The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0959-0552.htm

IJRDM
38,7 Effects of online store attributes
on customer satisfaction
and repurchase intentions
482
Ruby Roy Dholakia
College of Business Administration, The University of Rhode Island,
Received May 2009
Revised February 2010 Kingston, Rhode Island, USA, and
Accepted March 2010 Miao Zhao
Gabelli School of Business, Roger Williams University, Bristol,
Rhode Island, USA

Abstract
Purpose – The purpose of this paper is to identify website characteristics that affect customer
evaluations and satisfaction with online stores at two interaction points – when the order is placed and
after the order has been fulfilled.
Design/methodology/approach – Using data collected by bizrate.com, data collected from
customers of thousands of online stores, the analysis focuses on the changes in the relationships
between website characteristics and customer ratings. Data for two-time periods, 15 months apart, are
used to determine the stability in the observed relationships.
Findings – Order fulfillment variables, particularly on-time delivery, dominate the effects on overall
customer evaluations and satisfaction. The statistical significance of other online store attributes,
however, changes as differences are observed between 2003 and 2004.
Research limitations/implications – The online environment is dynamic and the paper captures
some of the changes in the relationships between website attributes and customer satisfaction. This
requires continuous monitoring of the online environment. Since the paper relies on secondary data
collected by bizrate.com, the research is limited by specific website attributes and measures of
customer satisfaction adopted by a commercial enterprise.
Practical implications – Online retailers must be strategic about fulfillment variables. When online
stores compete with each other, it is easier to copy certain attributes like “shipping options” than other
attributes such as “on-time delivery.” This suggests that the most creative, interactive, and vivid
online site will not compensate for weak fulfillment and customer support capabilities.
Originality/value – No other paper has looked at these data, collected from real customers making
purchases at actual merchant sites, over two time periods. The results provide insights regarding
stability of findings.
Keywords Internet shopping, Customer satisfaction, Consumer behaviour, Repeat buying
Paper type Research paper

Both authors contributed equally to the paper.

International Journal of Retail & The authors would like to acknowledge the helpful comments of two International Journal of
Distribution Management
Vol. 38 No. 7, 2010 Retail & Distribution Management reviewers, and the editorial assistance received from
pp. 482-496 Kathleen Micken that helped improve the paper. The authors are indebted to Research Institute
q Emerald Group Publishing Limited
0959-0552
for Telecommunications and Information Marketing at College of Business Administration,
DOI 10.1108/09590551011052098 University of Rhode Island for support.
Introduction Effects of online
With the steady growth of online retailing, e-merchants are experimenting with store attributes
attributes that are unique to the new media. There are many attributes to choose from
(e.g. search engine, ordering system, personalization, virtual reality display, etc.), each
performing a specific function and distinct from other attributes within the website.
In addition, online store managers are also concerned about the impact of an attribute on
customer satisfaction and loyalty. While there is an established body of literature and 483
decades of experience regarding the design of physical stores, the new world of online
stores and website attributes are now beginning to receive attention (Burke, 2002;
Eroglu et al., 2003; Jin and Park, 2005; Kim and Kim, 2006; Park and Kim, 2003).
Researchers have attempted to identify website characteristics that affect customer
evaluations and satisfaction (Zeithaml et al., 2002; Szymanski and Hise, 2000) as well as
those that affect marketing performance and e-tail success (Lii et al., 2004; Weathers and
Makienko, 2006).
In this paper, we examine effects of online retail store attributes on customer
satisfaction and repurchase intentions. We analyze customer ratings of real-world online
stores from data collected by bizrate.com. This differs from other studies that used
fictitious stores rated by student respondents (Dholakia and Zhao, 2009; Eroglu et al.,
2003; Kim and Kim, 2006). In addition, the secondary data allow us to separate the
contribution of website attributes at the time of placing an order as well as after receiving
delivery of the ordered merchandise. This is particularly relevant for e-tailing where
there is a temporal separation of the service encounters and this separation is likely to
affect consumer judgments (Posselt and Gerstner, 2005). Furthermore, because of the
dynamic nature of attribute-satisfaction relationship, we examine data for two years to
identify changes and stability in the observed relationships.
The paper is organized as follows. First, the paper briefly reviews the literature on
website attributes and their relationship to customer satisfaction and loyalty. Next, the
design of the empirical study, specifically the secondary data source, is described. After
presentation of the data analysis, the paper concludes with a discussion of the
implications of the findings for online stores.

Conceptual foundation
A website consists of multiple attributes and several attempts have been made to list
and categorize these attributes. The early attempts aimed at building a list of attributes.
For instance, Emerick (1995) identified several attributes adopted by internet presence
sites (IPSs) while Marrelli (1996) specifically analyzed the Zima website (www.zima.
com) and emphasized the role of operational attributes such as e-mail feedback loop to
build a highly interactive IPS. Continuing this stream of research, Ghose and Dou (1998)
listed 23 attributes used by IPSs and further classified them into five groups: customer
support, marketing research, personal-choice helper, advertising/promotion/publicity,
and entertainment while Lii et al. (2004) listed eight attributes, termed operational
factors, including content, attractiveness, ease of use, personalization, interactivity,
online community involvement, security, and maintenance level.
Research soon followed to differentiate attributes along various dimensions.
Zeithaml et al. (2002) differentiated attributes critical in the online environment from
those important in offline shopping. The focus of Dou et al. (2002) was to identify
attributes more relevant for communication sites as distinct from those more relevant
IJRDM for transactional sites. Burke (2002) organized 31 website features into four groups and
38,7 used consumer surveys to list “must have” and “should have” attributes.
Given the large number of possible attributes as well as the changing nature of
technology that makes new attributes increasingly possible, it is not surprising that
there is a lack of consensus regarding “must have” and “optional” attributes. Using
Ghose and Dou’s (1998) classification of website attributes, Table I lists potential
484 attributes across four different types of websites – communication, entertainment,
information, and transaction (online store). An “X” indicates that a specific attribute is
more likely to be included in corresponding type(s) of websites; however, a blank does
not mean that the attribute cannot be included in a specific type of website. For example,
the attribute “games” is more likely to be included in an entertainment website; however,
some information websites such as www.electrolux.com also offer several games that
can be related to the firm’s products, even though “games” is not a typical attribute for an
information site.
Table I suggests that online stores are “attribute rich” – potentially containing the
maximum number (16) of the 25 specific attributes. With so many website attributes

Website type
Attribute Information Entertainment Transaction (e-tail) Communication

Personal choice helper

Keyword search X X X X
Recommendation X
My account/file X X X X
Dealer location X
Virtual reality display X X X
Customer support
Software downloading X X
Online problem diagnostics X
E-form inquiry X X X X
Order status tracking X
Comment X X X X
Feedback X X X X
Marketing research
Site survey X X X X
Product survey X
New-product proposal X
Advertising, promotion, and publicity
Electronic coupon X
Online order X
Bulletin board X X
Chat room X
Short message X
E-mail X
Sweepstakes/prize X
Banner ads X X X X
Entertainment
Table I. Games X
Types of websites and Surfer postings X X
typical website attributes E-card X
to choose from, managers of online stores have to decide on what specific attributes to Effects of online
include as well as how to operationalize a specific attribute such as “customer support.” store attributes
These variations create important challenges for research on the relationships between
individual attributes and customer satisfaction and loyalty.

Relationship between store attributes, satisfaction, and loyalty

Our knowledge of the relationship between store attributes and customer loyalty and 485
satisfaction is still evolving. Not only are store attributes deployed differently in the off
and online environments, but the measures of loyalty/satisfaction differ as well. There is
little agreement about the impact of store attributes on loyalty and on satisfaction. These
issues are briefly discussed below.
Store attributes. The changing competitive environment affects store attributes
within both offline and online store formats. For instance, product assortment and retail
location have a great deal of influence on customer behaviours in offline stores (Baltas
and Papastathopoulou, 2003; Carpenter and Moore, 2006; Hansen and Solgaard, 2004).
Physical co-location of customers and salespeople as well as the presence of sensory cues
appear to be important as well. For instance, Baker (1986) proposed a general typology
that includes social factors (the people in the store, customers, and employees) and
ambient factors (non-visual cues such as smells and sounds). Similarly, Bitner (1992) in
her focus on the service retail context includes ambient cues (those cues that affect the
five senses).
Translation of offline store attributes into the online world is not easy. Eroglu et al.
(2003) propose two types of cues: high task-relevant cues (i.e. descriptions of the
merchandise, the price, terms of sale, delivery and return policies, pictures of the
merchandise, site maps, etc.) and low task-relevant cues (colors, borders, and
background patterns, typestyles and fonts, animation, music and sounds, entertainment,
decorative pictures, a web counter, site awards, and affiliations). They also argued that
high task-relevant cues influenced the utilitarian value of shopping while low
task-relevant cues influenced hedonic value.
Satisfaction and loyalty. In addition to differences in attributes, measures of
satisfaction and loyalty have been implemented differently in the contexts of online and
offline retail stores. For offline retail stores, loyalty is primarily measured through repeat
patronage (Sivadas and Baker-Prewitt, 2000; Corstjens and Lal, 2000). Even though the
importance of customers’ length of store visit has been recognized in both academia and
practice (Morrison, 2001), it has not been emphasized in research.
In the online environment, additional concepts of e-satisfaction (Szymanski and Hise,
2000) have been proposed. Park and Kim (2003, p. 18) proposed a specific concept of
“information satisfaction” conceptualized as “an emotional reaction to the experience
provided by the overall information service.”
Concepts of e-loyalty have also been proposed. Srinivasan et al.’s (2002) proposed
concept differed from the length of visit (or stickiness) offered as one of the important
measures of a website’s operational effectiveness (Crockett, 2000; Lii et al., 2004) and
associated with increased probability of requesting additional webpage (Bucklin and
Sismeiro, 2003). There have been few efforts, however, to measure stickiness
(Bansal et al., 2004; Dubelaar et al., 2003).
Online attributes and their effects. Using various measures of satisfaction and
loyalty, researchers have attempted to identify their relationships with online
IJRDM attributes. Eroglu et al. (2003) relate the influence of high- and low-task relevant
38,7 attributes on satisfaction mediated through responses such as pleasure and arousal.
Wu et al. (2008) used atmospheric factors (music and color), termed as low-task
relevant attributes by Eroglu et al. (2003), and found the two factors to affect
participants’ emotional response which in turn influenced their intentions to purchase.
Ballantine (2005) designed an online retailing experiment with digital cameras and
486 found interactivity and product information to be positively related to customer
satisfaction. Kim and Kim (2006), using multi-dimensional scaling, identified three
dimensions of online retailing – safe purchasing, shopping convenience, and vendor
reliability – that relate to customer satisfaction and identified online store attributes
such as delivery information and secure transaction over the internet that underlie each
of the dimensions. Srinivasan et al. (2002) surveyed online consumers to determine
which of eight factors impacted e-loyalty and found that customization, contact
interactivity, cultivation, care, community, choice, and character were positively
related to e-loyalty while convenience had no impact on e-loyalty. Yun and Good (2006),
on the other hand, focused on image attributes that affect e-patronage intentions and
e-loyalty. In their review of the literature, Bansal et al. (2004) find no consensus on the
drivers of e-quality and e-satisfaction.
Finally, timing plays an important influence on the relationship between online store
attributes and satisfaction. Posselt and Gerstner (2005), using the theory of order effect,
argue that satisfaction with a service will be influenced by the sequence of service
encounters. This becomes particularly important to non-store, including online, retailing
where there is a temporal separation between order placement, and delivery of ordered
merchandise. For instance, website design and visual appeal are more important in the
search stage and product assortment and security are more important in the ordering
stage (Koo, 2005). Otim and Grover (2006) separate website features into pre-purchase,
transaction-related, and post-purchase services to determine their impact on customer
loyalty and find post-purchase services to be most important.
The literature review above suggests that our knowledge regarding online store
attributes and their effects is still evolving. There is no consensus regarding store
attributes or the process by which they impact satisfaction and loyalty. Some of the
researches have focused on mediating variables such as pleasure and arousal
(Eroglu et al., 2003); others have used specific definitions of satisfaction such as
information satisfaction which is distinct from “overall satisfaction that refers to the
consumers’ overall evaluation” (Park and Kim, 2003). The following study is an attempt
to shed additional insights into the relationships between website attributes and their
effects on customer satisfaction and loyalty.

Empirical investigation
An empirical study was designed to test whether online retail attributes affect users’
satisfaction with and repurchase intentions from the website. Secondary data, collected
at the online store level, were compiled directly from www.bizrate.com which collects,
analyzes, and reports real customers’ ratings of online stores. Bizrate combines
consumer feedback from two different customer groups – directly from online
customers as they make purchases; as well as from a panel of Bizrate members who have
volunteered to rate online stores – to create comprehensive store ratings. After
purchasing at a Bizrate affiliated website, each customer is asked to rate the website
twice: first, immediately after completing the online transaction, a store’s customer Effects of online
is asked to evaluate his/her attribute-level shopping experience (after check-out). Second, store attributes
as a “post-delivery” follow-up, the customer is again asked to rate the remaining
shopping experience attributes, overall shopping experience and revisit intentions. An
online store’s ratings are publicly reported only when at least 30 customers have rated
the store. Each rating is the weighted average of evaluations from both customers and
Bizrate members during the past 90 days. 487
The major advantage of using the Bizrate data is that it collects data twice –
immediately after placing the order and again after receipt of the ordered merchandise.
As reviewed earlier, the influence of website attributes depends on the decision-making
stage (Koo, 2005; Otim and Grover, 2006; Posselt and Gerstner, 2005). The external
validity of the data is assumed to be high since the store ratings are collected from real
customers making purchases at actual merchant sites for a large number of product
categories.
There are several limitations to the data as well. The universe of online stores
changes regularly as older stores die or opt out of the Bizrate measures and new ones
join. Although all customers are asked to complete the merchant rating two times, not
all customers choose to respond both times. The set of attributes selected by
bizrate.com does not correspond exactly with the existing literature. Finally, the second
rating does not exactly replicate the first set of attributes used in the post-order
measures. Despite these disadvantages, the Bizrate data were considered to be quite
useful.

Data collection
Data were collected at different points of time – once in August 2003 and again in
December 2004. Both datasets were compiled directly from www.bizrate.com and
collected at the online store level. By December 2004, the number of stores included in the
Bizrate database had grown from 1,079 online stores to 1,242 stores for which data were
available. All available data for each time period were collected – 1,079 (2003) and 1,242
(2004) ratings.

Measures
Rating attributes and their explanations from www.bizrate.com are listed in Table II.
Shoppers and panel members rate online stores on eight attributes after checkout,
including “overall look and design of site” and on seven attributes post-delivery,
including satisfaction and repurchase intentions. Each item is rated on a 1-10 scale.

Analysis and results

Regression analysis was the primary analytical procedure used. First, multicollinearity
among the 13 attributes was assessed using the variance inflation factors (VIFs).
According to Belsley et al. (1980), a VIF of less than 10 indicates lack of a significant
multicollinearity; all 13 had a VIF of less than 10 and 12 of 13 VIFs in our model were less
than 5, indicating a low level of mulitcollinearity. Factor analysis of 12 attributes
(excluding “overall look and design of site”) indicates that the time at which the
measures were taken contributed to the factor loadings; hence, the two factors are
labeled “post-order” and “post-delivery” (Table III). The two factors explained 68 percent
of the variance in 2003 and 2004.
IJRDM
Rating Explanation Source
38,7
A. Attribute measures
Ease of finding what you are How easily were you able to find the product your After
looking for were looking for checkout
Selection of products Types of products available After
488 checkout
Clarity of product information How clear and understandable was the product After
information checkout
Prices relative to other online Prices relative to other web sites After
merchants checkout
Shipping charges Shipping charges After
checkout
Variety of shipping options Desired shipping options were available After
checkout
Charges stated clearly before order Total purchase amount (including shipping/ After
submission handling charges) displayed before order checkout
submission
Availability of product you wanted Product was in stock at time of expected delivery After
delivery
Order tracking Ability to track orders until delivered After
delivery
On-time delivery Product arrived when expected After
delivery
Product met expectations Correct product was delivered and it worked as After
described/depicted delivery
Customer support Any post-purchase activity such as: questions, After
complaints, replacements, and returns delivery
B. Overall measures
Overall look and design of site Overall look and design of the site After
checkout
Table II. Would shop here again Likelihood to buy again from this store After
Attribute ratings and delivery
their explanations Overall rating Overall experience with this purchase After
in Bizrate delivery

Regression analysis
The first analysis focused on the effects of individual attributes on site design. In both
2003 and 2004, the two most important attributes were “ease of finding what you
are looking for” and “clarity of product information” (Table IV); “product selection”
was not a significant contributor to the evaluation of site design. Similarly “clearly
stating charges before order submission” and “variety of shipping options” were
more important than “shipping charges” per se. Overall, the relationship between the
measured attributes and overall site design improved between the two years.
The primary analysis focused on the effects of individual online store attributes on
customer satisfaction and repurchase intentions (Table V). Both satisfaction and
repurchase intentions were measured post-delivery although consumers also rated
the online store soon after checkout. We find that satisfaction is impacted most by
“on-time delivery,” “product meeting expectations,” and “customer support.” “Product
availability” was also important. “Ability to track orders until delivered,” however,
 Effects of online
Factor 1 Factor 2
“post-delivery” “post-order” store attributes
2003 2004 2003 2004

Ease of finding what you are looking for 0.770 0.826

Selection of products 0.697 0.784
Clarity of product information 0.641 0.759 489
Prices relative to other online merchants 0.723 0.709
Shipping charges 0.770 0.627
Variety of shipping options 0.679 0.711
Charges stated clearly before order submission 0.746 0.564
Availability of product you wanted 0.781 0.804
Order tracking 0.901 0.873
On-time delivery 0.927 0.901
Product met expectations 0.773 0.693 Table III.
Customer support 0.893 0.846 Factor analysis of Bizrate
Variance explained (%) 36.18 55.03 32.02 12.54 store attribute ratings

2003 data 2004 data

a*
Ease of finding what you are looking for 0.482 0.628 *
Selection of products 20.016 0.038
Clarity of product information 0.403 * 0.230 *
Prices relative to other online merchants 20.158 * 20.179 *
Shipping charges 0.019 0.001
Variety of shipping options 0.031 0.068 *
Charges stated clearly before order submission 0.090 * 0.119 * Table IV.
Adjusted R 2 0.673 0.747 Online store attributes
and overall look and
Notes: Significant at: *, 0.05; aSTD b design of site

did not contribute to the measure of customer satisfaction. All these attributes were
measured at the same time as satisfaction.
“Shipping charges,” “clarity of product information” and “variety of shipping
options” also contributed positively to satisfaction; “charges stated clearly before order
submission” was negatively related. All these attributes were measured post-checkout
and were therefore temporally separated from the measure of satisfaction.
When we examine the effect of online store attributes on repurchase intentions, all
attributes except “prices relative to other online merchants” make a significant impact.
“Customer support” is most important followed by “product meeting expectations” and
“on-time delivery.” “Availability of product wanted” and “order tracking” are also both
significant but not as important as other attributes rated at the same time (after
delivery). Among the attributes rated after checkout, “shipping charges,” “selection of
products,” and “clarity of product information” are most important.
In order to eliminate the temporal co-variance of measurement, a second set of
regression analysis was conducted with only the first set of attributes rated after
checkout. The overall variance explained is significantly lower (adjusted R 2 ¼ 0.42) but
“charges stated clearly before submission” “clarity of product information” and “variety
of shipping options” remain significant for both satisfaction and repurchase intentions
IJRDM
Satisfaction Repurchase intentions
38,7 Combined 2003 2004 Combined 2003 2004

At checkout attributes
Ease of finding what you are
looking for 0.028a * 20.01 0.041 2 0.053 * * 2 0.076 * 2 0.044
490 Selection of products 0.008 20.01 0.020 0.061 * * 0.049 * 0.064 *
Clarity of product information 0.065 * * 0.013 0.096 * 0.071 * * 0.018 0.105 *
Prices relative to other online
merchants 0.001 0.019 20.018 2 0.007 0.008 2 0.032
Shipping charges 0.062 * * 0.035 * 0.073 * 0.048 * * 0.005 0.073 *
Variety of shipping options 0.031 * * 20.014 0.055 * 0.024 * 2 0.013 0.045 *
Charges stated clearly before order
submission 20.124 * * 0.032 * 20.191 * 2 0.089 * * 0.049 * 2 0.146 *
Overall look and design of site 0.002 0.024 20.017 0.036 * 0.035 * 0.034
After delivery attributes
Availability of product you wanted 0.111 * * 0.105 * 0.128 * 0.073 * * 0.048 * 0.097 *
Order tracking 0.005 0.038 * 20.013 0.039 * 0.061 * 0.020
On-time delivery 0.416 * * 0.398 * 0.424 * 0.323 * * 0.266 * 0.353 *
Table V. Product met expectations 0.221 * * 0.241 * 0.206 * 0.224 * * 0.286 * 0.189 *
Effects of online store Customer support 0.280 * * 0.245 * 0.292 * 0.334 * * 0.338 * 0.328 *
attributes on customer Adjusted R 2 0.906 0.942 0.886 0.864 0.902 0.845
satisfaction and
repurchase intentions Notes: Significant at: *, 0.05 and * *,0.01; aSTD b

(Table VI). “Selection of products” and “site design” emerge as significant predictors of
repurchase intention but lose their significance for satisfaction.

Relationship between satisfaction and repurchase intentions

The correlation between satisfaction and repurchase intentions is very high (r ¼ 0.958,
n ¼ 2,321). This is true in 2003 when the number of stores is lower (r ¼ 0.961, n ¼ 1,079)
as well as in 2004 (r ¼ 0.956, n ¼ 1,242).

Satisfaction Repurchase intentions

2003 2004 2003 2004

After checkout attributes

Ease of finding what you are looking for 0.014a 0.083 20.076 2 0.004
Selection of products 0.027 0.098 * 0.086 * 0.134 *
Clarity of product information 0.268 * 0.290 * 0.277 * 0.293 *
Prices relative to other online merchants 20.022 2 0.045 20.04 2 0.077 *
Shipping charges 0.023 0.052 * 20.001 0.056 *
Variety of shipping options 0.147 * 0.198 * 0.143 * 0.186 *
Charges stated clearly before order submission 0.242 * 0.142 * 0.260 * 0.172 *
Table VI. Overall look and design of site 0.073 2 0.028 0.098 * 0.027
Effects of online store Adjusted R 2 0.415 0.422 0.410 0.422
attributes on customer
satisfaction Notes: Significant at: *, 0.05; aSTD b
Changes over time Effects of online
The data analysis also captured some of the changes in the online store attributes’ store attributes
contribution to customer satisfaction and repurchase intention ratings. We used Chow
(1960) tests to examine the differences; the regression coefficients are different between
2003 and 2004 data when all variables are included to determine their effects on
satisfaction (F ¼ 13.93, df ¼ 14/2,293, p ¼ 0.000) and repurchase intentions (F ¼ 11.60,
df ¼ 14/2,293, p ¼ 0.001). Similar results are achieved when Chow tests are conducted to 491
test the coefficient differences of various regression models for the two subsets –
post-order and post-delivery (Table VII). The results show that online store attributes
contribute differently to customer satisfaction and repurchase intentions between 2003
and 2004.
Based on the analyses, the attributes are categorized into the following five groups:
(1) The first group is composed of those attributes that significantly and positively
impact satisfaction and repurchase intentions in both years:
“On-time delivery,” “customer support,” “product met expectations,” and
“product availability.” All these attributes relate to order fulfillment.
(2) The second group includes attributes that have an impact in both years, but the
direction changes:
“Charges stated clearly before order submission” becomes negative in its
impact in 2004 for both satisfaction and repurchase intentions. This attribute
relates to pre-order attributes measured immediately after checkout.
(3) In the third group are those attributes that have an impact both years, but not
on both satisfaction and repurchase intentions:
“Product selection” impacts repurchase intentions in both years but not
satisfaction; “shipping charges” impact satisfaction in both years but not
repurchase intentions in 2003.
(4) Fourth, those attributes that have an impact in 2003 but lose significance in 2004:
“Order tracking” is significant for satisfaction and repurchase intentions in
2003 but loses its significance in 2004; “ease of finding” impacts repurchase
intentions in 2003 but not in 2004.
(5) Finally, those attributes that gain significance in 2004:
“Variety of shipping options” and “clarity of product information”
significantly impact both satisfaction and repurchase intentions in 2004.

Discussion and conclusion

With online retailing continuing to grow steadily, we can also expect greater
experimentation with website attributes along with the concerns of online store

DV
IV Satisfaction Repurchase intentions

All variables F(14, 2,293) ¼ 13.93, p , 0.001 F(14, 2,293) ¼ 11.60, p , 0.001 Table VII.
Checkout variables F(9, 2,303) ¼ 4.19, p , 0.001 F(9, 2,303) ¼ 4.22, p , 0.001 Chow test – comparing
After delivery variables F(6, 2,309) ¼ 8.66, p , 0.001 F(6, 2,309) ¼ 9.55, p , 0.001 2003 data and 2004 data
IJRDM managers about the impact of an attribute or a set of attributes on customer satisfaction
38,7 and loyalty. Our literature review had suggested that there is no consensus regarding
online store attributes and how they affect customer satisfaction and loyalty. Based on
the analysis of this rather large data set of customer ratings, we can draw some
conclusions regarding the influence of online store attributes on satisfaction and repeat
purchase intentions:
492 .
First of all, fulfillment attributes dominate satisfaction and repurchase intention
judgments. Fulfillment variables are the biggest challenge to all non-store
retailing, including online retailing. Our analysis found “on-time delivery” to be
the most important attribute impacting satisfaction; its impact on repurchase
intentions increases in 2004. “Customer support” is next in importance. “Order
tracking” is less important and loses its significance in 2004. Otim and Grover
(2006) had also found fulfillment attributes to dominate the relationship with
customer loyalty but order tracking was significant in their study but not in this
analysis. While our results are consistent with Posselt and Gerstner (2005) who
had also looked at 2004 Bizrate data, our analysis is able to report on the stability
of importance ratings over a period of two years.
.
“Ease of finding what you are looking for” and “clarity of product information” are
the two most important attributes for generating positive ratings of overall look
and design of the site. This is consistent with other researchers who have focused
on the content and presentation of product and service information (Park and Kim,
2003). As Baker et al. (2002) conclude, design cues are the most significant and
consistent influence on shopping experience and e-stores have to pay attention to
design factors such as appearance and layout of home pages. This is particularly
important at the search stage (Koo, 2005).
.
The analysis also suggests how an online attribute is implemented makes a
difference. For instance, while price/cost-related attributes may be implemented in
various ways, “charges stated clearly before order submission” is very important;
it influences “overall look and design of site” as well as satisfaction and loyalty.
“Prices relative to other online merchants,” on the other hand, did not impact any
of the measures. For shipping-related attributes – both “variety of shipping
options” and “shipping charges” – which gained significance in 2004 needed to be
included for favorable satisfaction and repurchase intention judgments.
.
When there is a time gap between interacting with a site and making evaluative
judgments, not all dimensions of the interactions persist in their impact.
As Posselt and Gerstner (2005) noted, the importance of fulfillment variables may
be attributed to “the recency effect” accentuated by the proximity of measurement.
We had attempted to tease out this issue by analyzing the influence of attributes
measured after checkout separately (Table VI). It is clear that “charges stated
clearly before order submission” is very important – it is significant when we look
at its impact on site design (measured concurrently), and on satisfaction and
repurchase (measured later). “Clarity of product information” and “variety of
shipping options” are significant for both satisfaction and repurchase intentions
only when the set of attributes measured after checkout is considered; this
suggests that their salience gets drowned by other attributes measured
concurrently with satisfaction and repurchase intentions. “Ease of finding
 what you are looking for” and “selection of products” (site design factors) have Effects of online
different relationships to satisfaction and repurchase intentions suggesting that store attributes
recall of these attributes follows a path different from the other attributes. This
may explain why Otim and Grover (2006) did not find impact of pre-purchase
attributes on customer loyalty.
.
It is not surprising that fulfillment variables such as “on time delivery,” “product
met expectations” dominate ratings after product delivery. This is consistent with 493
Wolfinbarger and Gilly’s (2003) prediction of e-tail quality, and matches the
findings from 2004 Bizrate data analyzed by Posselt and Gerstner (2005) as well as
research conducted by Otim and Grover (2006).

These fulfillment variables are significantly different for all non-store retailing,
including catalog and online retailing. In offline stores, customers can touch and
examine products before they buy, which assures them of product meeting expectations
(Peck and Childers, 2003). Buyers usually take possession immediately after a purchase;
so on-time delivery is not an issue. Similar haptic information, however, is not available
online; neither is immediate possession. Therefore, on-time delivery is very important.
Furthermore, when online stores compete with each other, it is easier to copy certain
attributes like “shipping options” than other attributes such as “on-time delivery” which
requires distinct organizational and logistical capabilities. These results suggest that
the most creative, interactive, vivid online site will not compensate for weak fulfillment
and customer support capabilities. To satisfy online buyers and attract them to
repurchase, a website should control out-of-stock conditions, enable customers to track
orders online, deliver on time, ensure that the descriptions on the site actually match the
product, and offer timely support to customers with questions and problems.
The changes from 2003 to 2004 uncovered by our analysis hint at the learning taking
place in the online environment. The shifts capture some of the dynamics of the online
environment as e-tailers learn from each other and their own experiences and improve
existing attributes and introduce new ones. As individual attributes become more
standard or commonplace, their importance in predicting customer satisfaction and
repurchase intentions also change. They become necessary but not sufficient conditions
of satisfaction. This dynamic situation requires continuous monitoring of the online
environment.

Limitations and future directions

Methodological issues need to be considered in the interpretation of the results. Using
secondary data from Bizrate had several limitations, but the following are most relevant
for this analysis. While the customer ratings were acquired twice, the temporal
association is not the same. Some attributes were assessed after checkout and others
were assessed post-delivery. It would have been useful to obtain measures of satisfaction
and loyalty also at the time of checkout; this would have allowed capturing customer
evaluations immediately after interacting with the site. Second, the two dependent
variables – satisfaction and repurchase intentions – were both measured with a
one-item scale. Multiple-item scales would have provided more reliability. Eroglu et al.
(2003), for instance, used four items to measure satisfaction (a ¼ 0.81) but their research
design involved a hypothetical site and 48 percent of the respondents had never
purchased online. Given the attractiveness of Bizrate data involving real customers
IJRDM and real retail sites, it is likely that a longer scale with multiple items for each construct
38,7 would have depressed completion rates.
There are more attributes in a transactional website than the attributes selected by
Bizrate; therefore, our analysis was limited to the selected attributes. While our research
revealed the changing nature of relationships between online attributes and measures of
satisfaction and loyalty, it could not address attributes that are added or deleted over
494 time. Because of the repeated measures over time, however, the database allows
researchers to track the stability of relationships.
Finally, future research should address the stability of relationships between online
attributes and satisfaction and loyalty. This is complicated due to the dynamic nature of
websites and the continuing technological changes in the design of website attributes.
The challenge, however, must be tackled in order to develop a body of knowledge based
on reliable and valid research designs.

References
Baker, J.A. (1986), “The role of environment in marketing services: the consumer perspective”,
in Czepiel, J.A., Congram, C. and Shanahan, J. (Eds), The Services Marketing Challenge:
Integrated for Competitive Advantage, American Marketing Association, Chicago, IL,
pp. 79-84.
Baker, J.A., Parasuraman, A., Grewal, D. and Voss, G.B. (2002), “The influence of multiple store
environment cues on perceived merchandise value and patronage intentions”, Journal of
Marketing, Vol. 66 No. 2, pp. 120-41.
Ballantine, P.W. (2005), “Effects of interactivity and product information on consumer
satisfaction in an online retail setting”, International Journal of Retail & Distribution
Management, Vol. 33 Nos 6/7, pp. 461-71.
Baltas, G. and Papastathopoulou, P. (2003), “Shopper characteristics, product and store choice
criteria: a survey in the Greek grocery sector”, International Journal of Retail &
Distribution Management, Vol. 31 No. 10, pp. 498-507.
Bansal, H., McDougall, G., Dikolli, S. and Sedatole, K. (2004), “Relating e-satisfaction to
behavioral outcomes: an empirical study”, Journal of Services Marketing, Vol. 18 Nos 4/5,
pp. 290-302.
Belsley, D.A., Kuh, E. and Welsch, R.E. (1980), Regression Diagnostics: Identifying Influential
Data and Sources of Collinearity, Wiley, New York, NY.
Bitner, M.J. (1992), “Servicescapes: the impact of physical surroundings on customers and
employees”, Journal of Marketing, Vol. 56, pp. 57-71.
Bucklin, R. and Sismeiro, C. (2003), “A model of website browsing behavior estimated on
clickstream data”, Journal of Marketing Research, Vol. 40 No. 3, pp. 249-67.
Burke, R.R. (2002), “Technology and the customer interface: what consumers want in the
physical and virtual store”, Journal of the Academy of Marketing Science, Vol. 30 No. 4,
pp. 411-32.
Carpenter, J.M. and Moore, M. (2006), “Consumer demographics, store attributes, and retail
format choice in the US grocery market”, International Journal of Retail & Distribution
Management, Vol. 34 No. 6, pp. 434-52.
Chow, G.C. (1960), “Test of equality between sets of coefficients in two linear regressions”,
Econometrica, Vol. 28 No. 3, pp. 591-605.
Corstjens, M. and Lal, R. (2000), “Building store loyalty through store brands”, Journal of
Marketing Research, Vol. 37, pp. 281-91.
Crockett, R. (2000), “Keep ‘em coming back”, Business Week, Vol. 3681, p. EB20. Effects of online
Dholakia, R. and Zhao, M. (2009), “Retail website interactivity: how does it influence customer store attributes
satisfaction and behavioural intentions?”, International Journal of Retail & Distribution
Management, Vol. 37 No. 10, pp. 821-38.
Dou, W., Nielsen, U. and Tan, C.M. (2002), “Using corporate websites for export marketing”,
Journal of Advertising Research, Vol. 42 No. 5, pp. 105-15.
Dubelaar, C., Leong, M. and Alpert, F. (2003), “Impact of interactivity on the stickiness of online 495
gift stores”, Journal of Asia Pacific Marketing, Vol. 2 No. 2, pp. 22-41.
Emerick, T. (1995), “Media and marketing strategies for the internet: a step-by-step guide”,
in Forrest, E. and Mizerski, R. (Eds), Interactive Marketing, NTC Business Books,
Lincolnwood, IL.
Eroglu, S.A., Machleit, K.A. and Davis, L.M. (2003), “Empirical testing of a model of online store
atmospherics and shopper responses”, Psychology & Marketing, Vol. 20 No. 2, pp. 139-50.
Ghose, S. and Dou, W. (1998), “Interactive functions and their impacts on the appeal of internet
presence sites”, Journal of Advertising Research, March/April, pp. 29-43.
Hansen, T. and Solgaard, H. (2004), New Perspectives on Retailing and Store Patronage Behavior,
Kluwer Academic, Boston, MA.
Jin, B. and Park, J.Y. (2005), “The moderating effect of online purchase experience on the
evaluation of online store attributes and the subsequent impact on market response
outcomes”, Advances in Consumer Research, Vol. 33, pp. 203-11.
Kim, K. and Kim, E.B. (2006), “Suggestions to enhance the cyber store customer satisfaction”,
Journal of American Academy of Business, Vol. 9 No. 1, pp. 233-40.
Koo, D. (2005), “The fundamental reasons of e-consumers’ loyalty to an online store”, Electronic
Commerce Research and Application, Vol. 5 No. 2, pp. 117-30.
Lii, Y.-S., Lim, H.J. and Tseng, L.P. (2004), “The effects of web operational factors on marketing
performance”, Journal of American Academy of Business, Vol. 5, pp. 486-94.
Marrelli, C. (1996), “Anatomy of web advertisement”, in Forrest, E. and Mizerski, R. (Eds),
Interactive Marketing, NTC Business Books, Lincolnwood, IL.
Morrison, M. (2001), “The power of music and its influence on international retail brands and
shopper behavior: a multi case study approach”, Proceedings of 2001 Academy of
Marketing Conference, Cardiff, UK, July 1-4.
Otim, S. and Grover, V. (2006), “An empirical study of web-based services and customer loyalty”,
European Journal of Information Systems, Vol. 15 No. 6, pp. 527-42.
Park, C.-H. and Kim, Y.-G. (2003), “Identifying key factors affecting consumer purchase behavior
in an online shopping context”, International Journal of Retail & Distribution Management,
Vol. 31 No. 1, pp. 16-29.
Peck, J. and Childers, T.L. (2003), “Individual differences in haptic information processing: the
‘need for touch’ scale”, Journal of Consumer Research, Vol. 30, pp. 430-42.
Posselt, T. and Gerstner, E. (2005), “Pre-sale vs. post-sale e-satisfaction: impact on repurchase
intention and overall satisfaction”, Journal of Interactive Marketing, Vol. 19 No. 4, pp. 35-47.
Sivadas, E. and Baker-Prewitt, J.L. (2000), “An examination of the relationship between service
quality, customer satisfaction and store loyalty”, International Journal of Retail &
Distribution Management, Vol. 28 No. 2, pp. 73-82.
Srinivasan, S.S., Anderson, R. and Ponnavolu, K. (2002), “Customer loyalty in e-commence:
an exploration of its antecedents and consequences”, Journal of Retailing, Vol. 78 No. 1,
pp. 41-50.
IJRDM Szymanski, D.M. and Hise, R.T. (2000), “E-satisfaction: an initial examination”, Journal of
Retailing, Vol. 76 No. 3, pp. 309-22.
38,7 Weathers, D. and Makienko, I. (2006), “Assessing the relationships between e-tail success and
product and website factors”, Journal of Interactive Marketing, Vol. 20 No. 2, pp. 41-53.
Wolfinbarger, M. and Gilly, M.C. (2003), “eTailQ: dimensionalizing, measuring and predicting
e-tail quality”, Journal of Retailing, Vol. 79 No. 3, pp. 183-98.
496 Wu, C.S., Cheng, F.F. and Yen, D.C. (2008), “The atmospheric factors of online storefront
environment design: an empirical experiment in Taiwan”, Information & Management,
Vol. 45 No. 7, pp. 493-8.
Yun, Z.-S. and Good, L.K. (2006), “Developing customer loyalty from e-tail store image
attributes”, Managing Service Quality, Vol. 17 No. 1, pp. 4-22.
Zeithaml, V.A., Parasuraman, A. and Malhotra, A. (2002), “Service quality delivery through
websites: a critical review of extant knowledge”, Journal of Academy of Marketing Science,
Vol. 30 No. 4, pp. 362-75.

Further reading
Eroglu, S.A., Machleit, K.A. and Davis, L.M. (2001), “Atmospheric qualities of online retailing:
a conceptual model and implications”, Journal of Business Research, Vol. 54 No. 2,
pp. 177-84.

About the authors

Ruby Roy Dholakia has taught at University of Rhode Island (URI) since 1981 and has had
previous teaching experiences at Kansas State University, Indian Institute of Management –
Calcutta and Ahmedabad, India. One of her research interests lies in the areas of marketing and
market acceptance of new technology products and services. She has co-edited several books
including – New Infotainment Technologies in the Home (1996), M-Commerce: Global Experiences
& Perspectives (2006), and Marketing Practices in Developing Economy (2009) and published
extensively in journals. Her PhD is from Kellogg School of Management, Northwestern University.
Ruby Roy Dholakia is the corresponding author and can be contacted at: ruby@uri.edu
Miao Zhao, teaching since 2003, has research interests in the areas of internet marketing,
consumer behaviour, international marketing, and marketing of technology products and
services. She has published in Journal of Interactive Marketing, Managing Service Quality and
International Journal of Retail & Distribution Management. Miao Zhao has a PhD from URI.

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com

Or visit our web site for further details: www.emeraldinsight.com/reprints