Behavioral Segmentation Analysis of Online Consumer Audience in Turkey
by Using Real E-commerce Transaction Data
Farid Huseynov*
Middle East Technical University, Ankara, Turkey
Sevgi Özkan Yıldırım
Middle East Technical University, Ankara, Turkey
Abstract
Consumer segmentation is a marketing strategy which involves firstly dividing customers into
groups based on their underlying characteristics, needs and interests, and then designing and
implementing strategies to target them. One of the most common types of segmentation
approaches is behavioral segmentation analysis in which consumers are grouped based on their
certain behavioral characteristics such as decision making, spending, usage and etc. This study
carried out behavioral segmentation analysis based on real e-commerce transaction records of
10.000 online customers and found five different types of online consumer segments which are
Opportunist customers, Transient customers, Need-based shoppers, Skeptical newcomers and
Repetitive purchasers. Behavioral characteristics of each segment were discussed in detail and
recommendations were made about how to approach to each segment in order to increase their
online shopping rates. Understanding the behavioral characteristics of each segment will enable
the selling companies to develop marketing strategies accordingly.
Keywords: Online consumer behavior, Market segmentation, Segmentation analysis, Two-Step
Cluster Analysis, Behavioral segmentation analysis
*Corresponding Author: Farid Huseynov, Tel: +90 (312) 210-7711, Fax: +90 (312) 210-3745,
Email: [email protected]
1. Introduction
B2C (Business-to-Customer) e-commerce refers to the exchange of goods and services over the
internet between online sellers and customers. According to eMarketer (2014) forecasts,
worldwide business-to-consumer (B2C) ecommerce sales will reach to 1.5 trillion U.S. dollars in
2017. B2C e-commerce model enables companies to easily and cost-efficiently reach to customers,
suppliers and business partners both nationally and internationally. That is, it enables companies
to expand their market to national and international markets with minimum capital invest. On the
other side, it enables customers to carry out fast and convenient e-commerce transactions with
many national and international sellers. Research showed that online shopping is one of the top
five popular online activities in U.S.A. (Pew Research Center 2011). Online shopping was also
found to be one of the most popular online activities in Europe (Eurostat 2012). When it comes to
developing nations, while most of the internet users use the internet to socialize and get
information, less people use it to purchase products and services (Pew Research Center 2014). A
survey conducted in Turkey showed that getting social through social networking platforms,
watching movies and listening to music and reading news are top online activities of Internet users
(Huseynov and Yıldırım 2016). In this research, 71.3 percent of respondents stated that they have
online shopping experience but only 18.5 percent of respondents stated that they actively use the
internet for online shopping. These findings related to internet users activities coincides with the
findings of Turkish Statistical Institute (TSI 2014) which showed that online shopping is one of
the less common online activities in Turkey. Compared with both USA and Europe, online
shopping rates in developing nations are considerably less. Deloitte (2015) assessed the share of
e-retailing in total retailing in both developed and developing nations. According to this research,
while the average of developed nations was found to be 6.5 percent, this figure was found to be
4.5 percent in developing nations. In Turkey, online retail transactions constituted only 1.6 percent
of total retail sales in 2014 and this ratio was well below the averages of both developing and
developed nations. Despite to the very high credit-card penetration rate and very improved
logistics infrastructure, online shopping is not getting enough attention in Turkey.
In order to improve the online shopping rate, especially in the countries in which online shopping
rate is low, it is important to carry out careful examination of consumer shopping behavior.
Understanding consumer shopping behavior is very important step in developing successful
marketing strategies. In the relevant literature numerous researches were carried out by both
practitioners and academicians in order to understand consumer behavior in online shopping
platforms. Among these researches, it is possible to come across with consumer segmentation
studies. Customer segmentation is the practice of dividing an existing or potential customer base
into sets of similar individuals that are related from a marketing perspective. Segmentation process
is the first step of three-stage consecutive process which is followed by targeting and positioning
stages. In the targeting stage, companies decide on which segments they should follow. Finally, in
the positioning stage, companies develop or optimize their products and services for the selected
segments and develop a marketing mix for each selected segment. Successful customer
segmentation strategies enable companies to determine their most and least profitable customers,
focus their marketing efforts on the most profitable customers, improve their relationships with
existing customers, improve their products, improve their customer services and use scarce
company resources more efficiently.
2. Literature Review
The main objective of market segmentation is to divide a broad target market into groups of
consumers who share common needs, expectations and interests so that customized strategies can
be designed to target each determined segment. There are different types of market segmentation.
The most common types of them are geographic, demographic, psychographic and behavioral
market segmentation. While in geographic segmentation target market is segmented according to
geographic criteria such as countries, regions and cities, in demographic segmentation grouping is
carried out based on factors such as age, sex, occupation, education level, religion and etc.
Psychographic segmentation approach takes into consideration consumers’ lifestyle, activities,
interests and opinions to define market segments. In contrast to other types of segmentation
analysis, in behavioral segmentation process consumers are divided into groups based on various
types of behaviors they exhibit during shopping. Behavioral factors are considered to be the best
starting point for building market segments (Keller and Kotler 2012).
In the relevant literature researches carried out various types of segmentation analysis on online
consumer audience. Conducted studies showed that online consumer audience is not a single
market segment. That is, online consumer audience is comprised of different segments whose
members perceive online shopping differently and respond in different ways to the marketing
efforts. Some of the findings from selected articles are as follows. Ganesh et al. (2010) carried
out online consumer segmentation analysis based on online shopping motivations and e-store
attribute importance measures. While seven online shopper segments were found based on
shopping motivation measures, six segments were discovered based on e-store attribute
importance measure. In the study of Pradas et al. (2013), a segmentation analysis was carried out
on non-shoppers in online stores based on the barriers which discourage them to make online
purchases and the drivers which might encourage them to shop online. As a result of their
segmentation analysis, four different non-shopper segments were identified based on the barriers
against online shopping. On the other hand, six different non-shopper segments were found based
on the drivers to start purchasing on online stores. In another research which was conducted in
U.S. and Belgium by Brengman et al. (2005), segmentation analysis was carried out based on Webusage-related lifestyle scale. Segmentation based on Web-usage-related lifestyle resulted in the
same segments (i.e., four online shopping segments and four online non-shopper segments) in both
countries. In a study conducted in China by Liu et al. (2015) six different online consumer
segments were determined by using real e-commerce transaction data. Later, each segment’s
sensitivity to different promotion strategies were tested and the results showed that each segment
responds differently to different promotion strategies. In the literature, it is also possible to find
studies that carried out segmentation analysis based on consumer decision making style (Rezaei
2015), consumer cognitive style and involvement (Wang et al. 2006), shopping orientation (Gehrt
et al. 2012), shopping motivation (Hill et al. 2013), internet usage pattern (Aljukhadar and Senecal
2011) and relative importance of e-store attributes and features (Chen et al. 2010). All of these
studies showed the existence of different consumer segments with different attitudes, perceptions
and motivations toward online shopping.
3. Research Question
Extensive literature review on consumer behavioral issues in online shopping platforms showed
that consumer segmentation analysis is one of the less investigated fields and needs further
investigation (Huseynov and Yıldırım 2015). In the relevant literature there are barely any
researches that carried out behavioral segmentation analysis especially on real e-commerce
transaction data. This research aims to fill this gap in the literature by determining the different
types of online consumer segments and their main characteristics by carrying out segmentation
analysis based on real e-commerce transaction data.
The main research questions of this study are as follows: what segments exist in a broad online
consumer audience and what are their main behavioral characteristics in terms of online shopping?
Answers to these research questions are expected to show online retailers what types of online
consumers there exist and what their main behavioral characteristics are so that they can decide on
segments to go after and develop more effective marketing strategies for each segment to be
followed.
4. Method
Behavioral segmentation analysis was carried out by utilizing real B2C e-commerce data which
was based on online shopping transactions carried out at Markafoni.com. Markafoni.com is one
of the most popular private online shopping websites in Turkey with 7.2 million members.
Markafoni.com offers its members many selected brands from different categories such as
clothing, accessories, cosmetics, decoration and lifestyle. The data required for behavioral
segmentation analysis were provided by Markafoni.com after signing the non-disclosure
agreement with the company. The dataset which was provided by the company was comprised of
350.000 online shopping transaction records of 10.000 unique customers. Random sampling
method was utilized while retrieving customers and related transaction records from the database.
Several behavioral factors were extracted from the dataset to be used in segmentation analysis.
The extracted behavioral factors had different dimensions; therefore, data standardization was
carried out before conducting the segmentation analysis. After the data standardization process,
segmentation analysis was carried out by using the Two-Step cluster analysis technique in IBM
SPSS version 17. The Two-Step cluster analysis procedure is an exploratory tool which enables
to determine natural groupings within a dataset. The algorithm of this procedure can handle very
large datasets and it can also automatically determine the optimal number of groupings.
5. Data Analysis
5.1. Segmentation process
The Dataset provided by Markafoni.com contained online shopping details of 10.000 unique
customers. Demographic profile of the customers involved in behavioral segmentation process is
given in Table 1. While female customers constituted approximately 74 percent of the provided
dataset, male customers had a share of only nearly one-fourth (26%). Approximately half of the
dataset (53%) was composed of customers who aged from 25 to 34.
Table 1 - Demographic profile of consumer set
Demographic Profile
Gender
Male
Female
Age
Less than 25
25-29
30-34
35-39
40-44
45 and above
Frequencies
Percentage (%)
2616
7384
26.16
73.84
817
2428
2839
1931
1024
961
8.17
24.28
28.39
19.31
10.24
9.61
Table 2 shows behavioral factors used in segmentation process. These factors are shopping rate,
price payment, coupon redemption, product diversity and refund rate. In addition to these
behavioral factors, several other factors which were not included in segmentation analysis but used
to get more insight about the behavioral characteristics of each segment are as follows: online store
membership duration, recent online store visit, free shipping usage rate, subscription rate to receive
campaign related news and credit card storage rate for one-click payment option.
Table 2 - Behavioral segmentation factor list
Factor
Shopping rate
Description
Total number of products purchased by a consumer.
Price payment
Average price of online products purchased by a consumer.
Coupon redemption
How often a consumer uses discount coupons in online shopping
transactions.
Product diversity
It refers to extent to which a consumer purchases diverse product types.
Refund rate
How often a consumer returns products purchased online.
Each of the behavioral factor used in the segmentation analysis had a different dimension.
Therefore, data standardization was required before conducting the segmentation analysis. Data
standardization is an important step in segmentation analysis and it is required to eliminate
dimensional influences of different data items (Liu et al. 2015). Data standardization was carried
out as follows. Initially, for each behavioral factor (Table 2) the original data is sorted in an
ascending order and quintile points were identified. Quintiles are four cut-off points that split a
dataset into five equal parts, each being 20 percent of the range. While the first quintile represented
the lowest 20 percent of the range, the final quintile represented the highest 20 percent. For each
behavioral factor, original data is labelled between 1 (Low) and 5 (High) based on the determined
quintile points. Later, standardized data was further rescaled to have a mean of zero and standard
deviation of one. After the data standardization process, behavioral factors listed in Table 2 were
input into the Two-Step cluster analysis in IBM SPSS version 17. The algorithm of the Two-Step
clustering approach has two main steps. In the first step, pre-clustering, factor scores are grouped
into several small sub-clusters. In the second step, agglomerative hierarchical clustering, the small
clusters generated in pre-clustering step are used in generating groups of larger clusters. In contrast
to other types of clustering tools, the Two-Step clustering algorithm can handle very large datasets
and it can also automatically determine the optimal number of clusters (SPSS Inc. 2001). The TwoStep cluster analysis was conducted with Log-likelihood distance measure and BIC (Schwarz's
Bayesian) clustering criterion. Low “BIC Coefficient”, high “BIC Change” and high “Ratio of
Distance Measure” are criterions that the Two-Step cluster algorithm takes into consideration
while automatically determining the number of segments. The Two-Step clustering algorithm
extracted five segments from the submitted dataset (BIC=16027.41, BIC change=-1306.71, Ratio
of distance measures=1.40).
The results of segmentation analysis are given in Table 3. Based on these results, a radar chart was
constructed to show each segment’s average scores (i.e., centroid values) on each of five
behavioral factors (Fig. 1). Segments’ score on each behavioral factor is plotted along a separate
axis in a radar chart. The value 0.0 in the radar chart represents the entire sample mean and the
numbers lower or higher than 0.0 reflect the number of standard deviations from the average of
entire sample. Each segment was given an appropriate name based on the patterns observed in the
radar chart.
Table 3 - SPSS Two Step Cluster Analysis results
Factors
Opportunist
Customers
Transient
Customers
Need-based
Shoppers
Skeptical
Newcomers
Repetitive
Purchasers
Demographic
factors
Segmentation
factors (Centroids)
Segment size
Age
13.7%
34.7
24.8%
33.5
15.8%
33.6
21.8%
32.5
23.9%
35.8
37.2%
30.6%
31.6%
16.5%
20.5%
62.8%
69.4%
68.4%
83.5%
79.5%
0.99
0.19
-0.74
1.01
-0.43
-0.54
0.56
1.12
0.05
-0.77
0.35
-0.97
-0.37
-0.11
0.55
-0.86
0.19
1.06
-0.89
-0.32
Gender
Male
Female
Refund rate (RR)*
Shopping rate (SR)*
Product diversity (PD)*
Evaluation factors
Coupon redemption (CR)*
0.44
0.08
0.97
-0.90
-0.16
Price payment (PP)*
Credit card storage
4.3%
2.6%
3.3%
2.7%
5.5%
Free shipping usage (FR)*
0.25
0.04
0.11
-0.28
0.00
Membership duration (MD)*
0.12
0.14
-0.20
-0.20
0.11
Recent visits (RV)*
0.14
-0.21
-0.09
-0.11
0.30
Subscriptions
SMS
93.1%
94.9%
93.8%
94.6%
93.6%
e-Mail
68.0%
65.5%
66.2%
68.4%
66.4%
Note: *Shows standardized values (i.e., 0.00 represents the entire sample average and +/- represents number of
standard deviations above or below the entire sample average.)
Opportunist Customers
Refund rate
1.5
1.0
Need-based Shoppers
0.5
Skeptical Newcomers
0.0
Repetitive Purchasers
-0.5
Price payment
Transient Customers
Shopping rate
-1.0
-1.5
Coupon redemption
Product diversity
Fig. 1 – Behavioral characteristics of consumer segments
5.2. Description of Segments
The following subsections discuss the distinguished characteristics of each determined consumer
segment by taking into consideration segments’ scores on various behavioral factors (Fig. 1, Table
3) and results of variable importance charts (Fig. 2-6). Variable importance chart shows how
important the different behavioral factors are to the formation of each consumer segment. In the
variable importance chart, the vertical axis shows factors used in segmentation process and
horizontal axis indicates Student’s t statistics. While a negative Student’s t statistic indicates that
the particular factor predominantly takes smaller than average values within this segment, a
positive Student’s t statistic indicates the opposite. A particular factor is considered to statistically
significantly contribute to the formation of a segment if it’s bar line crosses the critical levels (i.e.,
p<0.05), either in positive or negative direction.
5.2.1. Opportunist Customers
Among all five segments, this segment is the one with the smallest size (13.7%). Consumers in
this segment are predominantly old members (MD=0.12) who visit the online store more
frequently (RV=0.14) than the entire sample’s average visit rate. Frequent online store visits by
this type of consumers resulted in higher online shopping rate (SR=0.19). Opportunist customers
generally shop for a specific type of products which can be understood from very low score of
product diversity factor (PD=-0.74). The most prominent behavioral characteristics of Opportunist
customers are their high usage of discount coupons (DC=1.01) and free shipping offers (FR=0.25)
in online transactions. Opportunist customers are more inclined to save money by taking advantage
of discounts and free shipping opportunities. Another notable point about Opportunist customers
is their fast and uncertain decision-making behavior while deciding on online products for
purchasing which can be inferred from their significantly higher refund rate than average
(RR=0.99). Furthermore, Opportunist customers tend to purchase products whose prices
(PP=0.44) are higher when compared with the entire sample average. In online shopping platforms,
certain amounts of discount coupons are usually offered when customers purchase products with
high prices. Higher price payment by Opportunist customers is probably due to their motivation to
obtain discount coupons. By taking into consideration all the points mentioned above, it makes
sense to call this segment as Opportunists. This type of customers are the ones who visit the online
store regularly, redeem more discount coupons and give importance to the free shipping options.
Variable importance chart (Fig. 2) shows that all of the behavioral factors significantly contribute
to the formation of this segment. All factors, except product diversity rate, generally take scores
larger than average within this segment. Coupon redemption, refund rate and product diversity rate
are the factors that contribute more to segment formation. The findings of variable importance
chart correspond with the factor centroid results (Table 3) of Opportunist customers.
Fig. 2 – Variable importance chart (Opportunist Customers)
4.2.2. Transient Customers
Transient customers segment is the one with the largest size (24.8%). Customers in this segment
are generally old members (MD=0.14). Transient customers visit the online store infrequently
(RV=-0.21) and they purchase online products occasionally (SR=-0.54). When they do online
shopping, customers of this segment purchase very diverse type of products (PD=0.56).
Furthermore, Transient customers generally do not return the products purchased online (RR=0.43) which shows their good decision-making characteristics. The outstanding behavioral feature
of Transient customers is their high usage of discount coupons (DC=1.12) in online shopping
transactions. More specifically, among all five segments, this segment is the one with the highest
usage rate of discount coupons. Transient customers’ online shopping rate is low but their discount
coupon redemption rate is high which means that most of their online product purchases were
triggered by offered coupons. In the light of findings above, it is reasonable to call this segment as
Transients. Consumers in this segment are mostly coupon-prone ones who do not visit the online
store frequently. What drives them to the online store is the availability of attractive discount
coupons. The results of variable importance chart (Fig. 3) coincide with the results of factor
centroids (Table 3) of Transient customers by showing that all behavioral factors, except average
price payment, significantly contribute to the formation of Transient customers segment. Price
payment factor is not important to segment formation. Coupon redemption and product diversity
are the factors that generally take scores higher than average within this segment. On the other
side, shopping rate and refund rate factors predominantly take lower scores than average within
this segment. Coupon redemption factor is the one that plays a major role in the formation of the
Transient customers segment.
Fig. 3 - Variable importance chart (Transient Customers)
4.2.3. Need-based Shoppers
This segment is the second lowest segment in size with a percentage of 15.8. Consumers in this
segment are generally old members (MD=-0.20) who visit the online store occasionally (RV=0.09). One of the most prominent behavioral characteristic of Need-based shoppers is that this type
of consumers generally purchase very diverse type of products (PD=0.35) which are very high in
price (PP=0.97). Customers usually get worried while purchasing high price items from the
Internet due to several reasons. In a survey study, majority of the respondents stated that while
shopping online they were concerned that they might not get what they actually ordered and the
product they purchased might get lost on delivery (Huseynov and Yıldırım 2016). Probably,
consumers in this segment prefer purchasing high price items from online store due to perceived
benefits of online shopping such as shopping convenience, broader product selection,
attractiveness of prices and etc. These perceived benefits were found to be positively correlated
with the amount of money spent online (Forsythe et al. 2006). Another notable characteristic of
this type of consumers is that they generally do not redeem discount coupons in online shopping
transactions (DC=-0.97). Redemption of discount coupons in this segment is so rare that it has the
lowest rate of discount coupon usage among all other segments. Online shopping rate of this type
of consumers is very low (SR=-0.77) which is an expected behavior by taking into consideration
this segment’s need-oriented shopping for high price items. By taking into consideration the
findings stated above this segment of consumers are called Need-based shoppers as they generally
visit the online store only to purchase products which they have decided on. Their online shopping
rate seems not to be influenced by offered discount coupons but by their actual product needs.
Fig. 4 – Variable importance chart (Need-based Shoppers)
According to variable importance chart (Fig. 4), all behavioral factors, except refund rate,
significantly contribute to the formation of Need-based shoppers segment and these findings
correspond to the results of factor centroids (Table 3) of Need-based shoppers. In this segment,
price payment and product diversity factors generally take scores which are higher than average,
while coupon redemption and shopping rate factors predominantly take scores which are lower
than average. Price payment and coupon redemption are two factors which contribute more to the
segment formation.
4.2.4. Skeptical Newcomers
The proportion of Skeptical Newcomers Segment is 21.8 percentage within entire sample.
Consumers in this segment are predominantly new members (MD=-0.20) who generally visit the
online store from time to time (RV=-0.11). Shopping rate of Skeptical newcomers segment is a
little below the entire sample average (SR=-0.11). However, when compared with Transient
Customers and Need-based Shoppers segments, Skeptical Newcomers tend to shop online more
often. A notable characteristics of consumers in this segment is that when they do online shopping
they generally purchase very diverse types of products (PD=0.55) which are most of the time low
price items (PP=-0.90). Furthermore, refund rate of Skeptical newcomers is very low (RR=-0.37)
which points to their good decision-making skills in choosing online products. Another notable
characteristic of Skeptical newcomers is their very low discount coupon usage rate (CR=-0.86).
Being relatively new member, shopping less often and purchasing especially very low price items
are the main characteristics of consumers in this segment that lead to call this segment as Skeptical
newcomers. Survey study showed that in online shopping context, perceived product and
perceived financial risks were among the major concerns of consumers (Huseynov and Yıldırım
2016). Majority of participants in this survey worried that difficulty of examining quality of online
products, possibility of receiving malfunctioning products and risk of not getting purchased
product can cause them to incur financial losses. Probably, due to perceived product and perceived
financial risks, Skeptical Newcomers tend to purchase low price items.
The results of variable importance chart of Skeptical Newcomers (Fig. 5) coincide with the results
of factor centroids (Table 3) by showing that all behavioral factors significantly contribute to the
formation of this segment. Price payment, coupon redemption and product diversity are three
factors that contribute more to the segment formation. According to the chart, all factors, except
product diversity, generally take scores which are lower than average within this segment.
Fig. 5 – Variable importance chart (Skeptical Newcomers)
4.2.5. Repetitive Purchasers
This segment is the second largest segment in size with a percentage of 23.9. Consumers in this
segment are predominantly old members (MD=0.11) who generally visit the online store more
frequently (RV=0.30) than all other segments. The most prominent behavioral characteristics of
Repetitive purchasers is their significantly higher online shopping rate than average (SR=1.06).
Among five determined consumer segments, this segment is the one with the highest online
shopping rate. Study of Forsythe et al. (2006) showed that frequency of online store visits and
frequency of online purchases are statistically significantly and positively correlated with
perceived benefits of online shopping. Perceived benefits can be shopping convenience, broader
product selection, ease of shopping and enjoyment. Frequent online store visits and high online
shopping rate indicate that consumers in this segment perceive the online stores to be convenient
and useful in terms of shopping. When compared with other segments, excluding Skeptical
newcomers, Repetitive purchasers tend to purchase products whose prices are relatively lower.
This can be understood from price payment factor score which is slightly less than average (PP=-
0.16). Repetitive purchasers generally visit the online store to meet their specific types of product
needs which can be inferred from very low score of this segment on product diversity factor (PD=0.74). That is, when they shop online, they do not purchase very diverse types of products. When
compared with Opportunist and Transient customer segments, coupon redemption rate of
Repetitive purchasers is considerably lower (CR=-0.32). In other words, consumers in this segment
do not redeem much coupons in their online shopping transactions. Furthermore, Repetitive
purchasers tend to apply for the refund of purchased products slightly more than entire sample
average (RR=-0.19).
Fig. 6 - Variable importance chart (Repetitive Purchasers)
The variable importance chart of Repetitive purchasers (Fig. 6) shows that all behavioral factors
significantly contribute to the formation of this segment and these finding overlap with the results
of factor centroids of this segment given in Table 3. While price payment and refund rate generally
take scores which are higher than average within this segment, the remaining factors
predominantly take lower factor scores than average. According to the chart, price payment and
product diversity are the factors that contribute more to the segment formation.
6. Discussion and Conclusion
In this study behavioral segmentation analysis was carried out by utilizing real e-commerce
transaction records of randomly selected 10.000 online customers. The dataset used in this study
was provided by Markafoni.com which is a very popular online shopping platform in Turkey.
From the supplied dataset, numerous behavioral factors were extracted and used in determining
groups of customers with similar characteristics. The Two-Step cluster analysis which is one of
the clustering tools of IBM SPSS statistics software was used in segmentation process.
Segmentation analysis results showed that a broad online consumer audience is not composed of
single type of consumers. Rather, online consumer audience is a collection of different consumer
groups whose members have different underlying behavioral characteristics in terms of online
shopping. By utilizing the findings of this study, online retailers can develop more effective
marketing strategies for each given segment in order to increase their online shopping rate.
As a result of behavioral segmentation analysis, five different online consumer segments were
determined. These segments are Opportunist customers, Transient customers, Need-based
shoppers, Skeptical newcomers and Repetitive purchasers. Each determined consumer segment
was found to have unique characteristics that differentiated it from other segments.
Opportunist customers regularly visit and shop at online store. They tend to take advantage of
discount coupons and free shipping offers during their online purchases. In order to increase the
online shopping frequency of Opportunist customers, free shipping offers and discount coupons
can be very effective incentives. Product refund rate of Opportunist customers was also found to
be high which indicates their poor decision-making characteristic in online product selection.
Integration of intelligent shopping tools to online stores can help these customers improve the
decision-making process as such tools were found to significantly improve consumers’ decision
quality and product selection process while shopping online (Haubl and Murray 2006; Huseynov
et al. 2016).
Transient customers segment is the largest segment in size which makes it worthwhile to go after.
Transient customers were found to be coupon-prone customers who redeem a relatively large
number of discount coupons. Attractive promotions and discount coupons can play an important
role in increasing the online shopping frequency of these customers. As they are driven by
attractive incentives, they can be easily lost to competitors who offer similar or better offers.
Therefore, for this segment, it is important to identify customers who can be persuaded by
promotions and coupons to make incremental purchases and become frequent visitors of online
store.
Need-based shoppers purchase very diverse types of online products. This type of customers are
driven by a specific product need. They visit the online store with a specific product in their minds
and look whether they can have that need filled in a better manner. Discount coupons redemption
rate of Need-based shoppers was found to be significantly low which implies that coupon-based
incentives might not be effective in making them repetitive purchasers. Need-based shoppers seem
to be influenced by their perceptions about benefits of online shopping. Therefore, ease and
comfort of shopping, a broad product selection, competitive prices and timely shipping can be
good strategies to prevent this type of customers from switching to competitors.
Skeptical newcomers segment is composed of consumers who are relatively new to the online
store. This type of consumers purchase very diverse type of products whose price are very low.
Similar to Need-based shoppers, coupon redemption rate of Skeptical newcomers is also very low.
Therefore, discount coupon rewards may not have any significant effects on their online spending
rate. Study of Forsythe et al. (2006) showed that product related risks and risk of incurring financial
losses negatively and statistically significantly correlated with online purchasing frequency and
amount of money spent online. Product and financial risks refer to consumers concerns such as
inability in examining the online product, choosing the wrong size of clothes, distrust to online
seller, being overcharged and even not getting the ordered product. In order to increase shopping
rate of these customers, it is important that online sellers reduce this type of consumers’ risk
perceptions related to online shopping.
Repetitive purchasers are predominantly old members who visit the online store on frequent basis
and purchase online products significantly higher than all other segments. This type of consumers
is highly loyal to the online seller. Loyalty in online e-commerce context refers to customer’s
favorable attitude toward the online retailer that leads to repeat buying behavior (Anderson and
Srinivasan 2003). Loyal consumers are the main source of profits and they also drives new
customers to the company by disseminating positive word-of-mouth messages about the company.
Therefore, it is important that the company maintains good relationship with loyal customers as
they are very important for the company’s long term success. Researchers found that high degree
of information quality, system quality, service quality, customization, interactivity and functional
look-and-feel in online stores are associated with higher customer satisfaction and higher customer
satisfaction in turn is associated with higher customer loyalty (Chang and Chen 2008; Chen et al.
2015). That is, if online sellers can improve their online stores’ functional and visual features along
with service and information quality, they can increase the loyalty level of not only repetitive
purchasers but also all other determined consumer segments.
Beside the differences, all determined consumer segments have similarities in different aspects.
In all segments, the sizes of female consumers are significantly higher than male consumers.
Furthermore, for all segments, the subscription rates to SMS notifications of the company are
higher than e-mail subscription rates. Moreover, for one-click payment option, the storage rates of
credit card details to the online system are very low in all consumer segments. These findings show
that consumers in all segments have concerns about the security of their financial details in online
shopping systems. These findings coincide with the findings of a survey study which show that in
online shopping platforms, the majority of consumers fear that their credit-card details may not be
secure and may be misused (Udo 2011; Huseynov and Yıldırım 2016). Online retailers are
recommended to take necessary security and privacy measures in their online stores to protect
personal and financial details of customers and they also advised to inform customers about how
their security and privacy are being protected in their online stores. Furthermore, online retailers
are advised to put necessary security and privacy seals in their online stores, as these seals were
found to increase consumers’ trust levels in online retailers (Huseynov and Yıldırım 2016).
7. Future Research Suggestions
This study utilized e-commerce log data of a single type of B2C e-commerce platform. Future
studies are recommended to utilize e-commerce data of different types of e-commerce platforms
such as market creators. Furthermore, future studies are recommended to carry out behavioral
segmentation analysis by using different behavioral factors from the ones used in this study. By
this way, how segment types and their characteristics differ according to different e-commerce
platforms and different behavioral factors will be understood.
8. References
Aljukhadar M and Senecal S (2011) Segmenting the online consumer market. Marketing
Intelligence & Planning 29(4): 421–435
Anderson RE and Srinivasan SS (2003) E-satisfaction and e-loyalty: A contingency framework.
Psychology & Marketing 20(2): 123–128
Brengman M, Geuens M, Weijters B, Smith SM and Swinyard WR (2005) Segmenting Internet
shoppers based on their Web-usage-related lifestyle: A cross-cultural validation. Journal of
Business Research 58: 79– 88
Chang HH and Chen SW (2008) The impact of customer interface quality, satisfaction and
switching costs on e-loyalty: Internet experience as a moderator. Computers in Human
Behavior 24: 2927-2944
Chen Y, Hsu I and Lin C (2010) Website attributes that increase consumer purchase intention: A
conjoint analysis. Journal of Business Research 63: 1007–1014
Chen J, Yen D, Pornpriphet W and Widjaja A (2015) E-commerce web site loyalty: A cross
cultural comparison. Information Systems Frontiers 17(6): 1283-1299
Deloitte (2015) E-commerce in Turkey: Market Size in 2014. Available at:
http://www2.deloitte.com/content/dam/Deloitte/tr/Documents/technology-mediatelecommunications/tubisad-infografik-turkiyede-e-ticaret-2014-pazar-buyuklu%C4%9Fufinal.pdf
eMarketer (2014) Global B2C Ecommerce Sales to Hit $1.5 Trillion This Year Driven by Growth
in Emerging Markets. Available at: http://www.emarketer.com/Article/Global-B2CEcommerce-Sales-Hit-15-Trillion-This-Year-Driven-by-Growth-EmergingMarkets/1010575
Eurostat (2012) Internet Use and Skills. Available at: https://ec.europa.eu/digital-singlemarket/sites/digital-agenda/files/DAE%20SCOREBOARD%202013%20-%203INTERNET%20USE%20AND%20SKILLS.pdf
Forsythe S, Liu C, Shannon D and Gardner LC (2006) Development of a Scale to Measure The
Perceived Benefits and Risks Of Online Shopping. Journal of Interactive Marketing 20(2):
55-75
Ganesh J, Reynolds KE, Luckett M and Pomirleanu N (2010) Online shopper motivations, and estore attributes: An examination of online patronage behavior and shopper typologies. Journal
of Retailing 86(1): 106–115
Gehrt KC, Rajan MN, Shainesh G, Czerwinski D and O’Brien M (2012) Emergence of online
shopping in India: Shopping orientation segments. International Journal of Retail &
Distribution Management 40(10): 742–758
Haubl G and Murray KB (2006) Double agents: Assessing the role of electronic productrecommendation systems. MIT Sloan Management Review 47(3): 8–12
Hill WW, Beatty SE and Walsh G (2013) A segmentation of adolescent online users and shoppers.
Journal of Services Marketing 27(5): 347–360
Huseynov F and Yıldırım SO (2015). Behavioral Issues in B2C E-commerce: The-state-of-the-art.
Information Development. DOI: 0.1177/0266666915599586
Huseynov F and Yıldırım SO (2016) Internet users’ attitudes toward business-to-consumer online
shopping: A survey. Information Development 32(3): 452–465
Huseynov F, Huseynov SY and Özkan S (2016) The influence of knowledge-based e-commerce
product recommender agents on online consumer decision-making. Information Development
32: 81-90
Keller KL and Kotler P (2012) Marketing management (14th edition). New Jersey: Pearson
education
Liu Y, Li H, Peng G, Lv B and Zhang C (2015) Online purchaser segmentation and promotion
strategy selection: Evidence from Chinese E-commerce market. Annals of Operations
Research 233: 263-279
Pew Research Center (2011) Search and email still top the list of most popular online activities.
Available at: http://pewInternet.org/Reports/2011/Search-and-email.aspx (accessed 27 May
2016)
Pew Research Center (2014) Online Activities in Emerging and Developing Nations. Available at:
http://www.pewglobal.org/2015/03/19/2-online-activities-in-emerging-and-developingnations/ (accessed 27 May 2016)
Pradas SI, Miguel FP, Garcia AH and Pelaez JC (2013) Barriers and drivers for non-shoppers in
B2C e-commerce: A latent class exploratory analysis. Computers in Human Behavior 29:
314–322
Rezaei S (2015) Segmenting consumer decision making styles (CDMS) toward marketing
practice: A partial least squares (PLS) path modeling approach. Journal of Retailing and
Consumer Services 22: 1–15
SPSS Inc. (2001). The SPSS Two Step cluster component. A scalable component to segment your
customers more effectively. White paper – technical report, Chicago.
http://www.spss.ch/upload/1122644952_The%20SPSS%20TwoStep%20Cluster%20Compo
nent.pdf
TSI
(2014) Research on Information Technologies
http://www.tuik.gov.tr/PreHaberBultenleri.do?id=16198
Usage
Rate.
Available
at:
Udo GJ (2011) Privacy and security concerns as major barriers for e-commerce: A survey study.
Information Management and Computer Security 9(4): 165–174
Wang HC, Pallister JG and Foxall GR (2006) Innovativeness and involvement as determinants of
website loyalty: I. A test of the style/involvement model in the context of Internet buying.
Technovation 26: 1357–1365