System Dynamics Methodology to Customer Loyalty
(Case Study: Internet Service Provider Company)
Alireza Bafandeh Zendeh
Department of Management, Tabriz Branch, Islamic Azad University, Tabriz , Iran
Samad Ali
Department of Management, Tabriz Branch, Islamic Azad University, Tabriz, Iran
Farzad Adelzadeh
Department of Business Adminttration, Aras, Institute of Higher Education, Tabriz, Iran
Masoud Askarnia
Department of Information Technology Management, Mizan, Institute of Higher Education, Tabriz, Iran
Abstract
Due to the complexity of the customer loyalty, we tried to provide a conceptual model to explain it in an
Internet service provider company with system dynamics approach. To do so, the customer's loyalty for
statistical population was analyzed according to Sterman’s modeling methodology. First of all, the
reference modes (historical behavior of customer loyalty) were evaluated. Then, dynamics hypotheses
were developed by means of causal-loop cyclic diagrams and stock-and-flow charts, based on theoretical
literature. In the third step, initial conditions of variables, parameters, and mathematical functions
between them were estimated. The model was tested and finally, advertising, quality of service
improvements and maintaining the status quo scenarios were evaluated. Results suggested improving the
quality of service scenario is more effective in comparison to others.
Keywords: System Dynamics, Customer Loyalty, Internet Service, Simulation
1. Introduction
One of the important concepts in marketing is customer loyalty to specific trademarks and
brands. This concept plays an important role in long-term benefits of the company since loyal
customers do not need any promoting actions; they are willing to pay more for obtaining
advantages and qualities of their favorite brand. Furthermore, since in online markets, loyalty of
customers to the brand can guarantee the life of the company or its destruction, customer loyalty
has a competitive aspect as well. Customer loyalty plays a significant role when businesses are
done in a competitive market. In these markets, customers have the right to choose and their
second purchase won’t be the result of lack of other choices. High speed Internet services work
in such competitive markets that customer loyalty modeling turns to be a remarkable element for
them. Like most disciplines in human sciences, customer loyalty is complex and multi-
dimensional ,too. Some of the reasons of complexity in these fields are reciprocal interactions
between different elements such as advertisement which is performed in different approaches by
spending money, the oral advertisement performed by customers, satisfaction or dissatisfaction
of customers with the quality of the product and service before and after sale, customers who
depart from the system that refused to extend their subscription, customers attracted to the
system, etc. Furthermore, these interactions are done dynamically, not statically. Over time, these
elements interact influencing each other and being influenced by other elements. Nearly all
researches done on customer loyalty and satisfaction, focus only on some of the related elements
on loyalty. Also, these researches are done statically and in the absence of time element; being
measured in mono-dimensional linear static measurement, these results are unable to predict the
results correctly. Reciprocal and dynamic interactions need appropriate tools. System dynamic
approach gives us the opportunity to model these complex problems more realistically, as a
result with respect to the mentioned limitations in researches on customer loyalty on the one
hand, and the ability to overcome these limitations by the use of system dynamic approach on the
other. In this article, we attempt to model the customer loyalty with this methodology.
2. Review of Literature
Between 1920 and 1950, a sales-based approach became dominant in the market. This approach
emphasizes the fact that a customer may not be willing to buy something and things could get so
hard that the organization may be forced to take special action in order to sell its products and
services. From the viewpoint of this approach, there are two important focuses: focus on product,
and focus on customer, and the conditions are in favor of customer (Harvey, 2002). Sales-based
approach was recognized for the first time by production and industrial organizations in the early
1980s. This approach emphasizes the fact that the more an organization is aware of its customers
and their needs, the less problems it will encounter in its sales processes (Jay, 2001). Since 1980,
academic researches and activities have made it clear that those institutions and organizations
which have been successful in terms of keeping their loyal customers satisfied, have made
considerable profits. In this regard, in the early 1990s, there was an exponential increase in the
use of customer loyalty programs for different subject areas (Margarita, 2001). In the final years
of the twentieth century, with the increase of competition and emergence of higher technologies
especially in the realm of communications, organizations realized that the only way to survive in
the market is not only to know but also be in touch with their customers (Johnson et al., 2001).
Considering its multi-dimensional and complicated nature, there are various acceptable
definitions for customer loyalty (Soderlund, 2006). Therefore, it is not easy to agree on a general
and comprehensive definition of loyalty. Jacooby and Keener define customer loyalty as brand
prejudice in which a person prefers a certain brand over others and decides on it based on a
psychological commitment (McMullan & Gilmore, 2008). In another place, loyalty is defined as
keeping a deep commitment to repurchasing the same product or service again and again (Chen
et al., 2002). Although repurchasing can be a sign of customer loyalty, Day & Jacooby’s research
reveals that sheer repurchasing of a product does not equal to loyalty. In some cases, people are
forced to purchase and repurchase a product from the same source because there are no other
options available. This kind of loyalty is defined as false loyalty. Loyalty can be real only when
customers are attracted to the same product or company with a higher priority and based on
having a clear understanding of a product’s difference with other alternatives. In response to
such criticisms, researchers have suggested that analysis and measurement of loyalty should be
done not only based on behavioral but also attitudinal aspects. Figure 1 illustrates a customer
loyalty model in which both behavioral and attitudinal aspects have been taken into
consideration.
Customer loyalty Customer loyalty
(behavioral aspect) (attitudinal aspect)
oF
Purchase ] >| etait ] aI Customer trust ] py Custonier ] DI Customer Loyalty
Figure 1: A dynamic model of customer loyalty in both behavioral and attitudinal aspects (Donio, 2006)
Generally, three different approaches have been offered for measuring customer loyalty:
behavioral approach, attitudinal approach, and the mixed approach. In behavioral approach,
number of purchases is emphasized as an indicator of loyalty. According to Chen and Bowen
(2001), in the attitudinal approach, the use of attitudinal information reflects a customer’s
feelings and psychological dependencies. In the third approach, according to Lovelock et al.
(1998), loyalty is measured based on customer presenter’s performance, brand prejudice,
frequency of purchase, the amount of purchase, and the most recent purchase. In the mixed
approach, in addition to a customer’s behavior and attitude, the effect of attitude on behavior is
also emphasized. Figure 2 illustrates this relationship.
Getting involved: dynamic = Participation in the hope an Poco)
opportunity
Figure 2: The effect of attitudinal loyalty on behavioral loyalty (Shang, 2006)
Griffin believes with an initial purchase, a customer should go through 5 stages (Griffin, 2002).
These five stages are: 1. becomes aware of the product; 2. makes an initial investment; 3. post-
purchase evaluation; 4. decision to repurchase; and 5. Repurchase. If the customer has the
freedom to choose and access to other alternatives, repurchasing the same product will be an
indicator of a true customer loyalty. Otherwise, repurchase is an indicator of false loyalty and not
repurchasing is product is a sign that customer is not loyal. Robinson and Baldinger (2000)
showed that a customer’s positive perception has a good influence on customer loyalty. In
addition, the positive effect of high quality service on customer loyalty was proved in this
research. Momeni (1389) showed that a perceived value can have a direct effect on customer
loyalty and satisfaction. Wolfinger and Gili (2004) studied the factors which influence customer
loyalty in bigger companies. In their research, the role of service quality and customer
satisfaction on loyalty was proved. Lusiano (2006) studied customer loyalty in the Internet
services. In his research, an appropriate customer support system and Internet speed were very
important factors and influenced customer loyalty.
Among other variables that may influence customer loyalty is advertising. Although there are
various ways to introduce and advertise products, it seems that mouth-to-mouth advertising can
have a lot of positive influence on customers’ consuming-related behaviors. Mousavi’s research
(1389) proved the effect of advertising on customer loyalty. Mirzaei-far (1390) showed that
recommendation given by others and customer satisfaction directly influences repurchase. One
of the important stages in producing and introducing a product is the time that it takes for the
product information to be spread among people and social networks. The key to this way of
introduction is mouth-to-mouth communication between the members of a social network.
Several researches have been conducted on the importance of mouth-to-mouth advertising to buy
products by Chevalier and Mayzlin (2006) and Mobius et al. (2006). In addition, Bughin et al.
(2010) showed that 20 to 50 percent of initial purchases were the result of moth-to-mouth
advertising and recommendation by others. Emotional commitment, related to the concept of
loyalty we are discussing here, is a positive emotional obligation which reflects a psychological
dependency on the other party (i.e. the provider of a product) (Sweeney 2008). The results of
Yen’s research show that most customers before repurchasing an item, search for retailers that
are selling the same product for a lower price, and they can easily change the website from which
they want to purchase (Yen, 2010). In the model they developed for explaining long-term
customer loyalty in electronic (Internet) purchasing, Mohd Kassim, N., & Ismail have mentioned
service quality, satisfaction and trust as factors influencing customer loyalty (Mohd Kassim,
2009). Also, Ramanathan’s research offers a model based on pre-purchase factors such as access
to services, e-pay details, comparing prices and post-purchase factors such as punctual delivery,
the way customer complaints are dealt with, and availability of customer support (Ramanathan,
2011). Park et al. conducted a research entitled “Social Perspectives of E-Care Centers and the
Creation of Customer Loyalty” in which the final model, the effect of service quality on social
values, and satisfaction with, and commitment to, e-care center are proved (Park, 2011). The
FRO model (Fast-Response Organizations) recognizes the six factors of price, quality, service,
time, trust, and flexibility as the factors leading to satisfaction and customer loyalty (Starr, 1990).
Figure 3 illustrates this model. Based on the theoretical principles of stable loyalty in e-retails,
and making use of models offered by Srinivasan et al. and Chang and Chen (Chang, 2009), the
proposed model for customer loyalty is elaborated in Figure 4.
eh.
relay
| |
A
services
Figure 3: Fast-response organizations and customer loyalty
customers esti
satisfaction Base ppRIng
regestered
shopping
OMPETITIVENESS
customers’
constant loyalty
interact
shopping
credit of
websites’ trade
brand
secure paym
Figure 4: Conceptual model of customer loyalty (Srinivasan, 2002)
Figure 5 shows another model of customer loyalty. This mode takes into account various factors
such as competence, commitment, ability to communicate and resolve issues and also mediating
factors such as trust and quality of communication and considers them as elements related to
customer loyalty (Ndubisi, 2007). In addition, Figure 6 shows a model without any mediator and
takes into account four independent variables of trust, commitment, communications, conflict
management and the only dependent variable of loyalty (Ndubisi, 2007).
ability of making
relations
Gualityof commitment
relationship
Figure 5: Customer loyalty model with mediation of Bank of Malaysia
loyalty K
—
communication
conflict
management
commitment loyalty
trust
Figure 6: A model without mediation
One of the important indicators of loyalty is American Customer Satisfaction Index (ACSI). This
model in addition to offering average and dispersion values for the variable of customer
satisfaction, evaluates the effect of variables on one another. ACSI includes hidden variables and
is calculated based on some measurable criteria and also customer surveys. This model’s
credibility is high because of calculating the cause-effect relationship between variables. In this
model, customer satisfaction is one of variables which are calculated using certain measurable
criteria. This approach has multiple criteria and these include overall satisfaction, customer’s
treatment of a product or service’s quality compared to his expectations, and product or service’s
quality compared to the ideal that customers have in mind. Figure 7 illustrates this model.
Europe’s Customer Satisfaction Index which includes expectations, quality and appreciated
value, satisfaction and loyalty, is one of other appropriate models that can be used to evaluate
customer loyalty. Figure 8 illustrates this model.
Customer
expectations
Gained
quality
Customer
complaint
Customer
satisfaction
Gained quality
Figure 7: ACSI (Donio 29, 2006)
Achieved
value
Satisfaction oe
perceived quality of
hardware
perceived
quality of
software
Like most topics related to humanities, customer loyalty is complicated and multi-dimensional.
In this research, modeling customer loyalty using system dynamics methodology is discussed
and is one of the important and modem non-linear theories based on systemic thinking in a
complicated world (Sterman, 2002). In system dynamics, implementation of system behavior,
analysis of results, testing of the simulated model, and information on the way system behaves
based on the considered conditions are offered to the system analyst (Hasret, 2007). Modeling
based on system dynamics, as a part of learning process, is repetitive and it is also a constant
process of formulating hypotheses, testing and controlling structured and mental models
(Kahzadi, 1382). In this research, according to a methodology offered by Sterman, five stages
have been considered for modeling. In the first stage, the problem was identified and defined as
follows: with regard to the data available in Hamara System Co.’s database, the company’s
activity during the past was studied (time horizon analysis). Then, in regard to the past time
horizon (history) analysis, problem and research literature, the key variables were identified. In
the second stage, the dynamic hypothesis was planned. In this stage, based on research literature
and a comprehensive analysis of previous research, and also based on expert views, a theory is
formed to explain the problem. This theory is described in the form of causal-cyclic and stock
and flow charts. Then, in the formulation stage, with mathematical estimations between
Figure 8: Europe’s Customer Satisfaction Index
variables, dynamic hypothesis was changed into a simulated model. The simulated model is
evaluated in stage four, and finally, system behavior is studied according to various policies
and scenarios and the best scenario is suggested for improving customer loyalty.
The first two steps in the methodology proposed by Sterman are qualitative and the following
two stages are quantitative. Therefore, the methodology of this research is the mixed type. The
statistical population being studied is Hamara System Co.’s customers in the years between 1389
and 1391. The variables connected with customer loyalty and the relations between them were
identified using library-based methods and based on the results of previous researches. The
information archive of Hamara System Co. was used in order to measure the relations between
variables. In addition, in some cases where documented information was not available, field
methods and interviews were used to estimate the mathematical relations between variables.
Finally, the model was implemented by VENSIM (software) and the information is (are)
analyzed.
3. Research findings
Considering the methodology used for this research, research findings are classified and offered
based on the five stages proposed by Sterman.
3.1.Identification and definition of the problem
The first step to be discussed in terms of framework is the time horizon. If we want to go back in
terms of time, the historical behavior of the number of high-speed Internet users in this company
can be explained using a chart like Chart 1.
40000 » 35000.
35000
30000 -
|
|
25000
|
|
|
33000
29500 —*— Number of
customers
20000 -
15000 -
10000 -
5000 -
Number of customers
2 2 2 oe eS eS ES > Time
o e 3
& Ca Roa Ra Ra # Ra Rd
Figure 9: number of customers in time horizon
The chart above shows the curve by which the number of customers has grown. As it can be
seen, in the beginning years because of lack of sufficient capacity and lack of competitors,
number of users is relatively high. As the capacity grows, the number of users grows as well but
from the first half of 1390 on, with the growth of awareness and number of competitors, the
number of customers descended to 29500. The company’s capacity is 52,000 users. The gap
between capacity and the number of users indicates a problem in the company’s relationship with
its users. The variable behavior in the number of customers shows that the organization is in the
regression state. The reason for this could be connected with various variables such as customer
satisfaction, system performance, the perceived values of customers, and services offered by
competitors.
3.2.Planning a dynamic hypothesis
In planning a dynamic hypothesis, two approaches can be used to identify the variables and the
relations between them. Literature review is a valuable resource that can be used to identify these
variables. In addition, experts, customers, and suppliers can be potential sources of information
to help identify the variables. Considering the literature review, the variables inside the system
that can influence the target variable (i.e. customer loyalty) are as follows: 1. the main functions
of the system; 2. the values perceived by the customer; 3. mouth-to-mouth advertising; and 4.
advertising.
The main functions of the system which are the main reason why customers interact with the
company, are connected to the company’s human resources and physical facilities. What is
meant by the value received is the customers’ idea of a comparison between the paid money
and the received service. The value received by customers, in addition to the service offered by
the organization, is a result of comparing this service with the price and service delivered by
competing companies. Customers transfer their positive and negative experiences regarding the
company to other people. Of course the intensity of this type of communication will be different
for positive and negative feedbacks. According to theories, in order to measure these
communications, this intensity has to be evaluated. Due to the above, negative experiences
will have more effect and intensity. This is known as mouth-to-mouth advertising. In order to
reduce the Company's vacant capacity, advertising plans and promotions are very important.
Customer satisfaction is a variable which is affected by system performance and the value
received by the customers. This variable has a great effect on mouth-to-mouth advertising done
by customers. Measuring loyalty is not an easy job due to its qualitative and mental nature and
also problems such as false and hidden loyalty that were discussed in the theoretical literature.
To solve this problem, loyalty is measured based on continuous extension of the contract with
the company. In this research, customers who have extended their subscription more than three
times were considered as loyal customers. Customers renew their subscription once every three
months. Customers initially enter the company through advertising or other customers and then,
renew their subscription according to their degree of satisfaction with the service. The dynamic
theory was provided based on an evaluation of literature review and interviews with managers of
Hamara System Co., and also by studying the company’s database, making use of causal-cyclic
charts, and stock and flow chart. Figure 10 illustrates the dynamic theory in the form of a stock
and flow chart. As it is seen in this chart, customers purchase products and services according
to advertising or recommendations by existing customers. After three months, based on his
satisfaction with the system, the customer decides to renew or opt out of his subscription. This
level of system is known as the customers at first level of loyalty. As time passes, customers will
evaluate their satisfaction with the system and after their subscription is over, will decide to
either renew or terminate their subscription. This is known as customers at second level of
loyalty. Based on the circles created, customers are divided into three levels of zero, one, and
two, and customers with a higher level of loyalty are at level two.
Figure 10: dynamic hypothesis of customer loyalty
3.3.Formulating and simulating the model
In this step, the formulas and the relations between them were gathered through documents and
interviews listed in Table 1.
Table 1: Estimating functions
er of customers = accepi
Satisfaction level = (quality) * 0.574 + (perceived value) * 0.424 + 0.25
Entrance rate = advertising + mouth-to-mouth advertising
Performance quality = Random uniform (0/7, 0/9, 99)
Perceived value = Random uniform ( 0/8,0/92,99)
Function for level of customer satisfaction at level zero = (satisfaction level) * 0.868) + 1/3
Customers leaving at level zero = Delay fixed (entrance rate, 3, 0)
Customer departure rate at level zero= exp(0) * (1 - gh(0))
Subscription renewal rate of level-one customers = subscription renewal of level-zero customers —
subscription renewal rate of level-two customers — leaving rate of level-one customers
Functions for satisfaction level and mouth-to-mouth advertising = (satisfaction level) * 1.02) +0.13
Acceptance rate = Random uniform (0/65, 0/99, 99)
Total number of potential customers = 10,000
Acceptance through mouth-to-mouth marketing = acceptance level * satisfaction rate and mouth-to-
mouth marketing * average of new customers
If then else (total possible customers/number of customers < 1, 1, 1/ total possible customers/number of
customers)
Advertising budget =
lookup ( [(0,0)-(6000,1000)],(0,1000),(3500,1000),(4000,550),(4500,500),(5000,400))
Advertising effectiveness = look up (([(0,0)-(4000,1000)],(1500,350),(1800,500),(2000,550),(3000,1000))
Total number of customers = level-zero customers + level-one customers + level-two customers
Potential attracted customers = Random uniform (6000, 8000, 99)
Function for degree of customer satisfaction at level one = ((degree of customer satisfaction) * 0.906)
Customers at level one = Delay fixed (entrance rate, 3, 0)
Customer departure rate at level one = exp(1) * (1- gh (1))
Level-one customers = renewal rate of level-one customers — renewal rate of level-two customers —
departure rate of level-one customers
Function for customer satisfaction degree at level two = ((degree of customer satisfaction) * 0.641) +1
Customers at level two = Delay fixed (entrance rate 2, 3, 0)
Customer departure rate at level two = exp(2) * (1- gh (2))
Level-two customers = renewal rate of level-two customers — departure rate of level-two customers
Subscription renewal rate of level-two customers = exp(2) * gh(2)
Subscription renewal rate of level-one customers = exp(1) * gh(1)
Subscription renewal rate of level-zero customers = exp(0) * gh(0)
Then, having equations and the necessary information, the suggested model was simulated in
VENSIM software.
2 4 6 8 10 12 14 16 18 20 22 24 26 28 3
Figure 11: number of customers entering the company
40,000
30,000,
20.000,
10,000 | _//
o 2 4 6 8 10 12 1) 16 18 2022 24 26 28 30
Figure 12: total number of customers
As it can be seen, up to month 14 in the time horizon, number of customers is increasing, but it
is decreasing after month 18. We should look for the reason among influential variables.
Satisfaction rate is shown in the following chart.
20.000
15,000 aa = =
V =] =
10,000
5,000 Xx -
°
0 2 4 6 R 10 12 14 14 18 20 22 24 246 28 20
Figure 13: number of level-three customers
3.4.Testing the model
Before testing different scenarios, we should be sure of the model’s validity. One of the
appropriate ways of validating a model is to compare the simulated behavior with the real
behavior. Figure 14 shows an example of these two behaviors. As it can be seen in the chart, the
simulated and real behaviors are almost the same.
g —
£ 30.000 -
8
3 20.000 | —e—|/ Time horizon
S20
8 10.000 | = Stimulated
z
2 0 Time periods (month)
Figure 14: comparing the number of customers in the time horizon (real behavior) with the simulated
behavior
3.5.Evaluating the scenarios
3.6.
At the end, considering the validity of the test, the scenarios are implemented on the model.
First scenario: “maintaining the status quo”
20,000
15,000
10,000
5,000
0
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Figure 15: number of level-three loyal customers in first scenario
By maintaining the status quo, the number of new customers may increase or decrease and
also, the number of departing customers at various levels will change due to low satisfaction rate,
the growing power of competing companies, availability of incentive plans, and loyalty.
Second scenario: “growth in the number of incentive plans”
20,000
15,000
10,000
5,000
0
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Figure 16: number of level-three loyal customers in second scenario
As this variable increases, it is possible that advertising and values perceived by customers about
the existing performances of the system in comparison to competitors, will increase the number
of customers and satisfaction rate as well. Both of these variables will directly influence the
arrival of new customers and keeping the existing ones in the system. The number of customers
will raise in the long run.
40,000
30,000
20,000
10,000
0
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Figure 17: total number of customers in second scenario
Third scenario: “hiring more staff and increasing the quality within the system”
20,000
15,000 —
10,000
5,000
°
Figure 18: number of level-three loyal customers in third scenario
In the variable aspects of system performance, one of the influential factors is the increase in
working quality of staff in the system. Customers will be loyal to the system if they perceive this
value. This variable will also affect satisfaction rate even more (it will increase the satisfaction
rate) and satisfaction rate, in its own turn, will influence loyalty at various levels. As it can be
seen, in this scenario the system will have an ascending trend in terms of total number of
customers. But it is worth mentioning that the growth rate of customers has decreased and this
can be a result of increase in the number of competitors, decrease in the company’s share of the
market, and market saturation.
40,000
30,000 S-
20,000
10,000
7)
°
o 2 4 6 8 410 42 34 16 18 20 22 24 26 28 30
Figure 19: total number of customers in third scenario
4, Summary and conclusion
In this paper, using a dynamic approach, the factors affecting customers were studied in Hamara
System Co. The analysis of the data revealed that all the aspects of a model influence customer
loyalty and the impact intensity of these factors are different from one another. All in all, the
results of the study show that customer satisfaction is influenced by two variables namely the
value perceived by the customer and service quality. With an increase or decrease in any of these
variables, satisfaction rate will change accordingly and affect the organization in two ways: 1. as
satisfaction rate drops, mouth-to-mouth advertising rate will drop as well and less people will
enter the system; or 2. as satisfaction rate increases or decreases, departure rate will change as
well. Of course, satisfaction rate depends on subscription timespan or departure time, but people
with a high level of loyalty and a low level of satisfaction do not depart easily.
Based on analyses of these scenarios, and considering the fact that service quality is one of the
important factors in creating customer loyalty, paying attention to service quality in organization
plans and approaches are recommended. Informing others about the service quality in marketing
and advertising activities could be helpful. In order to increase the rate of customer loyalty and
decrease the rate of customer departure, it is necessary to pay attention to the value perceived
indicator as well. Therefore, it is recommended that an appropriate organizational culture should
be promoted so that tolerance and customer-orientation approach would be within the
organization. The next important factor is time. An organization should appreciate people’s time
and arrange its plans in order to consume a fair amount of time of customers for their desired
services. The results of validity test show that delivering promises and earning customers’ trust is
very influential on satisfaction and the changing customers into zealous supporters and lifelong
loyal members.
References
Kahzadi, N.( 2003). Electronic Business. First Electronic Business congress. Tehran.
Chang, H.H. & Chen, S.W., (2009). “Consumer perception of interface quality, security, and
loyalty in electronic commerce”, Information & Management, 46, 411-417.
Chen, L., Gillenson, M. L., Sherrell, D. L. (2002). Enticing online consumers: an extended
technology acceptance perspective, Information & Management, NO. 39, 705-719.
Donio, J, Massari, P & Giuseppina, P (2006),'Customer satisfaction and loyalty in a digital
environment: an empirical test’, Journal of Consumer Marketing, vol. 23, no. 7, pp.445-57.
Griffin, J., (2002). Customer Loyalty: How to Earn It, How to Keep It, New and Revised
Edition, Weekly publisher
Hasret NUHOGLU, (2007). System Dynamics A pproach In Science and Technology Education,
Journal of TURKISH SCIENCE EDUCATION Volume 4, Issue 2, September 2007.
Harvey, D. (2002), Customers-The Hidden Threat to Your Business, Capstone, Oxford.
Jay, R. (2001) "winning minds-the ultimate book of business leadership" capstone.
Johnson, M Gustufsson. A, Andreassen, T, Lervik, L and chaJ (2001) "The evolution and future
of national customer satisfaction index models", Journal ofeconomic psychology, V ol.22, pp.
217-245
Margarita, Elorz. (2007). The use of loyalty cards databases: Differnces in regular price and
discount sensitivity in the brand choice decision between card and non-card holders, Journal of
Retailing and Consumer Services, no. 15, 52-62.
McMullan, R., Gilmore, A. (2008). Customer loyalty: an empirical study, European J ournal of
Marketing. Vol. 42 Iss: 9/10, pp.1084 — 1094
Mohd Kassim, N., & Ismail, S., (2009). “Investigating the complex drivers of loyalty in e-
commerce settings”, Measuring Business Excellence, Vol. 13 No. 1, pp. 56-71.
Ndubisi , nelson Oly (2007) « Relationship quality antecedents : the Malaysian retail banking
perspective » International Journal of quality & Reliability management, Vol.24 , No.8, PP.829-
845.
Ndubisi, O. N. (2007), "Relationship marketing and customer loyalty", Marketing intelligence &
planning, Vol. 25, No. 1, pp: 98-106.
Ramanathan, R., (2011). “An empirical analysis on the influence of risk on relationships between
handling of product returns and customer loyalty in Ecommerce”, International Journal
Production Economics, 130, pp 255-261.
Shang, R.-A.& Chen, Y .-C.& Liao, H.-J., (2006), "The value of participation in virtual consumer
communities on brand loyalty", Internet Research, vol. 16, NO.4, p.398-418
Soderlund, M (2006), ‘Measuring customer loyalty with multiitem scales: a case for caution’,
International J ournal of Service Industry Management, vol. 17, no. 1, pp. 76-98.
Srinivasan, S.S., Anderson, R., & Ponnavolu, K., (2002). “Customer loyalty in e-commerce: an
exploration of its antecedents and consequences”, J ournal of Retailing, 78 (1), pp 41-50.
Starr, Martin k., the role of project management in a fast response organization, Journal of
Engineering & Technology management, Volume 7, Issue 2, September 1990, page 89-110.
Sterman, J. D. (2011). Business dynamics. tehran: samt.
Sweeney, Jill & Joffre Swait (2008) « The effect of brand credibility on customer loyalty »
Journal of Retailing and Consumer Services, 15.
Park. J., Chung, H., & Rutherford, B., (2011). “Social perspectives of e-contact center for loyalty
building”, J ournal of Business Research, 64, pp 34-38.
Yen, Y-S., (2010). “Can perceived risks affect the relationship of switching costs and customer
loyalty in e-commerce? Internet Research ,Vol. 20 No. 2, pp. 210-224.
