. 12 (1), 2155.
Wedel, M., DeSarbo, W.S., Bult, J.R., Ramaswamy, V., 1993. A latent class Poisson
regression model for heterogeneous count data. J. Appl. Econ. 8 (4), 397411.
Wedel, M., DeSarbo, W.S., 2002. Market segment derivation and proling via a
nite mixture model framework. Marketing Letters 13 (1), 1725.
Wedel, M., Kamakura, W.A.,, 2000. Market Segmentation: Conceptual and Methodological Foundations. Kluwer Academic, Boston (MA).
Wolnbarger, M., Gilly, M.C., 2003. eTailQ: dimensionalizing, measuring and
predicting etail quality. J. Retailing 79 (3), 183198.
Yoo, B., Donthu, N., 2001. Developing a scale to measure the perceived quality of
internet shopping sites (SITEQUAL). Q. J. Electron. Com. 2 (1), 3147.
Zeithaml, V.A., Bitner, M.J., 2000. Services Marketing: Integrating Customer Focus
Across the Firm, second ed. Irwin/McGraw-Hill, Boston (MA).