HOW BANKS CAN IMPROVE CUSTOMER
RELATIONSHIP QUALITY THROUGH AN
OMNI-CHANNEL SERVICE OFFERING
Author:
Philip Ladiges
Student number:
10221433
Supervisor:
Dr. Shameek Sinha
Date of submission:
21.06.2018
MSc. Business Administration – Digital Business Track
University of Amsterdam
Statement of originality
This document is written by Student Philip Ladiges who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the content.
Abstract
Technological development has caused a shift in the way consumers use banking services. With mobile and online banking being widely adopted, customer preferences have changed, such that people now expect to be able to access their bank whenever they want, wherever they want. Banks have had to move their service offering away from the physical branch, into the digital realm. This shift introduced a different relationship between banks and their customers. This study has researched whether banks can use an omni-channel service offering to build stronger relationships with their customers, based on satisfaction, trust and commitment. Additionally, research was done into how the level of channel integration influences the performance of banks. Its main findings are that a better omni-channel service offering indeed leads to higher satisfaction, trust and commitment, and better customer relationship quality as a result of that. This effect is not influenced by the type of bank, or by recent bank branch closure experienced by customers. It is likely that banks also achieve higher transactions, traffic and customer base when the quality of omni-channel service offering is higher. This study serves as clear proof to bank managers that succesfully implementing an omni-channel strategy leads to improved customer relationship quality.
Table of Contents
STATEMENT OF ORIGINALITY 2 ABSTRACT 3 1 INTRODUCTION 6 1.1 BACKGROUND 6 1.2 RESEARCH OBJECTIVE AND RESEARCH QUESTION 7 1.3 CONTRIBUTION AND STRUCTURE 8 2 LITERATURE REVIEW 9 2.1.1 OMNI-CHANNEL BANKING DEFINITION 9 2.1.2 ADVANTAGES AND DISADVANTAGES OF AN OMNI-CHANNEL APPROACH 11 2.2 CUSTOMER RELATIONSHIP QUALITY 14 2.2.1 SATISFACTION 15 2.2.2 TRUST 17 2.2.3 COMMITMENT 20 2.3 BRANCH CLOSURE AND RELATIONSHIP QUALITY 22 2.3.1 BRANCH CLOSURES AND TRUST 22 2.3.2 BRANCH CLOSURES AND SATISFACTION 23 2.3.3 BRANCH CLOSURES AND COMMITMENT 24 2.4 BANK PERFORMANCE 25 3 METHODOLOGY 26 3. 1 RESEARCH DESIGN 273.2 SAMPLE & DATA COLLECTION 27
3.3 VARIABLES 29 3.3.1 CONTROL VARIABLES 29 3.3.2 DEPENDENT VARIABLES 30 3.3.3 INDEPENDENT VARIABLE 31 3.3.4 MODERATOR VARIABLE 33 4 RESULTS & ANALYSIS 34 4.1 ANALYTICAL STRATEGY 34 4.1.1 DESCRIPTIVE STATISTICS 34 4.1.2 RELIABILITY, NORMALITY AND VALIDITY 35 4.2 ANALYSIS 36 4.2.1 RELATIONSHIP QUALITY (H1) AND COMPONENTS (H2, H3.1, H4.1) 37 4.2.2 INTEGRITY (H3.1.2), COMPETENCE (H3.2.2) AND BENEVOLENCE (H3.2.3) 38 4.2.3 AFFECTIVE COMMITMENT (H4.1.1) AND CALCULATIVE COMMITMENT (H4.1.2) 38 4.2.4 BRANCH CLOSURE, TRUST(H5) AND SATISFACTION (H6) 39 4.2.5 BANK TYPE (H2.2 & H3.2) & BRANCH CLOSURE (5.1 & 6.1) MODERATORS 39
4.2.6 OMNI-CHANNEL AND BANK PERFORMANCE 40
4.3 OVERVIEW OF HYPOTHESES 41
5 DISCUSSION AND CONCLUSION 42
5.1 OMNI-CHANNEL SERVICE OFFERING AND CUSTOMER RELATIONSHIP QUALITY 43
6 MANAGERIAL IMPLICATIONS 47 7 LIMITATIONS AND FURTHER RESEARCH 48 REFERENCES 50 APPENDIX 53 A – SURVEY 1 53 B – SURVEY 2 57 C – NORMALITY CHECKS 62 62
1 Introduction
1.1 Background Over the last few years, the banking sector has been disrupted by a digital revolution, causing banks to increasingly move their service offering into the digital realm (Abishek, Geng, Zhou & Li, 2017). Digital innovation has transformed products, services and business models in the financial sector, which has led banks to increasingly adopt an omni-channel strategy. This allows customers to experience banking services as a unified whole, instead of through one single channel with limited service capabilities. This new strategy offers great opportunities for banks to reduce operational costs, and improve efficiency ratios (Liu, Abhishek & Li, 2017). One of the areas where the impact of the digital revolution is most notable is in the dramatic rise of popularity of online banking. Customers that are served through digital channels are generally more profitable for banks than traditional customers who interact and transact through physical branches. Research done by CEB TowerGroup (2013) shows that the average transaction costs of banking done through branches is 20 times higher than that of mobile banking, and 40 times higher than that of online banking. This has caused the leading banks in the United States to start scaling down their branch networks in recent years (Abhishek et al., 2017).Considering how profitable digital customers are to banks, it seems logical that online-only banks, which are not burdened by expensive physical branches, and have their operations built around big data integration, would have a significant competitive advantage over traditional banks. These new banks have, however, generally not yet been able to be profitable, as exemplified by British online Bank Monzo, which loses £50 per new customer it
attracts (Rumney, 2017), or the Dutch online bank Bunq, which reported millions of Euros in losses over 2016 (Fors verlies voor nieuwe bank Bunq, 2017).
Although research shows that customers appreciate the convenience of online banking, an increasing focus on digital interaction also means that the extent of personal contact with bank staff decreases, and with it, so do switching costs and ultimately long-term customer commitment (Yap, et al., 2010). This introduces a dilemma for banks: their physical presence is required to build trusted, long-term relationships with their customers, but they are simultaneously unable to afford expensive branches, without losing market share to low-cost, specialized Fintech competitors. Yap, et al. (2010) claim that this dilemma only exists because traditional and online banking are often seen as mutually exclusive or substitutable. Through the proper integration of online and offline channels, banks should be able to add value and improve the quality of the relationships with their customers (Brun and Rajaobelina, 2016). Furthermore, the retail sector has already shown that an omni-channel approach can lead to synergy effects on profitability and relationship quality. 1.2 Research objective and research question Current academic literature, has not fully investigated the opportunity that banks have to optimize customer relationship quality through an omni-channel approach. Some literature has, however, focused on how specific channels impact customer relationships. For example, Arcand, Promtep, Brun & Rajaobelina (2017) have studied the relationship between mobile banking service quality and relationship marketing. To date, the academic literature has not clarified the relationship between omni-channel service offering and customer relationship optimization. This study aims to address this gap. It investigates how factors leading to an improvement in relationship quality are impacted by a bank’s omni-channel service offering,
and whether the introduction of new technology and channels may lead to positive synergetic effects for banking customers. This leads to the following research question: “How does an omni-channel service offering in the banking industry influence the strength of customer relationships?” 1.3 Contribution and structure This paper contributes to the existing literature on customer relationships in omni-channel banking services. It has managerial value on account of the major shift that is currently taking place from offline channels to online channels. With a deeper understanding of the factors that influence customer relationship quality, managers will better be able to design their service offering in a way that maximum value can be extracted as a result of improved customer relationships.
This paper is organized as follows. In section 2, an overview will be given of the existing literature related to omni-channel services and customer relationship quality. This will lead to a presentation of the hypotheses related to the research question, as well as the conceptual model. Section 3 describes the research method, data and variables used for the empirical analysis. The results of the research will be analysed in section 4. A discussion of the results will be presented in section 5. Section 6 serves as an overview of the managerial implications of this research. Finally, recommendations for future research will be given in
2 Literature Review
This literature review will function as an overview of the academic literature on the main factors influencing customer relationship quality, omni-channel banking, the introduction of new technology channels and the closing of branch channels. 2.1.1 Omni-channel banking definitionTechnological development has transformed the ways in which individuals conduct their banking. Traditionally, banks would serve customers through a single, physical channel. This has developed into multichannel banking, where customers access banking services through diverse touch points such as a branch, mobile, online or through an ATM. Banks often have independent databases per channel, leading to a varying customer experience with minimal consistency between channels (Kumar, 2017). When bank processes operate in silos, it is difficult to combine channels, and banks often encourage customers to simply use the least expensive channel (Ericsson, 2012). Recently, this paradigm has been shifting. With an increasing number of consumers adapted to digital interaction, and the penetration of smartphones becoming mainstream, the demand for financial services to be available 24/7 has risen significantly (Cuesta, Ruesta, Tuesta & Urbiola, 2015). Simultaneously, Fintech companies offering specialized, low-cost alternatives for banking services have increased in popularity. This has prompted the need for banks to develop a holistic strategy, integrating their services to deliver an optimal service experience. Ericsson (2012) argues that it is important that this holistic strategy, known as omni-channel banking is significantly different from the popular multi-channel strategy. Through omni-channel banking, customers can experience consistency across channels, and seamless access to banking products, whenever and wherever they are needed. As such,
customers can have the best of both physical, and virtual channels. Kotarba (2016) adds that a omni-channel service offering enables customers to begin a process in one channel and continue it in another, with all available information from all channels being integrated in real time.
To be able to make a clear distinction between multi-channel banking and omni-channel banking, the interdependencies between channels need to be clarified. Research into the substitutive and complementary patterns between digital and physical channels was first conducted in the e-commerce and retailing context (Liu, et al., 2017). The opening of a new bricks-and-mortar store can result in customers moving away from online purchasing, while the number of in-store purchases increases due to the ‘access cost’ reduction associated with the opening of a new physical store (Kumar, Mehra and Kumar, 2014). A multi-channel service offering typically involves customers basing the decision to substitute channels on a cost calculation. Liu, et al. (2017) have researched channel interdependencies in the context of financial services, specifically focusing on how mobile services impact omni-channel banking behaviour. They found that mobile channels complemented other digital channels such as tablets and PCs when being used to search for information, and also led to customers conducting more activities in these channels. Additionally, a higher branch density is found to have a positive impact on digital service consumption, meaning that customers use omni-channel banking services complementarily, instead of solely relying on digital or physical banking channels. In short, when customers experience a multi-channel service offering, they substitute between channels, while channels are being used complementarily in an omni-channel service offering.
Kumar (2017) exemplifies omni-channel banking by describing a customer who starts a banking transaction in one channel, views it in another, and finishes it in a third channel,
made possible by the convergence of virtual and physical channels. As different channels use the same database, the customer is able to select their preferred channel.
Research has been conducted by Parise et al. (2017) into how companies with an omni-channel service offering can best leverage technology to improve the customer experience and strengthen customer relationships. They maintain that companies need to stop thinking only in terms of online versus offline channels, and should instead focus on integrating the advantages of both channels into an optimized omni-channel customer experience. This customer experience is made up of a diverse set of touchpoints, in which companies need to provide immediate, personalized content in order to be successful. Although this research was done in the context of the retailing sector, this study assumes that the concept is similar in the banking sector. Most literature regarding omni-channel integration and preferences focuses on product-centric markets, where certain retail specific factors such as product fit and product look-and-feel are important to consider. This is different for the banking sector, which is primarily service driven. This study will use the definition provided by Kotarba (2016), who has investigated the impact of an omni-channel service offering on customer relationship management (CRM), highlighting how the approach of Fintech firms differs to that of traditional banks. In this study, the focus will be on the influence of an omni-channel approach on specific elements of relationship quality. 2.1.2 Advantages and disadvantages of an omni-channel approach
When banking services were still performed through a single channel, customers relied completely on a bank’s branch network. This was a highly personal, but expensive operation for the bank. With the introduction of self-service, multi-channel service offerings, banks
were able to reduce transaction costs, while increasing customer profitability (Liu, et al., 2017). Research by PWC (2012) shows that the average transaction cost of a branch transaction is $4.00 for a bank, while the same transaction done digitally costs the bank only $0.09. Such cost savings seem to be an obvious driver for banks to stimulate digital transactions among their customers. This assumption is in line with transaction cost theory, which states that organizations are comparative beings, always examining processes in relation to other feasible forms. This theory examines the overall structure of an organization and states that firms make decisions primarily on the basis of efficiency. As such, according to transaction cost theory it is only logical for a bank to switch to an omni-channel service offering, when this leads to efficiency and costs advantages (Williamson, 1981, 1996). While customer preferences have generally been shifting towards digital transactions, this is not the case for all types of transactions, as most customers still prefer going to a bank branch for investment transactions, mortgages and loans (Graupner, Melcher, Demers and Maedche, 2015). This is confirmed by Nam, Lee and Lee (2016), who investigated the impact of the shift towards a digital service on the user experience. They state that, although the number of digital banking users is rising steadily, customer satisfaction and loyalty is dropping due to the lack of touch and interaction associated with self-service channels. Conversely, Nam et al. (2016) found that a shift from traditional branches to digital channels has a significant positive effect on a bank’s profitability.
With the introduction of an omni-channel strategy, customers should ideally be able to experience all benefits of all banking channels in an optimal way. Customers increasingly become part of the service delivery process, and receive time-and location-independent services. Such a service offering would improve efficiency in customer interactions by shifting from individual interaction points, to one consistent holistic interaction. (Nüesch, Alt &
Puschmann, 2015). According to Hossain, Akter, Kattiyapornpong & Wamba (2017, p. 786), one of the more important factors in customer perception of omni-channel service quality is integration quality. This is defined as “the ability to provide customers with a seamless experience across multiple channels”. When integration quality is high, this leads to higher service quality, which in turns increases factors such as word of mouth (WOM) intention and trust. Although Hossain et al. (2017) have researched this phenomenon specifically in the retail sector and not in a banking context, this study will adopt the same definition, and use integration quality as a marker of omni-channel quality. Herhausen, Binder, Schoegel & Herrmann (2015) have also researched channel integration in the retail sector. They add that online-offline channel integration quality can be judged on the basis of the degree to which different channels interact with each other. Additionally, Accenture (2016) has researched omni-channel integration quality in their annual retail banking customer survey by asking banking customers whether their experience their bank’s branch, online and mobile channels is a seamless one.
Nüesch, et al. (2015) have researched omni-channel banking from a business management perspective. They state that banks need to reshape their organizational structure through a customer-driven structure of value chains, such that optimal channel convergence is achieved. This will enable seamless customer interaction through a maximum number of channels and technologies. Increasingly closing bank branches, the traditionally preferred banking channel, impacts upon service delivery and customer relationships. This impact will be discussed in greater detail in the following chapters.
2.2 Customer relationship quality
The amount of digital banking options available to consumers has grown immensely in a relatively short period. While the transition of banking services into the digital realm has introduced many benefits to both consumers and banks, it has also diminished the physical and social proximity that banks have long relied on. This has caused the relationships that banks have with their customers to dehumanize, which threatens the quality of the existing relationships (Brun, Rajaobelina & Ricard, 2014). In order to engage and retain their online customers, banks need to optimize perceived service quality, which can be done by building strong relationships (Brun & Rajaobelina, 2016). Much research has been done into relationship quality in the traditional sense, which can be defined as “a meta-construct composed of several key components reflecting the overall nature of relationships between companies and consumers” (Hennig-Thurau, Gwinner & Gremler, 2002, p.234). Hennig-Thurau et al. (2002) state that in order to get a better understanding of the drivers and outcomes of customer relations, relationship marketing needs to identify the key factors that impact the outcome of the relationship between a firm and a customer. They found that the two main dimensions of relationship quality are commitment and satisfaction. This is supported by Palmatier (2008), who researched the relational drivers of customer value. However, he adds a third factor, and claims that commitment is another key factor in determining relationship quality. When related specifically to the customer value that is extracted from a relationship, satisfaction is also seen as a key factor. Both Palmatier (2008) and Hennig-Thurau et al. (2002) have researched customer relationship quality in a traditional sense, in a time when omni-channel did not yet exist. There are multiple studies that confirm that satisfaction, trust and commitment are still important in an online environment (Chung and Shin, 2010; Rafiq et al., 2012; Walsh et al.,
2010 in Brun, et al., 2014). Within their research into the key factors of online relationships, these studies have confirmed the importance of satisfaction, trust and commitment. The impact of these factors has primarily been studied separately, yet Brun et al. (2014) claim that online relationship quality needs to be seen as an integrated, multidimensional construct. They have introduced a scale to measure online relationship quality, which confirms that trust, commitment and satisfaction are also the key dimensions in online relationship quality. Based on the advantages of an omni-channel integration described in existing literature, the following hypothesis was formed:
Hypothesis 1: An omni-channel service offering has a direct positive effect on customer relationship quality. 2.2.1 Satisfaction Satisfaction is seen as the main component in explaining relationships between two parties. It is often described as the result of a comparison between actual performance, and expected performance (Brun, et al., 2016). As such, satisfaction is described as backword looking, and as an assessment of overall consumption experience. It can be defined as “the extent to which benefits actually received, meet or exceed the perceived equitable level of benefits” (Brun, Rajaobelina & Ricard, 2016, p 223). Brun, Rajaobelina, Ricard & Fortin (2017) further define this definition in the context of banking websites by saying that satisfaction is an evaluation that is cumulative over time, and comprehensive (existing of total purchase and consumption experience). Sampaio, Ladeira & Santini (2017) have studied the role of satisfaction specifically in relation to mobile banking apps. They found that satisfaction is closely related to trust, and that satisfaction leads to loyalty and positive WOM. When customers use mobile banking
through an app as a channel, this leads to benefits that are positively related to satisfaction. This means that when customers use mobile banking apps to enhance satisfaction, this is a predictor of trust, loyalty and WOM. In addition, Amin (2016) states that especially satisfaction is an important factor in building and maintaining customer relationships. It also reduces the perceived risk of online banking. This is particularly relevant for this study, as this would be a main factor of stronger relationships in an omni-channel environment, which does not revolve around one particular channel.
One of the first studies conducted specifically into the determinants of customer satisfaction in retail banking was done by Levesque & McDougall (1996). They found that there are many different variables that explain customer satisfaction. Some of the most relevant ones are service quality dimensions, such as core and relational performance, problems encountered and satisfaction with problem recovery. Although, of course, banks did not yet offer an omni-channel service offering in 1996, the way that these constructs were studied leads to believe that these variables are still relevant. In their research, participants were requested to judge statements such as ‘my bank performs the service right the first time’, ‘employees in my bank have the required skills and knowledge to perform the service’, and ‘considering everything, I am extremely satisfied with my bank’. It seems likely that a customer’s satisfaction with their bank can still be judged through similar statements.
Ayo, Oni, Adewoye and Eweoya (2016) researched customer satisfaction with e-banking. They also found (e-)service quality to be an important factor in influencing customer satisfaction. Other variables influencing satisfaction in an online banking environment are the range of products and services offered, and the ability to personally interact with an e-banking system. Service quality is defined by Ayo et al. (2016, p. 351) as “a long-term judgement regarding an organization’s excellence or superiority”. In addition, Amin (2015, p.282) states
that customers perceive e-services differently than traditional services. E-satisfaction is, for example, also driven by website characteristics. He defines e-service quality however as “a consumer’s overall evaluation and judgement on the quality of the services that is delivered through the internet”. Because of the relative overlap in the two definitions, the first more general definition will be used to define both for this study. The relevance of the constructs described above, is generally supported by Liébana-Cabanillas, Munoz-Leiva and Rejón-Guardia (2013), who investigated determinants of customer satisfaction with e-banking. They found that in addition to service quality, primarily accessibility, trust, ease of use and usefulness impact customer satisfaction with e-banking. Accessibility in this context is seen as the ease of access to e-banking applications. Ease of use is generally defined as the perception that a system is free of effort, while usefulness is defined as “the probability that using a system will enhance job performance in an organisation context” (Liébana-Cabanillas, et al., 2013, p. 756). On the base of the different contexts in which previous research has been conducted, this paper will use (e-)service quality, accessibility, ease of use and usefulness as the main constructs of satisfaction. Based on the literature, the following hypothesis was formed: Hypothesis 2: An omni-channel service offering has a direct positive effect on satisfaction. 2.2.2 Trust Among the many factors that are relevant for success in banking, trust is considered one of the most important ones for online banking continuance; a lack of customer trust is seen as a potential major obstacle for widespread acceptance of online banking (Yu, et al. 2015). Trust establishes the belief in customers that banking providers have the ability to provide useful services. Yousuf & Wahab (2017) have researched this specifically in the context of mobile
banking, and state that trust is a critical factor for the success of digital financial activities. This is because of the open and global nature of the internet, which can lead to uncertainty and risk when doing online transactions. There are many possible definitions of trust, depending on the context it is used in. Brun, et al (2017, p 142) studied trust in the context of online banking websites. They define trust as “the belief that one party’s word or promise is worthy of the trust of the other and that the party in question will fulfil his or her obligations”. They claim that there are three main facets of trust that need to be grasped, in order to fully understand the concept. These factors are competence (related to the skills required to perform required behaviour), integrity (related to honesty) and benevolence (related to honouring the needs and interests of related parties).
When discussing trust in banking, it is important to make the distinction between online banking (through digital channels), and offline banking (through physical branches or ATM’s). The experience that customers have with an online bank is often perceived as more distant and impersonal when compared to an offline experience with a traditional bank. When the switch is made from offline to online, customers perceive personal contact and their own control of the situation to be lower, which reduces trust in online banking compared to offline banking (Yu, Balaji & Khong, 2015). Parise et al. (2016) have shown, however, that it is also possible to build trust by simulating an employee’s physical presence in an omni-channel environment by adding video services. This is supported by Ericsson et al. (2012), who state that video is a key enabler to build trust in a situation where there is no human physically present. This indicates that a proper integration of channels and services can improve trust.
Yu, et al. (2015) have researched how online-only banks can best rebuild the trust that is lost with the switch to online banking. They found that belief in competence, integrity and shared values are the key factors in increasing trustworthiness and trust, such that it increases a customer’s intention to keep using online banking services. Based on the existing literature on trust and banking channels, the following hypotheses have been formed: Hypothesis 3.1: An omni-channel service offering has a direct positive effect on trust. The following sub-hypotheses are based on the three main components of trust: Hypothesis 3.1.1: An omni-channel service offering has a direct positive effect on perceived competence. Hypothesis 3.1.2: An omni-channel service offering has a direct positive effect on perceived integrity. Hypothesis 3.1.3: An omni-channel service offering has a direct positive effect on perceived benevolence. Because online-only banks do not have branches, they cannot offer their customers the same types of channels as traditional banks, nor can they make use of reputation advantages that traditional banks have because of their existing branches. Liu, et al. (2017) add to this and state that customers use channels complimentarily in an omni-channel environment, and that branches play an important role in taking people from physical to digital banking. When both types of channels can be used, customers feel more secure and more diverse digital transactions take place. This leads to believe that the absence of bank branches influences the effect of an omni-channel service offering. Based on this, the following hypothesis is made:
Hypothesis 3.2: The positive relationship between a bank’s omni-channel service offering, and improved trust, is moderated by the type of bank, so that this relationship is weaker for online-only banks. 2.2.3 Commitment Commitment in respect to relationship quality is defined by Palmatier (2008, p77) as “a desire to maintain a valued relationship, and thus, an exchange partner’s relationship motivation toward a partner.” Palmatier (2008) researched this in a traditional marketing perspective, rather than in the context of omni-channel banking. Brun, et al. (2016) adopt the same definition when researching a model on online relationship quality in the banking industry. They extend the definition of commitment by stating that it ensures relationship continuity over time, and that it is equally important in both traditional relationships as online relationships. Shaikh, Karjaluoto & Chinje (2015) researched relationship commitment specifically in the context of mobile banking usage. They add to the definition that commitment is a key variable in relationship marketing, which keeps customers loyal, even if they are unsatisfied, or if there is a competitive offering.
Relationship marketing research identifies three different types of commitment: affective commitment, calculative commitment (also known as continuance commitment) and normative commitment (Shaikh, et al., 2015). Affective commitment is defined as “a positive emotional attachment or psychological bond”. It implies that customers identify with, and like an organization. Calculative commitment is more dependency based, linked to phenomena such as switching costs. Finally, normative commitment arises from a feeling of obligation (Brun, et al., 2016). Some of the main drivers of commitment are found to be
having shared values, attraction of the services, and perceived user empowerment (Shaikh, et al., 2015).
Affective and calculative commitment are seen as having the most significant impact on customer loyalty and retention in online banking, and for this reason only these two types will be relevant for this study. Based on the literature the following hypothesis has been formed: Hypothesis 4.1: An omni-channel service offering has a direct positive effect on commitment. The following sub-hypotheses were formed based on the two relevant components of trust: Hypothesis 4.1.1: An omni-channel service offering has a direct positive effect on affective commitment. Hypothesis 4.1.2: An omni-channel service offering has a direct positive effect on calculative commitment.
The relationship between the use of online channels and customer loyalty has been researched by Levy (2014). This was researched for banking consumers who use both offline and online banking. He found a strong negative relationship between the use of online services, and customer relationship strength, specifically due to decreasing commitment and loyalty. The main reason for this negative relationship was the lack of personal, human mediation in the case of service failure or dissatisfaction. This leads to believe that commitment is lower for customers of online-only banks, as all their interactions are through online services. Based on this, the following hypothesis was formed:
Hypothesis 4.2: The positive relationship between a bank’s omni-channel service offering, and improved commitment, is moderated by the type of bank, so that this relationship is weaker for online-only banks.
2.3 Branch closure and relationship quality
As discussed above, it is clear that there are many efficiency and cost reduction benefits for banks when their customers switch to digital banking. Banks are comparative beings that evaluate processes in relation to other feasible forms of doing business. This has been the basis of the decision by banks to increasingly scale back on their branch and ATM networks. An example of the steady decline in bank branches can be seen in
figure 1 (The closing of American bank branches, 2017). The impact of branch closures on customer relationships can be significant, especially in local communities, The Economist states. Small businesses in local communities often rely heavily on personal relationships within a bank branch. When these branches close, it can lead to an estimated 13% decrease in small-business lending, and even a 40% decrease in low-income neighbourhoods. Overall, Accenture (2016) shows that the banking branch is still a popular channel, with 20% of customers visiting their branch at least once a week. This raises the question as to whether banks can find ways to integrate interaction and touch into digital channels, such that existing customer relations are not harmed. 2.3.1 Branch closures and trust When customers have personal interactions during business interactions, this instils a sense of security, which is one of the main advantages of bank branches in relation to trust. This belief is enforced by a large-scale survey amongst retail banking customers done by Accenture Figure 1: Branch closures in the US (The Economist)
(2016), showing that the main reason people use bank branches is because they trust the bank more when speaking to someone in person at a banking branch. When a branch closes, and customers are forced to switch to digital channels, the perceived control and personal contact are reduced, which causes trust to decrease (Yu, Balaji & Khong, 2015). Parise et al. (2017) state that companies can create feelings of personal contact, by interacting digitally with their customers through augmented reality and video communication. This would connect a customer with an employee in real time, which could stimulate the feeling of interaction. These studies don’t mention whether the loss of trust due to a decrease in personal interaction can be reversed completely by simulated personal interaction. Because previous studies mention a decrease in trust repeatedly, this study will assume that this is not the case, and that perceived branch closure had a negative impact on the positive effect that integrating digital services has on trust. On the basis of this previous research, the following the hypotheses are formed: Hypothesis 5: An increase in bank branch closures, leads to a decrease in trust. Hypothesis 5.1: The positive relationship between omni-channel service offering and trust is moderated by perceived branch closures, so that this relationship is weaker when branch closures are higher.
2.3.2 Branch closures and satisfaction
When banks scale down on their branch networks, this naturally means that consumers who previously used these branches have fewer channels to choose from, making a service offering less complete. Determinants of customer satisfaction with banking, such as service quality dimensions, problem solving and support response are all influenced by branch closures. When customers switch to digital channels, oftentimes other determinants become
important for a customer’s perception of service quality (Amin, 2015). Current academic research does not explicitly explain whether the negative effect of branch closures on customer satisfaction is greater than the positive effect of omni-channel integration on customer satisfaction. The far-reaching impact of branch closures has however been researched by Nguyen (2014), who found that closings have a large impact on credit supply and lender-specific relations, especially in low-income areas. The following hypotheses are formed on the basis of previous research:
Hypothesis 6: An increase in bank branch closures, has a direct negative effect on satisfaction. Hypothesis 6.1: The positive relationship between omni-channel service offering and satisfaction is moderated by perceived branch closures, so that this relationship is weaker when branch closures are higher
2.3.3 Branch closures and commitment
There is a lack of clear academic evidence regarding commitment and branch closures. Research by Accenture (2015) has shown that 19% of responding banking customers claim to switch banks if their local bank branch closes. This number has fallen significantly compared to similar research by Accenture done in 2013, when 48% claimed that they would switch if their local branch were to close. With the rapid development and the continuing rise of customer acceptance of digital channels, there is no reason to believe that this number will not have dropped significantly in 2018 (though more recent Accenture research is not available). Sufficient research is lacking to make claims about the relation between an omni-channel service offering, customer relationship commitment and branch closures.
2.4 Bank performance
Banks have been described as comparative beings that act based on transaction cost considerations (Williamson, 1981, 1996). If a switch to an omni-channel service offering leads to increased bank performance, this would logically mean that banks are likely to make the switch. Liu, et al. (2017) found that transactions and traffic have increased for banks after mobile banking has been integrated in a bank’s channel portfolio. Additionally, they found that an omni-channel integration creates efficiency advantages when dealing with customers. This is confirmed by a case study performed by Inetco (2015) into a leading African Bank, UBA. They found that an omni-channel integration leads to improved call resolution rates, and a more streamlined troubleshooting process. Based on these findings, it is likely that service waiting time decreases when banks adopt an omni-channel service offering. Finally, using the balanced scorecard presented by Ittner, Larcker & Meyer (2003), it is determined that customer base growth is one of the main objectives and performance measures of banks. Based on the description presented in previous sections, it is likely that customer relationship quality is higher when banks offer omni-channel services. Because of the extra benefits that customers experience, this study assumes that an omni-channel integration will lead to a growing customer base. The following hypotheses have been formed:
Hypothesis 7.1: An omni-channel service offering has a direct positive effect on bank channel traffic.
Hypothesis 7.2 An omni-channel service offering has a direct positive effect on a bank’s transaction amount.
Hypothesis 7.3 An omni-channel service offering has a direct negative effect on a bank’s service waiting time.
Hypothesis 7.4: An omni-channel service offering has a direct positive effect on a bank’s customer base. All hypotheses are summarized in the following conceptual model:
3 Methodology
The following part of this thesis study will outline how the research is designed, and how the choice of methodology was motivated based around the topic of an omni-channel service offering in the banking industry. The research design will be explained in section 3.1. In section 3.2 an explanation will be given of the sample and data collection methods that were used. Finally, section 3.3 will present an overview of variables that were investigated.3. 1 Research design
The data for this quantitative research was collected through two cross-sectional surveys. These surveys were administered as web-based self-completion questionnaires (Appendix A & B). The first questionnaire was distributed among banking customers, the second was distributed among banking employees and managers. To ensure proper representation of the moderator variables, the surveys will be distributed to diverse banking customers, and banking employees and managers. The hypotheses that will be tested in this study were formed using a deductive approach, as they are formed based on existing literature. 3.2 Sample & Data Collection The population for the first survey consists of all Dutch banking consumers. As this a large group, the exact sampling frame was difficult to define, however only consumers who are equipped to autonomously handle their own bank account were approached. As such, this research only includes people who are at least 18 years old, and are officially of legal capacity to make decisions regarding banking matters. The sampling frame of the first survey consisted all legal adults who make use of banking services. The population for the second survey consisted of banking employees and managers of Dutch retail banks. This was limited to only include full time employees. This study made use of non-probability convenience sampling. This approach was used because access to banking employees is difficult, and the population of banking customers is extremely large and diverse. Respondents for both surveys were targeted through e-mail, social media, LinkedIn, online bank support forums and bank branch visits. The researcher aimed to reach a demographically diverse sample, and to find as many respondents as possible. It was difficult to make an accurate prediction of the response rate,
as many different distribution techniques were combined, and the response rates from previous research that the scales are based on was not specified. In total, 240 participants filled in the survey for banking customers, while 35 people filled in the survey for bank employees and managers. Not all respondents filled in the survey completely however. All incomplete surveys were omitted from the research, which were a total of 27 respondents for the first survey, and 4 respondents for the second survey. This left n = 213 for the first survey, and n = 31 for the second survey. To facilitate descriptive statistics, the first survey started with questions related to the demographics of the respondents. The number of men and women was approximately normally distributed, and their age ranged from 18 to 71 with the largest group being between 18 and 24 years old. Most respondents finished a Bachelor’s degree, and spent 2-4 hours using the internet privately every day. The relevant descriptive statistics that can be inferred from the results from demographic questions are summarized below in table 1. To research potential differences between men and women’s age, level of education and time spent on the internet, Levene’s test was used. This showed that there were no significant differences variances between men and women related to age (F = 3.187, p = 0.864), education (F = 1.333, p = 0.192) and daily internet usage (F = 0.181, p = 0.193).
Measure Item Amount (N = 213) %
Gender Female 139 34.3 Male 73 56.3 Other 1 0.5 Age 18-24 90 42.3 25-34 81 38 35-44 12 5.6 44-54 5 2.4 54+ 25 11.7 Education Master’s degree or higher 72 33.7 Bachelor’s degree 103 48.8 High school 32 15
Other 6 2.8 Daily internet usage 0 – 30 minutes 8 3.8 30 minutes – 1 hour 20 9.4 1 – 2 hours 53 24.9 2 – 4 hours 81 38 4+ hours 51 23.9 An overview of the type of bank the respondents of the second survey worked for is shown below in table 2.
Measure Item Amount (N = 31) %
Bank Type Online and offline bank 19 61.3 Online-only bank 12 38.7 Offline-only bank 0 0 3.3 Variables This section describes the main variables that were researched. Specifically control variables, the independent variable, dependent variables, and the moderator variable will be discussed. 3.3.1 Control variables The following control variables were included in the first survey: 1. Gender, nominal scale (male, female or other). 2. Age, ratio data. 3. Level of education, ordinal data (ascending). 4. Frequency of private internet use, ratio data (ascending).
No questions were asked in the second survey about the respondents’ personal demographics, because this survey reflected their banks’ perspective and not their personal
Table 2: Descriptive statistics survey 2 Table 1: Descriptive statistics respondents survey 1
opinion. The following control variable and related follow-up questions were included in the second survey: 1. Type of bank, nominal scale (Both online and offline, online-only, or offline-only). a. Can the service offered in digital channels match the quality of the services offered in branches? b. Would the bank better be able to service customers when they also made use of physical branches? c. Would the bank better be able to service customers when they also made use of digital channels? 3.3.2 Dependent variables The dependent variables related to customer relationship quality described in this research are trust, commitment and satisfaction. These variables were measured using 5-point Likert scales, validated by Brun, Rajaobelina & Ricard (2014). Ten items are used to measure the components of trust, of which the first four measure competence, the next four measure integrity and the final two measure benevolence. Five items are used to measure commitment, the first two measure affective commitment specifically, and the final three measure calculative commitment. Additionally, two more items to measure commitment are used from a verified scale by Ganesan, Shankar, Brown & Mariadoss (2010). Four items are used to measure satisfaction. The scale used by Brun et al. (2014) shows great reliability in measuring trust with Cronbach’s Alpha being 0.94, satisfaction (α: 0.93), calculative commitment (α: 0.81), and affective commitment (α: 0.95). The items were worded slightly differently than in the verified scales they were retrieved from, to ensure a fit better with the research topic.
Multiple measures of performance were included in the second survey. Bank performance was measured through general questions related to traffic, customer base, deposits, transaction volume and customer service waiting time. The aim of these questions was to research whether scoring higher on omni-channel service offering has a positive effect on performance related metrics. The results of both surveys could then be compared. A full overview of the survey questions from both surveys can be found in Appendix A & B. 3.3.3 Independent variable The independent variable used in the first survey, is the perceived level of a respondent’s bank’s omni-channel service offering. To research channel usage and its effect amongst consumers, questions were asked in the first survey about which banking channels consumers use, and with which frequency, as well as whether they feel that these channels are integrated well together. Because the perceived quality of channel integration is a complex construct, that is difficult to measure, the evaluation of integration quality was conceptualized using the adequacy-importance model. This model was used by Schramm-Klein, Wagner, Steinmann & Morschett (2011), when researching cross-channel integration in a retailing context. In this model’s two-component perspective, separate data is used to first measure the evaluation of relevant channel integration factors, then the importance of these functions is evaluated. These two factors together are seen as an appropriate evaluation of total integration quality, which is calculated by using the mean of weighting factors, multiplied by the corresponding expression value (Schramm-Klein, et al., 2011). The questions were altered slightly to fit an omni-channel banking context, while irrelevant questions were left out, and others were added to reflect the topics previous research found
most import in defining an omni-channel service offering. These dimensions are shown below in Table 3.
Dimension ‘Integration’ Indicator ‘Importance’ Indicator
Channel usage Banking channels can be used
interchangeably. Option of using banking channels interchangeably. Channel switching -
information There is no difference in available information about me when switching channels.
Option of having no difference in available when switching channels.
Seamless experience The experience when switching between bank channels is seamless.
Option of having a seamless experience between bank channels.
Banking product
information Banking product information is available in all channels. Option of having banking product information available in all channels.
Pricing information Price information about banking products is available in all channels.
Option of having price information available in all channels.
Communication The quality of communication is equal among all bank channels used.
Option of having the same communication quality in all channels.
The second survey also made use of the adequacy-important model to measure bank employee and manager perception of their bank’s omni-channel service offering quality. The dimensions of the questions are summarized below in table 4. The ‘integration’ indicator asked respondents on their opinion on how integrated certain aspects of their channels were for their customers. The ‘importance’ indicator asked respondents about how important they felt each dimension was to their bank. Similar to the method used for survey 1, integration quality was calculated by using the mean of weighting factors, multiplied by the corresponding expression value. To be able to compare this with the other questions about omni-channel integration that did not make use of the adequacy-importance model, Z-scores were computed for all questions related to omni-channel banking.
Table 3: Omni-channel integration quality, customer perspective.
Dimension ‘Integration’ Indicator ‘Importance’ Indicator
Information disparity There is no disparity in available information when switching between channels.
Option of having no disparity in available information between channels.
Seamless switching Switching between channels is
a seamless experience. Option of being able to seamlessly switch between channels.
Interchangeable use of
channels Channels interchangeably. can be used Option of being able to use channels interchangeably. Complementing
channels A bank’s different channels complement each other. Option of having channels complement each other.
Banking product
information Banking product information is available in all channels. Option of having banking product information available in all channels.
Pricing information Price information about banking products is available in all channels.
Option of having price information available in all channels.
Communication quality Quality of communication is
equal among all channels Option of having equal quality of communication on all channels.
3.3.4 Moderator variable
As discussed in section 2, hypotheses have been formed based on prior research about the moderating role of the amount of branch closures, and the type of bank someone uses. Whether respondents were influenced by branch closures was researched by asking them if they had noticed bank branches closing around them recently. This was measured by using a 5-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’. Type of bank was measured by asking respondents whether their bank serviced them through both online and offline channels, or exclusively through online channels. Table 4: Omni-channel integration quality, bank perspective
4 Results & analysis
This chapter presents an overview of the results from the research that was done into the hypotheses that were formed. The structure is as follows: section 4.1 gives an overview of the analytical strategy, including an overview of descriptive statistics. Section 4.2 will present an explanation of the analytical tests that were performed per hypothesis. Finally, an overview of whether each hypothesis is supported or not is shown in section 4.3. 4.1 Analytical strategy To ensure that the data set is complete, the researched variables were checked for missing data. This was done by performing a frequency test, which determined that all responses to the variables under investigation were complete. Several variables that were based off reverse-scored question were recoded before they were analysed. Dummy variables were created for several control variables, and interaction variables were created before analysing the moderator variables. 4.1.1 Descriptive statistics The descriptive statistics of the main variables that were used in the analyses are summarized below in table 5. These include all dependent, independent, moderating and control variables. Furthermore, a correlation matrix has been computed to test for potential significant correlations between the control variables and the dependent variables. One significant correlation was found when comparing the control variable education with satisfaction (r=0.179, p = 0.009). This result was taken into account when testing the hypotheses related to satisfaction.Group N Mean Std. Deviation SatisfactionTOT 213 4.0164 0.77457 CommitmentTOT 213 3.2220 0.69866 OmniWeighted 213 12.9845 3.79619 TrustTOT 213 4.0793 0.63108 CompetenceTOT 213 4.2453 0.65829 IntegrityTOT 213 4.0246 0.75644 BenevolenceTOT 213 3.8561 0.84276 BankType 213 0.1934 0.39590 BranchClosure 172 2.8140 1.29788 Age 213 30.79 13.468 Gender 213 0.6651 0.48299 Education 213 0.8726 0.76524 Daily Internet Use 213 3.70 1.050 A table of the descriptive statistics related to the second survey can be found below. This includes the mean and standard deviation related to the computed Z-scores of the independent variable, and the dependent variables.
Group N Mean Std. Deviation
OmniWeighted 31 0.000 0.55878 BankType 31 0.3871 0.49541 Traffic 29 3.5517 1.18280 Transaction Volume 28 3.5000 1.10554 Customer base 28 3.5000 1.13855 Service waiting time 27 3.1111 0.84632 4.1.2 Reliability, normality and validity The scale reliability for all items related to trust, commitment and satisfaction has already been measured by Brun, Rajaobelina & Ricard (2014), and the corresponding Cronbach’s Alpha values have been reported in section 3. To measure the reliability for omni-channel related variables, the items first needed to be computed according to the
adequacy-Table 5: Descriptive statistics variables survey 1
Table 6: Descriptive statistics variables survey 2
importance model that was used to measure integration quality. This was done by first multiplying the ‘integration indicator’ of each question with the related ‘importance indicator’, resulting in 6 new items (Omni1xOmni2, Omni3xOmni4, etc.) After this was done for each question, a weighted average of all questions was computed, and named ‘OmniGewogen’. Cronbach’s alpha was computed for the new items, which were reliable with a score of 0.754.
To test the normality of satisfaction, trust, commitment, competence, integrity, benevolence, bank traffic, bank transactions and banking customer base, histograms and normal p-plots were computed for each of these variables. These show that all variables are approximately normally distributed. This is less clear for the variables related to the banking side, as the number of respondents is much lower for these variables. An overview of all computed histograms and p-plots can be found in Appendix C. No separate validity tests were performed as all scales that were used, were validated by other authors when they created them. Only minor changes have been made to these, based on factors that were described as particularly relevant for each construct in previous research. 4.2 Analysis To test the hypotheses that were formed in this research, multiple tests were performed using SPSS Statistics software. Simple linear regressions were calculated to predict satisfaction, trust and commitment based on omni-channel service offering. The effects of these tests will be described below. After the results for the hypotheses related to these variables are described, the results for the moderator variables will be analysed and described.
4.2.1 Relationship quality (H1) and components (H2, H3.1, H4.1)
A regression analysis was performed to research the direct relationship between an omni-channel service offering, and the mean of all items related to customer relationship quality. A significant regression equation was found (F(1,211) = 50.782, p < 0.000 with an R2 of 0.194 and a β of 0.440). Respondent’s relationship quality is equal to 2.984 + 0.061 units of omni-channel service offering, when relationship quality is measured in units of omni-and a β of 0.440). Respondent’s relationship quality is equal to 2.984 + 0.061 units of omni-channel service offering. Respondent’s satisfaction increased 0.061 for each unit of omni-channel service offering
A significant regression equation was also found when inspecting the mean of items measuring satisfaction (F(1,211) = 41.918, p < 0.000) with an R2 of 0.166 and a β of 0.410. Respondent’s predicted satisfaction is equal to 2.938 + 0.083 units of omni-channel service offering, when satisfaction is measured in units of omni-channel service offering. Respondent’s satisfaction increased 0.083 for each unit of omni-channel service offering. This effect has been controlled for education. The impact of omni-channel service offering on satisfaction remains significant when education is removed.
A regression analysis was performed first for the all items measuring trust, and then separately for the three components parts of trust, namely integrity, competence and benevolence. The results for the analyses of the components parts of trust will be described below. A significant regression equation was found for the total 10 items measuring trust (F(1,211) = 33.418, p < 0.000) with an R2 of 0.137 and a β of 0.370. Respondent’s predicted trust is equal to 3.281 + 0.061 units of omni-channel service offering, when trust is measured in units of omni-channel service offering. Respondent’s trust increased 0.061 for each unit of omni-channel service offering.
Finally, the same linear regression model was used when inspecting the mean of items measuring commitment. A significant regression equation was found (F(1,211) = 16.250, p < 0.000 with an R2 of 0.072 and a β of 0.267). Respondent’s predicted commitment is equal to 2.583 + 0.049 units of omni-channel service offering, when commitment is measured in units of omni-channel service offering. Respondent’s trust increased 0.049 for each unit of omni-channel service offering. 4.2.2 Integrity (H3.1.2), competence (H3.2.2) and benevolence (H3.2.3)
Significant regression equations were found when inspecting perceived integrity, competence and benevolence (F(1,211) = 22.657, p < 0,000), (F(1,211) = 14.450, p < 0.000) and (F(1,210) = 47.858, p < 0.000) with R2’s of 0.097, 0.064 and 0.186, and β’s of 0.311, 0.253 and 0.431 respectively.
Respondent’s predicted perceived integrity is equal to 3.219 + 0.062 units of omni-channel service offering, predicted perceived competence is equal to 3.675 + 0.044, and predicted perceived benevolence is equal to 2.615 + 0.096. Perceived integrity, competence and benevolence increased with 0.062, 0.044 and 0.096 respectively for each unit of omni-channel service offering. 4.2.3 Affective commitment (H4.1.1) and calculative commitment (H4.1.2) Significant regression equations were found when inspecting the separate components of commitment, namely affective and calculative commitment (F(1,211) = 25.009, p < 0.000), (F(1,211) = 5.034, p < 0.0026) with R2’s of 0.106 and 0.023, and β’s of 0.326 and 0.153 respectively. Respondent’s predicted affective commitment is equal to 1.873 + 0.096, and predicted calculative commitment is equal to 2.867 + 0.031. Affective commitment and
calculative commitment increased with 0.096 and 0.031 respectively for each unit of omni-channel service offering.
4.2.4 Branch closure, trust(H5) and satisfaction (H6)
A linear regression analysis was performed to determine the direct effect of branch closure on respondent’s trust. A significant regression equation was found (F(1,170) = 2.014, p < 0.000) with an R2 of 0.012, and a β of -0.108. Respondent’s predicted trust is equal to 4.221 -
0.049 units of perceived branch closure, when trust is measured in units of perceived branch closures. Respondent’s trust decreased 0.049 for each unit of perceived branch closure. No significant regression equation was found between branch closure and satisfaction, with p(0.489) > 0.05.
4.2.5 Bank type (H2.2 & H3.2) & branch closure (5.1 & 6.1) moderators
A hierarchical linear regression was calculated to test whether the relationship between omni-channel service offering and trust, and omni-channel service offering and commitment was moderated by the type of bank a consumer uses. To test the potential impact on trust, in the first block of the regression analysis the direct impact of omni-channel service offering on trust was calculated, in the second block the direct impact of bank type on trust was calculated, and for the third block an interacting variable between omni-channel and bank type was computed, and its impact on trust was calculated. No significant regression equation was found between omni-channel, trust and bank type, with p(0.079)>0.05.
A similar hierarchical linear regression was calculated to test whether bank type moderated the relationship between omni-channel service offering and commitment. The direct impact of omni-channel service offering on commitment was calculated in the first
block, then the direct impact of bank type was calculated, and finally an interactive variable of omni-channel and bank type was added. There was no significant regression equation, with p(0.705)>0.05.
Interactive variables of omni-channel and branch closure were computed to measure the moderating effect of branch closures. In the first block of the hierarchical linear regressions that were performed the direct impact of omni-channel service offering on satisfaction and trust was calculated. Then the direct impact of branch closure was calculated, and finally the impact of the interacting variables was added. Neither of these regression equations were significant, with p(0.545)>0.05 for satisfaction, and p(0.729)>0.05 for trust.
4.2.6 Omni-channel and bank performance
To examine the relationship between the level of a bank’s omni-channel service integration and their performance, a correlation analysis has been performed. Respondents were surveyed about the following performance measures: increase in traffic on any channel (TRA), increase in the amount of transactions (TXN), increase in customer base (CUS) and finally a decrease in service waiting time (SER). The results of this correlation analysis are shown below in table 7. Because the relationship between different performance measures was not relevant for this study, only the relation to omni-channel was relevant, therefore only these results are shown below. The analysis shows that no significant correlation was found related to service waiting time, meaning that service waiting time did not decrease with a better omni-channel service offering. This can be explained by a number of different factors which will be discussed in more depth in section 5.2.
Significant regression equations were found related to traffic (F(1,27) = 5.760, p < 0.024), transaction amount (F(1,26) = 5.604, p < 0.026) and customer base (F(1,26) = 4.485, p
< 0.044.) with R2’s of 0.176, 0.177 and 0.147, and β’s of 0.419, 0.421 and 0.384 respectively.
A bank’s predicted traffic growth is equal to 3.566 + 0.902 units of omni-channel service offering, predicted transaction growth is equal to 3.528 + 0.844, and customer base growth is equal to 3.527 + 0.791. Traffic, transaction amount and customer based grew with 0.902, 0.844 and 0.791 respectively for each unit of omni-channel service offering. The analysis shows that it is beneficial for banks to integrate their services when it comes to traffic, amount of transactions, and customer base. OMNI SER 0.990 TRA 0.024 TXN 0.026 CUS 0.044 4.3 Overview of hypotheses Table 8 shows an overview of the hypotheses that were supported or not supported after having been tested in this research. H1: An omni-channel service offering has a direct positive effect on relationship quality. Supported H2: An omni-channel service offering has a direct positive effect on satisfaction. Supported H3.1: An omni-channel service offering has a direct positive effect on trust. Supported H3.1.1: An omni-channel service offering has a direct positive effect on perceived competence. Supported H3.1.2: An omni-channel service offering has a direct positive effect on perceived integrity Supported H3.1.3: An omni-channel service offering has a direct positive effect on perceived benevolence. Supported H3.2: The positive relationship between a bank’s omni-channel service offering, and improved trust, is moderated by the type of bank, so that this relationship is weaker for online-only banks. Not supported Table 7: Performance measures bank side
