Drives of Customer Equity in the US and Dutch Cloud industry and the mediating role of loyalty intentions

(1)

Drives of Customer Equity in the US and Dutch Cloud

industry and the mediating role of Loyalty Intentions

Author:

Xaviër Halfhide

Student number:

10784322

Supervisor:

Tom Paffen

Submission date:

27-1-2017

Final version

MSc. In Business Administration - Marketing Track

Amsterdam Business School - University of Amsterdam

(2)

Statement of originality

This document is written by Student Xaviër Halfhide who declares to take full responsibility forthe contentsof thisdocument.I declare that the text and the work presented inthis document is originaland

that no sources other than those mentioned in the text and its referenceshave been used in creating it.The Faculty of Economics and Business is responsible solely for thesupervision of completion of the

work, not for thecontents.

Abstract

In this research an answer to the following question has been attempted to given: “What is the impact of the Customer Equity sub-drivers on the likelihood of becoming a paying user in the Dutch and American Cloud-based industry and to what extend can the impact be explained by loyalty intentions?”. The results show that brand equity, efficiency and fulfillment are amongst the most important drivers and all have a significant impact on loyalty intentions. Also, loyalty intentions fully mediated the effect of these three variables on the likelihood of becoming a paying customer. No evidence was found that indicates a moderating effect of the country of residence. From a academic perspective, it extends the existing literature and combines two streams of research by focusing on the impact of customer equity drivers in the context of cloud-based business models. From a managerial perspective this research provides valuable insight for firms with a cloud-based business model. By knowing which equity drivers are most important to cloud customers strategic trade-offs can be made in terms of marketing investments, product innovations and service improvements.

(3)

1. Introduction

Customer Equity is a commonly used marketing metric. It contains a set of techniques that helps companies evaluate their portfolio of customers. Rust, Lemon & Zeithaml (2004) define Customer Equity as “the total of the discounted lifetime values summed over all of the firm‟s current and potential customers” (p. 110). It reflects the sum of the total individual customer life time values. Using data, technology and surveys, Customer Equity analyzes allow companies to discover key customers and customer segments that contribute to the long term value of the company. Within the marketing literature the Customer Equity concept is widely recommended to evaluate customer-based assets and optimize marketing investments. (e.g., Venkatesan & Kumar, 2004; Rust, Lemon & Zeithaml, 2000, 2004; Kim, Jung, Suh & Hwang 2006; Reinartz and Kumar, 2000). Managers who have to trade off strategic marketing and innovation initiatives can do so based on a data-driven approach rather than solely intuition and gut feeling. The three drivers of Customer Equity (Value, Brand & Relationship Equity) can be analyzed and linked to financial performance. In this way, managers can estimate the return on investment from improvements in a driver.

1.2. Research gap

Despite the importance and popularity of the Customer Equity construct, there is still little known about the Customer Equity Sub-drivers in the context of an innovative environment where intangible digital products are sold. One relatively new but exciting industry in the digital area is the Cloud sector. This industry is expanding rapidly and the strong economic impact is undeniable. According the International Data Corporation (IDC), the cloud computing market was $56.6B in 2014 and is growing annually to $127.5B in 2018. This represents a five-year compound annual growth rate of 22.8%, which is about six times the rate of growth for the overall IT market (Gens, 2015). Another study by Cisco predicts that by 2019, more than four-fifths (86 percent) of workloads will be processed by cloud data centers and only 14 percent will be processed by traditional data centers. Also, a 2014 research publication of Deloitte concludes that Small businesses using cloud technology to overcome growth challenges grow 26% faster and deliver 21% higher gross profits (Deloitte, 2014).

(6)

Even though the cloud phenomenon is rapidly expanding, there has no literature been found by the researcher the that bridges the gap between Cloud-based business models and Customer Equity. More specifically, no marketing literature about Customer Equity has been found thus far that focuses on the Cloud-based industry and country differences. There are measures of electronic services quality in the IT and Service literature (Parasuraman, Zeithaml & Malhotra, 2005), but the link to Customer Lifetime Value has not been made yet. For example, Rust, Lemon and Zeithaml (2004) only investigated the Sub-drivers in traditional industries such as aviation and grocery. Ramaseshan, Rabbanee and Hui (2013) investigated Customer Equity in the telecommunication industry but the authors did not look at the effect of Sub-drivers which is essential to estimate the financial impact.

Another relatively unexplored area in the marketing literature is the mechanism through which the drivers impact the Customer Lifetime Value. Some authors have focused on the antecedents of the drivers (Zhang, Ko & Lee, 2013) and others on trust as a mediating variable (Ramaseshan, Rabbanee and Hui, 2013). However, customers who have the intention to stay loyal are more likely to focus on long-term benefits and are more profitable than others (Doney and Cannon, 1997; Ganesan, 1994). Also, due to the shift of software-ownership from the client to the provider and the subscription based business models, loyalty intention is seen as an important element in the Cloud-based SaaS industry (Katzan, 2009).

1.3. Research goal

This article attempts to bridge the gap between Customer Equity and Cloud-based services by focusing on the drivers of Customer Equity in the Dutch and US Cloud-based industry and the mediating role of loyalty intentions. Based on the above arguments an attempt will be made to answer the following main research question:

“What is the impact of the Customer Equity sub-drivers on the likelihood of becoming a

paying user in the Dutch and American Cloud-based industry and to what extend can the

impact be explained by loyalty intentions?”

(7)

formulated.

Qualitative sub-question:

“What are the most important sub-drivers of Cutomer Equity in the Cloud-based SaaS industry in The Netherlands and the USA?”

Quantitative sub-question:

“To what extend is the relationship between the Customer Equity drivers and loyalty intentions moderated by the country of residence?”

Academic contribution

By answering these questions, this paper contributes to the customer equity literature in the following ways: First, it extends the existing literature and combines two streams of research two streams of research by extending the Customer Equity framework with the mediating variable “Loyalty Intentions” so that it can fit within the classical Theory of Planned Behavior. Second, this study is the first of its kind to examine customer equity in the cloud industry. Third, this research adds to the Customer Equity literature on a cross country level. Most studies thus far are based on users in a specific country who may exhibit idiosyncratic preference patterns. By focusing on the users from The Netherlands and the Unites States, this study adds to the knowledge about differences in countries. Managerial contribution

From a managerial perspective this research will provide valuable insight for firms with a cloud-based business model. First, it is highly relevant to know which equity drivers are most important to cloud customers. With this knowledge strategic trade-offs can be made in terms of marketing investments, product innovations and service improvements. Second, it is important to know under which circumstances an equity driver will be most likely to impact CLV so that managers can take into account contingency factors. Third, marketing managers in the cloud industry can use this information to optimize their acquisition and retention activities by using the CLV categories as strategic customer segments which can be targeted. Lastly, a fast majority of the technology firms with a cloud business

(8)

model have an international strategy. This research contributes to the management of these firms by presenting a cross-country analysis.

To conduct this study several research methods will be used. This research will have two distinct phases. Because the potential drivers in the Cloud industry are unknown thus far, Phase 1 will be an exploratory qualitative research. After this qualitative explorative phase, a quantitative explanatory approach will be taken to estimate the impact of the drivers on the dependent variable.

2. Literature review

2.1. Customer lifetime value (CLV)

According to the resource-based view (RBV), resources that are valuable, rare, inimitable and non-substitutable (Bamey, 1991) enable firms to develop and maintain competitive advantages. Firms need to utilize these resources to reach a superior performance level (Collis & Montgomery, 1995; Grant, 1991; Wemerfelt, 1984). The marketing discipline adopted a resource-conscious view between 1996 and 2004, with a focus on the use of organizational resources to improve marketing effectiveness and customer profitability (Kumar, 2015). Many marketing scholars have accepted the general RBV logic because it gives a way of looking at the role that customers play in the creation of value for the firm. In this way, customers and their relationships with the firm can be treated as critical resources that function as relational market-based assets. These assets contribute to competitive advantage and can be developed, leveraged and valued in a similar way as the traditional resources of the firm are valued (Srivastava, Shervani & Fahey, 1998).

Customer Lifetime Value or CLV, is a way to measure market-based assets as it evaluates the long-term value of customers with the company. It was firstly defined by Kotler (1974:24) as the present value of the future profit stream expected over a given time horizon. CLV has been studied under different names, such as Customer Profitability (CP), Lifetime Value (LTV), Net Present Value

(9)

(NPV), or Customer Equity (CE). In general, the CLV framework makes a bridge between marketing and finance and is, to a great extent, based on the popular DFC technique in financial valuation.

Nonetheless, the Discounted Cash Flow (DCF) method is different from the CLV method in two ways. (1) DCF is typically not defined and estimated at an individual customer or segment level, allowing no room for targeting and segmentation and (2) unlike in CLV techniques, DCF methods do not incorporate the possibility for future customer churn, which defined a retention rate (Gupta & Lehmann, 2006; Gupta & Lehmann, 2008). These differences allow CLV to be used to measure the return on marketing and value of the total customer base within a certain time period referred to as Customer Equity.

2.1.1. Customer Equity and Return on Marketing

The concept of Customer Equity was first suggested by Blattberg and Deighton (1996) and later refined by several marketing researchers (Reinartz and Kumar, 2000; Rust et al., 2000, 2004). The conceptual framework in figure 1 is made by Rust et al., (2004) and is one of the first methods of measuring customers as market-based assets. The Customer Equity framework integrates a strategic decision-making approach that is customer-centered, theoretical, and practical enough for managerial application.

The model is used as the criterion that balances spending between getting and keeping customers. In other words, it

is the sum of two net present values: the returns from acquisition spending and the returns from retention spending. According to Blattberg and Deighton (1996), the key question for judging new products and customer service initiatives is, “will it grow our customer equity?”. Likewise, Rust et al.,

(10)

(2004) linked marketing investments to company performance by looking at specific drivers that can be improved and cause an increase in the CLV. The sum of all the individual CLVs is the Customer Equity of the firm.

Considering the profound impact of the Return on Marketing framework in the CLV literature, this current paper will use the framework as a basis for further examination. In this way marketing is viewed as an investment that produces an improvement in customer perceptions which will lead to an improvement of a driver of customer equity if customers become or remain loyal. As a result there can be an in increase in customer attraction and retention that leads to a higher CLV. The increase in the sum of the total CLVs, minus the cost of marketing investment results in a return on marketing investment. Using the equation of Rust et al., (2004) the CLV for each customer can be calculated:

Tij = the number of purchases customer i makes during the specified time period;

dj = the firm‟sdiscount rate;

fi = the average number of purchases customer i makes in a unit time (e.g., per year);

Vijt = the customer i‟s expected purchase volume of brand j in purchase t;

pijt the expected contribution margin per unit of brand j from customer i in purchase t*;

Bijt the probability that customer i buys brand j in purchase t (or retention rate).

*Contribution margin, or dollar contribution per unit, is the selling price per unit minus the variable cost per unit

This current research takes the Rust et al., (2004) framework as a basis and focuses on the the probability that customer i buys brand j in purchase t and how this is impacted by the drivers of customer equity. This element is mainly influenced by the company‟s marketing investments when

(11)

specific drivers are improved. Since no behavioral retention data can be obtained, this current research will look at the likelihood of a user becoming paid customer. Even though this current research takes the Rust et al., (2004) framework as a basis, there are still some major points of critique to be made that challenge the main arguments of the authors.

First, the authors claim that managers in general can use the framework to estimate the impact of strategic investments. However the article lacks further generalizability since it only examines US customers in traditional industries. There is reason to doubt whether the same drivers can be readily copied to 1) other geographical areas and 2) innovative digital industries where software is sold as a service.

Second, the research does not examine mediating variables through which CLV is impacted. Thus it lacks some theoretical explanation for the mechanism that can potentially explain the relationships in the framework.

Third, the authors take the variable “last purchased” as a proxy for consumer inertia which may lead to biased results. By doing this the authors try to separate the impact of the individual drivers on CLV from the effect that customer inertia has on CLV. However, in many digital industries, online services are free to use and a big challenge is to convert users into paying premium customers. Therefore this current research will focus on the intention of customer to stay loyal and the likelihood of them to convert into a paying customer rather than the “last purchased”.

2.1.2. Drivers of Customer Equity

In order to improve CLV on an individual customer level, firms can improve the customer perception of the company and the company‟s offerings. This can be done by improving the popular customer equity drivers as mentioned by Rust, Lemon and Zeithaml (2001, 2004). These authors propose three main customer equity drivers (see figure 2). The first and most important is Value Equity and this is the customer‟s objective assessment of the utility of a product, based on perceptions of what is given up for and what is received (Rust, Zeithaml & Lemon, 2000, pag 68; Lemon, Rust & Zeithaml, 2001). Sub-drivers of Value Equity include quality, price and convenience. The second equity driver is Brand

(12)

Equity and this is the “...customer‟s subjective and intangible assessment of the brand, above and beyond its objectively perceived value.” (Rust, Zeithaml & Lemon 2000 pag. 81). Sub-drivers include ad-awareness, corporate citizenship, information community events, ethical standards. The last driver is Relationship Equity. This is everything above and beyond the objective and subjective evaluations of the firm. Sub-drivers of Relationship Equity may include loyalty programs, preferential treatment, being known, recognizes as being special, community and trust.

Figure 2. Customer Equity Framework Rust et al., (2004)

2.1.3. Relevance of the Sub-drivers

It is evident that the formulation and use of the sub-drivers have a big impact on the outcome of the predictive power of the model. Rust and other scholars have formulated these drivers in the beginning of the 21st century. Since that time economic changes have transformed current organizational structures and business models which may have led to a change in the drivers and their sub-drivers. Also, media usage patterns have endured substantial changes the last ten to fifteen year. Specifically, the variations in customer preferences toward media channels have increased: people are spending more time on interactive media (Internet and mobile services) than on traditional media (radio and print) (Kumar 2015).

Sub drivers

Customer

Equity Drivers

Customer

Lifetime Value

CLV

Value Brand Relationship Sub-drivers

(13)

Therefore, it can be assumed that the content of the drivers in the Cloud industry might be significantly different. For example, Rust et al., (2004) studied the effect of drivers in traditional industries such as aviation and groceries without explaining the mechanism through which the drivers impact the customer life time value. Ramaseshan Fazlul and Hui (2013) look at the customer equity drivers in the telecommunication industry and they used trust as a mediating variable. However, the authors do not describe any sub-drivers and they took loyalty as a dependent variable instead of CLV. In a more recent publication by Ivens, Walsh and Schaarschmidt (2015) the degree of trust is used as a mediator. The authors tried to explain the relationship between a company‟s reputation and customer equity and thus they do not make use of customer equity sub-drivers either.

2.2. Loyalty Intentions: Extending the Customer Equity framework

Loyalty Intentions is addressed by several marketing scholars (Johnson, Herrmann & Huber, 2006; Bendapudi and Berry 1997; Morgan and Hunt 1994). Loyalty intention can be termed as a buyer‟s favorable attitude toward the seller that results in repeating buying behavior (Srinivasan et al., 2002). Drawing on the organizational behavior literature (Meyer and Allen 1997), commitment to stay or become loyal has also been defined as a desire to maintain a relationship with a business (Moor- man, Deshpandé, and Zaltman 1993; Morgan et. al., 1994). The concept of “loyalty” also is used in traditional key account models (e.g. Sengupta et al., 2000; Workman et al., 2003) and is considered essential to successful business partnerships (Morgan et. al., 1994). Loyal customers who are committed to the relationship are more likely to focus on long-term benefits and engage in cooperative actions beneficial to both partners and thus reducing transaction costs (Doney and Cannon, 1997; Ganesan, 1994).

In the current marketing literature it is not clear whether the customer equity framework would be adequate in predicting the effectiveness of marketing investment to gain business customers‟ loyalty. However, some researchers have found that relational commitment can impact financial outcomes. For example, Morgan and Hunt (1994) developed the commitment – trust theory. In this theory

(14)

relationship marketing is explored and the main idea is that loyalty intentions and trust should serve as two key mediating variables between relationship antecedents and different account performance. Also, Palmatier, Dant, and Grewal (2007) found evidence that loyalty intention is an important mediator and direct antecedent to performance outcomes such as sales growth, financial performance.

Based in the above arguments, the prediction in this study is that loyalty intention mediates the Customer Equity drivers-retention relationship and it represents a significant move away from the traditional view of Customer Equity drivers as depicted by Rust et al., (2004). The main shift in reasoning is: rather than drivers being related directly to retention and acquisition, drivers only significantly influence intention. To back up this critical extension to the traditional Customer Equity framework, we turn to one of the main theories in social psychology: Theory of Planned Behavior.

2.2.1. The Theory of Planned Behavior

The Theory of Planned Behavior focuses on the relationship between belief, attitudes and behavior and is an extension of one of the three classic persuasion models of psychology (Fishbein, 1967a, 1967b; Fishbein & Ajzen, 1975; Wicker, 1969). According to this theory, salient beliefs (e.g., beliefs about behavior, social norms and control) are held to determine our attitudes, the subjective norms we possess and the extent we perceive to be in control over our behavior (Peceived Behavioral Control, PBC). However, the implication in psychological research in the 70‟s and 80‟s was that there seemed to be a discrepancy between the actual behavior and attitudes. Wicker (1969) concluded: "taken as a whole, studies suggest that it is considerably more likely that attitudes will be unrelated or only very slightly related to overt behaviors than that attitudes will be closely related to actions" (p. 64).

Thus, further research was required to gain a deeper understanding about attitude-behavior relations and consequently psychologist considered factors that might mediate the relationship between attitudes and behavior. These efforts led to an addition to Theory of Planned Behavior by adding the mediator „‟Behavioral Intention‟‟. This mediator is regarded as a “summary of the motivation required to perform a particular behavior, reflecting an individual's decision to follow a course of action, as well as an index of how hard people are willing to try and perform the behavior” (Armitage &

(15)

Christian, 2003 p. 190). Behavioral intention is important to the theory because these intentions are determined by drivers or antecedents depicted in figure 4.

Figure 4. Theory of Planned Behavior (source: Armitage & Christian, 2003).

2.2.2. Customer Equity framework linked to Theory of Planned Behavior

As a result of the additional mediator “behavioral intention” psychologist were able to generate better predictors of specific behaviors. Likewise, to get a deeper understanding of Customer Equity, one needs to examine the variable that represents the generative mechanism through which the independent variables “the drivers” are able to influence the dependent variable “likelihood of becoming a paying customer". Just like the mediator “behavioral intention” is an index of how hard people are willing to try and perform a behavior, “loyalty intention” is an index for effort at preserving and the willingness of advancing the relationship and maintain its value (Dwyer, Schurr & Oh, 1987; Moorman, Zaltman & Deshpande, 1992; Morgan & Hunt, 1994). Morgan and Hunt (1994,p. 23) argue that commitment occurs when a relationship is so important that people are motivated to put for effort at maintaining it. Stronger intentions of loyalty lead to increased effort to stay with the firm and it can be argued that this will increase the probability that customer i buys brand j in purchase t (retention rate). In figure 5 the extended Customer Equity framework is therefore linked to the classical Theory of Planned Behavior.

(16)

Figure 5. Extended Customer Equity framework linked to Theory of Planned Behavior (Source: Author)

To conclude, Rust‟s et al., (2004) review of the Drivers-Retention relationship does not reflected the fact that intentions are the principal proximal determinant of behavior. The authors omit a major element within the study of behavioral change in psychology that can possibly explain retention and acquisition to a greater extend. Simply because: If there is no intention to becoming or remain loyal, churn is more likely. This study attempts to combine two streams of research by extending the Customer Equity framework with the mediating variable “Loyalty Intentions” so that it can fit within the classical Theory of Planned Behavior. To the best of knowledge, no empirical study has been conducted about this issue so far, hence the current study will aim to fulfill this key research gap.

2.3. Country differences

Another important element that has had little attention in the literature is that country differences may have an impact on the content of the drivers. More specifically, a driver that is profound in the US may not be so important in The Netherlands because of cultural dissimilarities. However, all the studies mentioned thus far that focus on Customer Equity and CLV, are based on users in a specific country who may display distinctive preference patterns. This is an important limitation because "one should take extreme caution in generalizing results found in a given country to other countries" (Jaffe

Improvement in

customer

Equity Drivers

Improvement in

customer

Perception

Improvement

in Loyalty

Intentions

Improvement in

actual

Retention/

Acquisition

Actual

Behaviour

Behavioral

intention

Attitude Behavioral Beliefs Subjective Norms Normative Beliefs Perceived Behavior Control Control Beliefs

(17)

& Nebenzahl 2001, p.53). Also, customer preference is a dynamic occurrence, which means it changes over time and replication research in multiple countries is necessary to establish the stability of the findings in other contexts.

2.3.1. Hofstede

These potential different findings can be contributed to the difference in core values across nations. Research on cultural values (Hofstede 2003) suggests that nation populations differ in their value systems. For example, American people are known for underscoring individualism more than Dutch people. Because brands may help consumers express themselves to and differentiate themselves from other people, brands might play a greater role for consumer‟s decisions in the United States than in the Netherlands (Hofstede 2003).

2.3.2. Brand Relevance in Category (BRiC)

A more relevant and recent study in the literature that focuses on country differences is the paper about brand relevance in categories (Fischer, Völckner & Sattler, 2010). The authors researched how relevant brand-building activities are for a company‟s success compared with other investment alternatives. They found out that the importance of brands varies across countries and introduce a construct called “brand relevance in category” (BRiC). This construct measures the overall role of brands in customers‟ decision making in a specific category. One of the main conclusions of the article is that the people from the Unites States are, in general more sensitive to brands than other people in the study (Fischer, Völckner & Sattler, 2010). These insights can be valuable the Customer Equity literature since “Brand Equity” is one of the three equity drivers in the model. As there was no literature found about customer equity differences between nationalities, this study can help to fill the gap in research about the impact of equity drivers in the Dutch and US cloud industry.

(18)

2.4. Cloud computing

Cloud computing is a IT concepts that enables convenient, on-demand network access to a shared pool of configurable resources e.g., networks, servers, storage and applications (Chong and Carraro ,2006). Even though the Cloud grows fast, the main idea behind cloud computing is not a new one. John McCarthy in the 1960s already envisioned that computing facilities will be provided to the general public like a utility. (Parkhill, 1966).

The National Institute of Standards and Technology (NIST) promotes the U.S. economy and public welfare by providing technical leadership for the nation‟s measurement and standards infrastructure. This institute defined cloud computing in the following way: “Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction” (NIST, 2011, p2).

According to the institute there are five essential characteristics of cloud computing. The first is “On-demand self-service”. This refers to the customer‟s ability to access computing abilities as needed without the interference of human interaction with the provider. The second is “Broad network access”. This refers to the digital capabilities being available over the network and accessed through standard mechanisms that support the use of various devices such as mobile phones, tablets, laptops, and workstations. The third is “Resource pooling”. This refers to the provider‟s digital resources being combined to serve multiple consumers while the resources are dynamically assigned according to consumer demand. The fourth is “Rapid elasticity”. This refers to the ability of the digital resources to scale rapidly up and down according to demand. The last is “Measured service”. The refers to the fact that usage can be measured, controlled, and reported in full transparency for both the provider and consumer. Like electricity or municipality water, IT services are charged per usage metrics e.g., pay per use, storage in GB and bandwidth (NIST, 2011).

(19)

Figure 6. Cloud model overview. Created based on the text in NIST (2011) (source: Author).

2.4.1. Cloud-based business models

Chong and Carraro (2006) define software-as-a-service models as software deployed as a hosted service and accessed over the Internet. Business models based on cloud computing can be rapidly expanded and released with minimal management effort or service provider interaction. The key features of cloud- based business models are location and how the software is accessed by the customer. The business model for cloud computing reflects how service providers can increase revenue and how clients can reduce operational costs of services. According to Youseff, Butrico, and Da Silva (2008) the prominent features of the cloud computing business model are:

1. The ownership of the software is transferred to the cloud service provider.

2. The responsibility for hardware, application software, storage facilities, and professional services resides with the provider.

3. Systems software is available from a trusted vendor for supporting cloud services.

The could customer will have to give up a certain level of control to benefit from the economy-of-scale supplied by the provider. This changes the traditional interaction with the provider in which

Deployment models

Private cloud Public cloud Community cloud. Hybrid cloud

Service Models

Software as a Service (SaaS) Platform as a Service (PaaS) Infrastructure as a Service (IaaS).

Essential Characteristics

On-demand self-service. Measure

d service elasticityRapid Resource pooling.

Broad network

(20)

software was bought and controlled on a local scale in house. Also, because of the change in ownership of the software, the high responsibility of the provider and subscription based business models, it can be expected that loyalty intention are important in this context (Youseff, Butrico, and Da Silva, 2008). This is an element that, to the best of knowledge has not been investigated before in the literature of cloud business models.

(21)

3. Conceptual framework

In this chapter the researcher elaborates on the relationships and effects between the subjects previously introduced. The section starts with the main question and then shows how this relates to the other subjects in the conceptual model. This model and the qualitative pre-study guide my hypotheses, which are addressed afterwards.

As discussed in the first chapter, the drivers of customer equity in the Cloud industry are unknown thus far. Changes in business models and the high penetration of digital services are likely to have altered user‟s perspective of and reactions to the equity drivers. Thus, promissory expectations can be unmet and the customer equity can decrease as organizations might misinterpret the current content of the drivers and sub-drivers. In order to minimize these negative effects in the rapidly growing cloud industry, research first needs to clarify the content of the equity drivers from a user‟s perspective. Therefore this study has a qualitative sub-question that will be used to discover the most important drivers.

3.1. Main research question

“What is the impact of the Customer Equity drivers on the likelihood of becoming a paying user in the Dutch and American Cloud-based industry and to what extend can the impact be

explained by loyalty intentions?”

Quantitative sub-question:

“To what extend is the relationship between the Customer Equity drivers and loyalty intentions moderated by the country of residence?”

3.1.2. Qualitative sub-question:

“What are the most important sub-drivers of Customer Equity in the Cloud-based SaaS industry in The Netherlands and the USA?”

(22)

3.1.3. Framework

The variables in the main question are depicted in the following conceptual model. This can be used to understand the subject matter and will function as the base of the quantitative phase of the research. Note that the dependent variable “Likelihood of become a paying customer” is used instead of the retention variable the Customer Life Time Value model. Rather than the traditional “retention rate”, I focus on the probability that user “x” becomes a paying customer since this reflects the crucial conversion rate in a freemium business model of cloud storage services. Other variables in the equation (e.g., the firm‟s discount rate, average number of purchases, purchase volume, contribution margin) can easily gathered in the database of a firm. When these variables are entered in the equation mentioned in section 2.1.1, the CLV can be estimated.

Figure 7. Conceptual Framework “Customer Equity in the cloud industry” (source: Author).

NL vs USA

Loyalty

intentions

Likelihood

paying user

Customer

Equity

Sub-drivers

(23)

4. Method Section

The following chapter will show the research design of this paper. In the first sub-section the research philosophy will be outlined. Next, the study design is explained, which is followed by the qualitative pre study.

4.1. Research philosophy and approach

There are two ways to think about research philosophy, which are ontology and epistemology. These two views indicate how the researcher views the world and makes basic assumptions that will lead to a specific research design (Saunders, 2011). Positivism is the epistemological view being most often used in management studies Gephart (2004). The positivist approach attempts objectively to discover the truth, whereas post-positivists do not try to find the “truth” but rather interpret findings and emphasize meaning, seeing and experience (Ryan, 2006). One of the most common forms of post-positivism is a philosophy called critical realism. This current study can best be categorized in the post-positivistic view in which the researcher leans strongly towards critical realism. A critical realist believes that there is a reality independent of our thinking about it that science can study. This leads to the main assumption that all observation is imperfect and has error. Thus, the author of this study is critical of our ability to know reality with certainty. Because the assumption that all measurement is imperfect, the researcher emphasizes the importance of multiple measures and observations to try to get a better understanding of what is going on in reality. Also, because the author believes that all observations and researchers are inherently biased by their cultural experiences and world views, he attempts to limit these biases by adoption a mixed methodology.

4.1.1. Methodology

As stated above, this study is a mixed method research that aligns with the main philosophy of the researcher. Two data collection techniques and two analytical procedures were used to answer the research questions. Since both quantitative and qualitative methodologies are combined in multiple stages of the study, it can be considered a fully integrated mixed method research (Leech &

(24)

Onwuegbuzie, 2009; Ridenour & Newman, 2008). The integration of the two methods is even more noticeable by the way the data is merged and transformed, with the qualitative data being „quantified‟.

Literature suggests that mixed method research may be conducted sequentially or concurrently (Creswell & Plano Clark, 2007). In concurrent mixed method research both qualitative and quantitative methods are involved with a single phase of data collection and analysis. However, in this current study a sequential exploratory design has been chosen. Rather than one single phase of data collection and analysis, sequential design involves at least two phases. The researcher has followed the use of qualitative data collection and analysis with the quantitative survey in order to expand and elaborate on the initial set of Equity drivers. This was to establish the issues that were important to the users of the cloud storage software. In detail the researcher counted specific events in the data related to the Equity sub-drivers as frequencies so that the highest occurring drivers could be statistically analyzed.

Figure 8. Overview of research methodology. Source: Author

Research philosophy & approach

Critical

realism

Deduction

Method & strategy

Mixed method complex

Quantitative and qualitative

Cross-sectional

Data collection

Structured

(25)

4.1.2. Design of the study

To conduct this cross-sectional study a mixed methodology has been used which consist of two phases. In the first phase a qualitative approach has been be taken in to select the measures (drivers) that has been be used in de second phase. This second phase is quantitative and the statistics were used to test the hypotheses and generalize the findings.

4.2. Qualitative data collection and analyses (Phase 1)

To the best of the researcher‟s knowledge, no literature exists about the customer equity drivers in the Cloud industry. For this reason the first phase of the research is exploratory and qualitative. To collect the data 14 semi-structured interviews are conducted with user in the cloud data storage industry. The interviews are transcribed and the date coded into short quotations that are subsequently designated to categories. The categories are pre-determined based on potential drivers from other industries that can be measured using pre-validated scales.. After each quotation has been designated to a category, a frequency distribution has been made to provide insight into which drivers are most likely to be important in the Cloud industry.

For this qualitative phase interviews were conducted with seven Dutch people and seven Americans. Six out of the fourteen people are female and the average age is 26. The goal was to gain insights into their perception of cloud based services and to see which drivers were most important to them. Based on the interview transcriptions a total of 227 quotes were collected. Some quotes were assigned to two initial drivers of Rust et al. (2004) which are Relationship and Brand Equity. However, in this paper the third driver of Rust et al. (2004) which is Value Equity is split out into more relevant by sub-drivers related to software quality. Instead of only looking at Price, Quality and Convenience, in this paper the term Quality is more thoroughly analyzed by incorporating the Quality drivers used by Parasuraman, Zeithaml and Malhotra (2005) in their article about E-Service Quality Assessment. Subsequently the quotes were assigned to the following drivers: Brand, Relationship, Price (acquired from Rust et al. 2004), Efficiency, Fulfillment, Privacy, System Availability, Contact (acquired from

(26)

Parasuraman et al . 2005 ). Convenience is replaced by Efficiency since the authors use similar descriptions for these two drivers.

In this section each driver will be discussed. If one quote or multiple quotes from a participant was assigned to a driver, then that driver was considered to be present. The frequency occurrence in terms of percentages is depicted by “F” and the number of quotes assigned to that driver is depicted by “n”. In order to provide a clear perspective of the categories, each dimension contains a definition and an example quote.

Efficiency: The ease and speed of accessing and using the application. (Parasuraman et al., 2005, p8).

Interview Quote: “I use the very limited version of DropBox, it seems to very

simple and user friendly and it seem to be easy to share with people. Also the

fast and efficient files uploading and downloading”.

(F=25%, n=57)

Fulfillment: The extent to which the site‟s promises about file storage and synchronization are fulfilled. (Parasuraman et al., 2005, p8).

Interview Quote: “If there would be much larger storage capacity (1TB+) and

if the platforms can be more personalized based on my preferences such as

adding theme”.

(F=24%, n=54)

System availability: The correct technical functioning of the site. (Parasuraman et al., 2005, p8).

Interview Quote: “However, I find other online file hosting sites (Rapidshare,

ICloud, Mediafire, Depositefiles) less reliable as there were occasions that

(27)

(F=11%, n=33)

Privacy: The degree to which the site is safe and protects customer information. (Parasuraman et al., 2005, p8).‟

Interview Quote: “If the company gives third parties access to the files I store

on the cloud, or if the company provides third party my personal usage

information I will consider to leave”.

(F=13%, n=29)

Contact: The availability of assistance through telephone or online representatives. (Parasuraman et al., 2005, p8).

Interview Quote: “Furthermore, if I can’t figure things out by myself I can

easily go to the Mac store, knowing that they will be able to sort my problem

out”.

(F=1%, n=24)

Price The competitiveness of the prices each of these services (Rust et al., 2004, p3).

Interview Quote: “There are so many optional ways to get things for free, I

can have so many other ways to use the storage cloud so I don’t I will pay”.

(F=7%, n=16)

Brand Focuses on the overall perception of brand image (Rust et al., 2004, p3). Interview Quote: “Well-known brands with reliable image make it more

convincing for me to use their services as I can rest assured that they provide

safe and quality services”.

(28)

Relationship Emphasis on long-term customer relationships rather than short-term transactions (Rust et al., 2004, p3).

Interview Quote: “I feel that I have a very trusting relationship with Google to

be able to have important docs. On the cloud drive and the peace of mind that

I know it won't go anywhere”.

(F=5%, n=3)

Figure 9. Interview quotes per sub-driver. Source: Author 57 54 33 29 24 16 11 3 0 10 20 30 40 50 60 N u m b er o f q u o tes

Customer Equity sub-drivers

(29)

5. Hypotheses

5.1.1. Customer Equity sub-drivers

The previous qualitative analysis provides evidence of which driver will be best suitable for further analysis. Due to the length limitation of this thesis, only the first three top drivers will be used in the quantitative part. In this section the hypothesis will be formulated based on the qualitative analysis and literature. According to Rust et al., (2004) marketing can be viewed as an investment that produces an improvement in customer perceptions. Thus, the goal of the investments made is to affect one of the underlying drivers that will positively impact the perception of a customer or prospect about the product or service of the firm. Loyalty intentions can be viewed as a customer‟s psychological disposition toward a service. In a purchase situation, loyalty intentions reflect favorable attitudes toward the brand or firm (Dick and Basu 1994). As result of improved perceptions there is an increased chance of better retention that leads to higher loyalty of customers. Based on the previous qualitative pre-study and the argument in the literature, the following hypotheses were formulated:

H1. There is a positive relationship between the Brand Equity driver and Loyalty Intentions.

Efficiency and fulfillment have been cited as important facets of e-SQ (Yen & Lu, 2008; Wolfinbarger & Gilly 2003). In fact, Wolfinbarger and Gilly (2003) found that reliability and fulfillment ratings were the strongest predictor of customer satisfaction and quality, and the second strongest predictor of intentions to repurchase at a website. Therefore the follow hypotheses are formulated:

H2. There is a positive relationship between the Fulfillment driver and Loyalty Intentions.

H3. There is a positive relationship between the Efficiency driver and Loyalty Intentions.

5.1.2. Loyalty-Intentions

In this current paper Loyalty Intentions is strongly related to the retention rate because without this a customer will tend to churn more easily (e.g. Sengupta, Krapfel & Pusateri, 2000; Workman, Homburg & Jensen, 2003). Higher retention rates mean that customers engage in a relationship with the firm longer. Also, the Loyalty Intentions are “forward looking.” For example, satisfaction is a

(30)

function of current performance, whereas Loyalty Intentions captures the strength of intention the to continue relationship and proceed forward on the long run. On the basis of the literature, it is predicted that Loyalty Intentions has a positive impact on customer retention probability. Since this research is limited in measuring the actual retention rate, the likelihood of a user to become a paying customer will be measured. This adds extra value because most of the cloud based services are based on a freemium business model and looks at retention in terms of keeping their paying customers. It can be assumed that the effect of the customer equity drivers on the likelihood of becoming a paid customer will go through the Loyalty Intentions mediator. Therefore the following hypotheses were formulated:

H4. There is a positive relationship between the Loyalty Intentions and likelihood of becoming a paid

customer. The drivers only impact the likelihood of becoming a paid customer through the intentions

of users to stay loyal

5.1.3. Country differences

According to Hofstede (2003), people from the USA have a higher sense of individualism that the Dutch. This might lead to a stronger desire to express and differentiate. Also in the study of (Fischer) the United States leads overall in how relevant brands are. US citizens use brand more strongly than others to reduce risk and to demonstrate themselves socially. According to the author “The United States has implemented the idea of economic freedom for a long time, and the principles of modern marketing were born here” (Fischer, Völckner & Sattler, 2010 p831). These conditions produced highly competitive product markets with a large variety of products and services. Brands play an important role in guiding the consumer decision under such circumstances. Also the vast majority of cloud providers in the B2C market are American. Therefore the following hypothesis is formulated:

H5. The relationship between the Brand driver and Loyalty Intentions is moderated by country of

(31)

5.2. Quantitative data collection and analyses (Phase 2)

After the qualitative explorative phase and hypotheses formulation a quantitative explanatory approach will be taken to estimate the impact of independent variables on the mediator and dependent variable. Since all the newly determined drivers can be measured using pre-validated scales, the data for the independent variables can be collected using surveys. The data sources are self-reported.

5.2.1. Sample

The sample was selected by means of non-random sampling technique using a quota. This was done because approximately 50% of the respondents needed to be from The Netherlands and 50% from the USA. Thus the sampling frame for the research objectives is multi-country. Non-probability sampling was done in a non-random manner. There were members of the target population who had zero chance of being selected for the survey sample since there is no proper list of all the members available in the population. One important element of the sample is that they must have used one of the cloud storage services at some point is time. To ensure that the respondents in the sample fall in this category and that the sample is representative, a pre-selection question was asked in the beginning. The sample size for this research is set to a minimum of 300. Confidentiality was ensured by making the survey completely anonymous.

5.2.2. Measures

In the study measures with demonstrated reliability were used. To ensure construct validity, standardized validated measures from the literature were used. For the independent variable a pre-validated Likert scale was used with 3 items measuring the construct “Brand Equity” (Vogel, Evanschitzky, & Ramaseshan, 2008). For the mediation variable a pre-validated Likert scale was used also but with 3 items measuring the construct “Loyalty Intention” (Vogel et. al., 2008). Both are a 6-point scales (completely disagree – completely agree) at interval level. For the moderating variable the Country of Residence was taken with the two countries “The United States” or “The Netherlands”. Several control variables were used, gender, degree and device. See the appendix for this list of complete questions.

(32)

5.2.3. Sample and missing values

Data were collected by means of a paid online survey service. Survey administration started on December 3, 2016. The survey was closed five weeks later on January 3, 2017. To perform the statistical analyses, the Statistical software Package for Social Sciences (SPSS) was used. The two reversed items, that were implemented to ensure validity, were recoded. Also, the variables Country and Gender were transformed (1=0 and 2=1). A total of 315 respondents have filled in the survey of which 304 have completed it fully. To ensure a proper division of the countries, 52% of the respondents are Dutch. All variables were analyzed to see to what extent data was missing. Based on a frequency test it was shown that the amount of missing data was < 4% for all variables which resulted 304 usable responses.

Table 1. Sample overview

Frequency Percent

Cumulative Percent

Cloud Google Drive 113 37,2 37,2

OneNote Microsoft 14 4,6 41,8 Dropbox 84 27,6 69,4 iCloud 87 28,6 98,0 Other 6 2,0 100,0 Total 304 100,0

Country The United States 146 48,0 48,0

The Netherlands 158 52,0 100,0 Total 304 100,0 Age 18 – 29 96 31,6 31,7 30 – 44 104 34,2 66,0 45 – 59 54 17,8 83,8 60+ 49 16,1 100,0

(33)

Total 304 99,7 Gender Male 174 57,2 57,4 Female 130 42,4 100,0 Total 304 99,7

Device iOS Phone /

Tablet 123 40,5 40,6 Android Phone / Tablet 47 15,1 55,8 Windows Desktop / Laptop 116 38,2 94,1 MacOS Desktop / Laptop 15 4,9 99,0 Other 3 1,0 100,0 Total 304 99,7

5.2.4. Outliers

After standardizing the scores of the main variables Fulfilment, Efficiency, Brand Equity, and Loyalty Intentions, three outliers were detected (inter-quartile range multiplier > |3|). The distribution was checked and these cases were isolated and therefore deleted to prevent biases in the data.

5.2.5. Normality check

After deleting the outliers, the data was tested for normality. First a histogram of frequencie

distributions was computed. For Brand Equity in particular the histogram showed no “Bell” shape but a heavy negative skewness. To follow up, the descriptive statistics were computed. The skewness and kurtosis of all variables except Brand Equity were within the empirical criteria of acceptable values between 1 and +1. For Brand Equity the skewness statistic was -1,051 a substantial negative skewness and the kurtosis 2,471.

(34)

Table 2. Descriptive Statistics

N Skewness Kurtosis

Statistic Statistic Std. Error Statistic Std. Error

Fulfilment 304 -,688 ,140 ,352 ,279 Efficiency 304 -,580 ,140 ,142 ,279 Brand 304 -1,051 ,140 2,471 ,279 Loyalty 304 -,441 ,140 ,143 ,279 Valid N (listwise) 304

For extra validation the Kolmogorov-Smirnova Shapiro-Wilk Test of Normality was done. The result were significant (p <.05) and confirmed the previous results of no- normality. The null hypothesis of normal distribution was rejected.

Table 3. Tests of Normality

Kolmogorov-Smirnova Shapiro-Wilk

Statistic

Df Sig. Statistic df Sig.

Brand ,180

304 ,000 ,910 304 ,000

5.2.6. Normalization

To normalize the distribution of Brand Equity and prepare it for further analysis, a variable

transformation technique was used. The following technique was used: X=Log10(K-X), where K= the highest value of the variable X, +1. After variable transformation the descriptive statistics were computed again. This time the skewness statistic was -,122 and the kurtosis -.380 which is conform the empirical criteria between -1 and 1.

(35)

Table 4. Descriptive Statistics after variable transformation

N Skewness Kurtosis

Statistic Statistic Std. Error Statistic Std. Error

Fulfilment 304 -,688 ,140 ,352 ,279 Efficiency 304 -,580 ,140 ,142 ,279 Brand 304 -,122 ,140 -,380 ,279 Loyalty 304 -,441 ,140 ,143 ,279 Valid N (listwise) 304

5.2.7. Reliability and correlation

To examine the consistency of measurements a reliability analysis was done. Reliability checks were done for Brand Equity, Fulfillment, Efficiency and Loyalty intentions. To estimate the internal consistency, Cronbach‟s alpha was used to verify if all the items in one scale measure the same thing. Table 5 shows that all variables have a Cronbachs alpha > .7, which indicates high level of internal consistency (Saunders et al., 2009). All variables have a sufficient reliability. The corrected item-total correlations indicate that all the items have a good correlation with the total score of the scale (all above .30).

To prepare the data further for hypothesis testing, new variables were created based the current ones to get scale means. The means and standard deviations are shown in table 2 together with the correlation coefficients for all the variables. Brand Equity, Fulfillment and Efficiency are stronger predictors of Loyalty Intentions with a Pearson correlation coefficient of respectively r=0,557**, r=0,527**, r=0,682** and the significance values less than 0.01.

(36)

Table 5. Means, Standard Deviations, Correlations and Reliability scores (Cronbach’s Alpha) Variables M SD 1 2 3 4 5 6 7 8 9 10 11 1. Country 1,5 0,50 - 2. Cloud Service 2,5 1,30 -,034 - 3. Age 3,2 1,05 -,295** _,192** _- 4. Gender 1,4 0,50 ,096 ,167** ,147* - 5. Device type 2,5 1,50 -,159** _-,174** _,145* _-,065 _- 6. Usage 4,2 1,30 ,213** _,004 _-,024 _-,089 _,069 _- 7. Paying likelihood 2,6 1,43 -,046 ,101 -,143* _-,184** _,068 _,421** _- 8. Fulfilment 4,8 0,84 ,271** _-,014 _-,231** _-,029 _-,074 _,336** _,195** _(,777) 9. Efficiency 4,6 0,98 ,281** -,165** -,313** -,041 ,018 ,449** ,236** ,629** (,788) 10. Brand Equity 0,3 0,17 ,127* _-,111 _-,192** _-,121* _-,127* _,264** _,289** _,352** _,422** _(,753) 11. Loyalty Intentions 4,3 1,09 ,222** _-,181** _-,206** _-,078 _,002 _,392** _,356** _,527** _,682** _,557** _(,849)

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed). n=304

(37)

5.2.8. Hypotheses testing

To test the hypotheses, a regression analysis was conducted. Overall the model explains 13.4% of the variance of Likelihood Paying User, which is statistically significant (p < 0.01). It also explains 55.3% of the total variance of Loyalty Intentions which is significant (p < 0.01).

Table 6. Regression analysis, drivers and mediation

Consequent

Loyalty Intentions (M) Likelihood Paying User (Y)

Antecedent Coeff SE P Coeff SE P

Brand Equity (X1) H1 .309 .051 .000** .143 .092 .120 Fulfillment (X2) H2 .109 .051 .033* .013 .100 .895 Effeciency (X3) H3 .487 .052 .000** -.037 .116 .751

Loyalty Intentions (Med) - - - H4 .450 .076 .000**

Constant -.002 .057 .971 2.575 .076 .000* R2=.553 R2=.134 F(73.973) = .453, p<.000 F(11.592)= 1.795, p<.000 *p<0.05, **p<0.01

The effect of Brand Equity on Loyalty Intentions is significant (SE=.051, p < 0.01). This indicates that hypothesis 1 is supported and that two users that differ by one unit on Brand Equity are estimated to differ 0.051 on Loyalty Intentions. The sign is positive, meaning that those relatively higher in Brand Equity are estimated to be also higher in their loyalty intentions. The effect of Fulfillment on Loyalty Intentions is also significant (SE=.051, p < 0.05). This indicates that hypothesis 2 is also supported and that two users that differ by one unit on Fulfillment are estimated to differ 0.051 on Loyalty Intentions. The sign is also positive, meaning that those relatively higher in Fulfillment are estimated to be also higher in their loyalty intentions. Hypothesis 3 is also supported. The effect of Efficiency on Loyalty Intentions is statistically significant (SE=.052, p < 0.00). This means that there is a positive

(38)

Hypothesis 4 is also supported with statistically significant evidence (SE=.076, p < .000). Two users that differ by one unit on Loyalty Intentions are estimated to differ 0.076 on the likelihood of being a paid customer of the cloud service. The sign is also positive, meaning that those relatively higher in Loyalty Intentions are estimated to be more likely to pay for the service. There is no direct effect of Brand Equity (p=.120), Fulfillment (p=.895) and Efficiency (p=.751) on the dependent variable Likelihood Paying User. The effect is only indirect via loyalty intentions. Thus, the data supports full mediation taking place.

Hypothesis 5 is not supported. No evidence was found that indicates that the effect of Brand Equity on Loyalty Intention is contingent on the Country of Residence. The interaction affect between of Brand Equity and Country of Residence was not significant (p=.737).

Table 7. Testing moderation

Conditional indirect effect

Loyalty Intentions (M) Antecedent Coeff SE P Brand Equity (X1) .309 .051 .000** Country (Mod) .008 .081 .915 Brand*Country (X1*Mod) H5 -.028 .084 .737 Constant -.002 .057 .971 R2=.553 F(73.973), = .453 p<.000 *p<0.05, **p<0.01

(39)

Figure 10. Complete model and results *p<0.05, **p<0.01

Brand Equity

(X1)

Fulfillment (X2)

Effeciency (X3)

Loyalty Intentions

(Med)

Likelihood Paying

User (Y)

Country of

residence (Mod)

H1**

H2*

H3**

H5

H4**

(40)

6. Discussion

In this section the significance of the findings will be discussed. First, the answer to the research question will be argued. Second, how the theory and concepts relate to the data that was gathered will be analyzed. Third, the implications for real-world practice will be discussed.

6.1. Qualitative pre-study: Most important sub-drivers of Customer Equity

This research looks at the drivers of customer equity and to what extent they impact the likelihood of

a user becoming a paying customer via the mediating variable Loyalty intentions. The findings of this study shed light on the relatively new field of cloud-based services which, up until this point, was not yet linked to the Customer Equity marketing construct. Because this construct was not yet previously used in the field of digital cloud services, a qualitative pre-study added more insights to the relevance of the drivers. In contrast to the article from Rust et al. (2004), this article shows that other drivers are important to consider when working in a field with digital cloud services. The qualitative research question, “What are the most important sub-drivers of Cutomer Equity in the Cloud-based SaaS industry in The Netherlands and the USA?”, can be answered. The qualitative results show that Efficiency, Fulfillment and Brand perception are among the most top of mind elements for users with a frequency occurrence of 25%, 24% and 14% respectively.

When we link this back to the theory it is noteworthy to mention that the driver “Price” scored very low in this study while several other studies have ranked this as one of the most important drivers of Value Equity (Rust et.al, 2004; Zhang, et.al, 2014; Ramaseshan, et.al, 2013). One reason for this discrepancy can be that the cloud based storage services are mostly based on freemium models in which users can freely access the basic functions of the application. Price only becomes a point of consideration when a user is willing to pay for the premium part of the services. However, most of the users are non-paid users. Thus, it can be argued that price is less top of mind within the users of cloud services than, for example, a traveler who seeks to make use of an airline flight service which was researched by Rust et al. (2004). Another driver that is prominent in other studies but ranked low in this current study was “Relationship”. Several authors have used this driver in their research and often

Drives of Customer Equity in the US and Dutch Cloud industry and the mediating role of loyalty intentions