T
HE
E
FFECTIVENESS OF
C
USTOMER
S
OLUTIONS
T
HE
E
FFECTIVENESS OF
C
USTOMER
S
OLUTIONS
I
TS ANTECEDENTS ANDE
FFECT ONM
EASURES OFC
USTOMERF
EEDBACK.
University of Groningen Msc: Business Administration Specialization: Marketing Management
Qualification: Master Thesis
Faculty supervisor: Dr. T. Wiesel Co-assessor: Dr. J. van Doorn Organization: Wim Bosman Holding B.V.
External supervisor: M. Schoofs
Author: S. van de Pavert
Address: Overboslaan 74, 3722 BM, Bilthoven E-mail: s.van.de.pavert@student.rug.nl
Student number: 1718339
M
ANAGEMENT
S
UMMARY
The aim of this research is to provide more insight in selling solutions within a business-to-business environment. The conceptualization of this research is divided into two parts. The first part will examine the effect of customer antecedents of solution effectiveness. The second will analyze the effect of solution effectiveness on well-known measures of customer feedback (satisfaction, loyalty and customer profitability). A survey has been conducted, whereby about 180 customers of a Dutch logistic service provider evaluated the solution delivered by Company X1. The respondents gave answers to questions regarding the factors
affecting solution effectiveness, solution effectiveness itself and the effect of solution effectiveness on the measures of customer feedback. It appears that the customer antecedents political counseling (the extent of the exchange of political information in the relation between buying en selling team), operational counseling (exchange of operational information) and customer adaptiveness (the willingness of customers to adapt) all together have positive effects on the effectiveness of solutions. It seems that stimulating the exchange of information between buying and selling team regarding the political and operational issues would effectuate a solution. Furthermore, this research will demonstrate that there is no evidence that the increase of tie intensity between buying and selling teams or the increase of matching ties (people with the same kind of expertise) between these teams would stimulate this effect. Moreover, this research will also explain that there is no reason to believe that an effective solution would increase the satisfaction, loyalty or profitability of the customers.
1 In this research the term ‘Company X’ is used for a Dutch logistic service provider at which this research
P
REFACE
This thesis is my last step to complete my Master of Business Administration with a specialization in Marketing Management. This paper has been written during an internship at logistic service provider Wim Bosman in the Netherlands.
T
ABLE OF
C
ONTENTS
Management Summary ... 3
Preface ... 4
Table of Contents ... 5
1 Introduction ... 7
2 Conceptual Framework ... 11
2.1.2 Solution Effectiveness ... 13
2.2 Customer Antecedents of Solution Effectiveness ... 14
2.2.1 Customer adaptiveness ... 14
2.2.2 Political Counseling ... 15
2.2.3 Operational Counseling ... 16
2.3 Moderating Effects of Relational Variables ... 16
2.3.1 Tie Intensity ... 17
2.3.2 Matching ties ... 17
2.4 The Effect of Solution Effectiveness on Measures of Customer Feedback ... 18
2.4.1 Satisfaction ... 18
2.4.2 Loyalty ... 19
2.4.3 Profitability ... 20
2.5 Control variables ... 22
3 Research design ... 23
3.1 Sample & Data Collection ... 23
3.2 Causality ... 25
3.3 Measurement scales ... 26
3.4 Reliability ... 27
4 Results ... 30
4.1 Descriptive statistics ... 30
4.2 Results Reliability analysis ... 30
4.3 Results Factor analysis ... 31
4.4 Model assumptions ... 32
4.5 Hypothesis testing ... 33
4.6 Results multiple regression analysis ... 33
4.6.1 Modering effects of relational variables ... 35
4.7 The effect of solution effectiveness on measure of customer feedback ... 35
5 Discussion ... 39
5.1 Antecedents of solution effectiveness ... 39
5.2 Moderating effects of relational variables ... 40
5.3 Solution effectiveness and the effect on measures of customer feedback ... 41
6 Managerial Implications ... 42
7 Limitations and Directions for Further Research ... 44
1
I
NTRODUCTION
Top managers constantly face the pressure of external factors in their attempt to increase the value of their company (e.g. intensified competition and the growing needs of the customers). Because of this the role of marketing in generating this value has changed during the last couple of years. Models of economic exchange converted from a goods-dominant into a service-goods-dominant view (Vargo & Lusch, 2004). In a goods-goods-dominant view tangible output and transactions are central. However, in a service-dominant view intangibility, exchange processes and the relationships are more important. In this perspective the economic exchange is no longer a transaction of goods or services provided by a manufacturer but the exchange is a customized offering where the customer is a co- producer instead of only the receiver of it. In addition, Lusch and Vargo (2006) refine their view by changing the term ‘co-producer’ into a ‘co-creator of value’ because they recognize two components of co-creation of value (during production and when it is in use). Furthermore, in their research about the effects of service transition strategies on firm value, one of the advises of Fang, Palmatier and Steenkamp (2008) is to focus service initiatives on closely related businesses in order to allow the firm value grow. Elaborating on this advise “solution selling” can be the way to enhance what they call “synergistic spillover benefits”. Their results suggest that companies should avoid unrelated service initiatives because of the strong interaction between service ratio and service relatedness. According to this view, offering customers a bundle of related products or services in order to meet their needs can be the remedy to increase firm value.
and services and typified it as a set of four relational processes. They suggest that there are four relational processes between customer and supplier when offering a solution; comprising (1) the definition of the requirements, (2) the customization and integration of goods and services, (3) the deployment and (4) the post-deployment support of the solution. The purpose of this relational process view is to meet customer’s business needs in a better way. This relational perspective is consistent with the dramatic growth of relationship management in marketing literature last years (Boulding, Staelin, Ehret & Johnston, 2005; Palmatier, Dant, Grewal & Evans, 2006; Krasnikov, Jayachandran & Kumar, 2009). Managing customer relationships can increase the knowledge about customers in order to meet customer needs more precisely, which would enhance customer value in the long term.
new developed products or services more effective (Hoyer, Chandy, Dorotic, Krafft & Singh, 2010). Furthermore, Kumar et al. (2010) argue that customers should not be evaluated solely by their purchase behaviour, therefore, a more comprehensive assessment of the customer is needed. Finally, taking these findings from extant literature into account an effective solution can be created by having a closer look at the relationship between buying and selling organizations. In this relationship it seems to be considerable to analyze a solution as four relational processes, stimulate matching ties between these buying and selling teams and having more insight in how customers engage to a firm in order to effectuate customer solutions and meet the needs of the customer.
In order to be able to answer the problem stated in the introduction, the problem statement of this study is:
Which factors affect solution effectiveness and do they affect measures of customer feedback?
To answer the problem statement, the following research questions are formulated: 1. Which antecedents affect solution effectiveness in a b2b contractual context? 2. Are there relational variables that mediate the relationship between customer
antecedents and solution effectiveness?
3. What is the effect of solution effectiveness on measures of customer feedback?
Answering these questions will contribute to literature in the following ways. First, factors that affect the effectiveness of solution selling have not been investigated adequately and have not been tested in practice. Secondly, to the best of my knowledge the relation between solution effectiveness and measures of customer feedback (i.e. satisfaction, loyalty and profitability) have not been researched before. Therefore, this research is interesting for marketing managers, as well as general manager, to get a deeper insight in selling solutions
to customers.
2
C
ONCEPTUAL
F
RAMEWORK
To provide an answer to the problem stated in the introduction a conceptual framework is developed (figure 2.1). It will analyze the effect of factors that influence solution effectiveness and the effect of solution effectiveness on measures of customer feedback. To start with, solution effectiveness and their relationships with measures of customer feedback (loyalty, satisfaction and profitability) will be analyzed. Next , the relationship between customer antecedents (Customer Adaptiveness, Political- and Operational counselling) and solution effectiveness will be examined. Then these outcomes of the research will be analyzed to see whether relational variables will mediate the relationship between the factors that affect solution effectiveness. The conceptual model contains two parts. The first part will examine the solution effectiveness and their customer antecedents. The second part will analyze solution effectiveness and the effect on measures of customer feedback2.
Figure 2.1: Conceptual model
In the following chapter the constructs of the conceptual model will be outlined in detail and hypotheses will be introduced.
2 The terms first and second part of the conceptual model will be used in this research.
2.1 C
USTOMERS
OLUTIONS ANDS
OLUTIONE
FFECTIVENESS2.1.1 Customer Solutions
Throughout the past couple of years, marketing literature have evolved from a product dominant into a service dominant conceptualization of economic exchange (Vargo & Lusch, 2004). Customers do not buy goods or services alone but want an offering which create value for them. Anderson, Narus and Narayandas (2009) distinguish the business market process into three value creating processes, including; understanding, creating and delivering value. When focussing on the creation of value in a B2B market, offering business customers a solution could be the way to realize this (Lillien et al., 2010).
To define a customer solution, Tuli, Kohli and Bharadwaj (2007) present three commonalities in customer solution literature. The first common definition they introduce is a solution based on a bundle of goods and services. Second, a solution is defined as a combination of goods and services customized to the specific needs of a customer. Third, each part of a solution (product or service) must be, what they call, “working” together. This would result in an integrated set of goods and services. Furthermore, they argue in their research based on in-depth interviews that a solution is more than an integrated or customized set of services or products. They state that it is a combination of four relational processes which is consisting with Vargo and Lusch’ S-D Logic. Moreover, a solution is viewed from a customer’s point of view. I will use the definition of Tuli, Kohli and Bharadwaj (2007) of customer solutions in this research:
A solution is a set of customer-supplier relational processes comprising (1) customer requirements definition, (2) customization and integration of goods and/or services and (3) their deployment, and (4) postdeployment customer support, all of which are aimed at meeting customers’ business needs.
When comparing this definition with the others, one of the main differences is the fact that a customer views a solution as a relational process. Besides that, Tuli, Kohli and Bharadwaj (2007) add postdeployment customer support because they see it as an underemphasized relational process that customers consider as crucial.
strengthen relational ties between supplier and buying team to analyze the business needs of the customer precisely now and in the future in order to effectuate the solution (Murtha, Bharadwaj & Van den Bulte, 2009). Second, Customization and Integration of a solution is consisting with the former definitions of solutions. Customization exists of designing, modifying, or selecting products to meet the needs of the customer (Tuli, Kohli & Bharadwaj, 2007). Integration of a solution involves designing, modifying, or selecting goods and services so that they work properly with one another. Third, the method to deploy the goods or services is essential in delivering a solution. Deployment of a solution is the actual delivery of products or services and their installation into a customer’s environment. During the implementation process new customer needs can arise, therefore, analyzing this stage and anticipating on these needs is crucial. Finally, Postdeployment Support is the last but one of the most critical steps in this process. A proactive postsales service can benefit from new product ideas and can enhance supplier’s satisfaction and reputation (Challagalla, Venkatesh & Kohli, 2009). Postdeployment support is not only the maintenance or delivery of additional spare parts, but it also means the response of the supplier to the evolving needs of their customers (Tuli, Kohli & Bharadwaj, 2007).
Regarding to these customer insights, a solution is more effective when interpreted as a relational process, which is consisting with dramatic growth in the attention to relationships in marketing literature (Palmatier et al., 2006; Verhoef, 2003; Reinartz & Kumar, 2003). Moreover, offering solutions to customers in order to meet their needs is in line with the change from a goods dominant into a service dominant view in marketing literature (Vargo & Lusch, 2004). This research will test the relationship between solution selling and customer behavioural outcomes as customer satisfaction, loyalty and their relative profitability to the firm.
2.1.2 Solution Effectiveness
incentive externality, customer interactor stability and process articulation. Moreover, in their research customer variables have an impact on the effectiveness of solutions. These variables consist of customer adaptiveness, political counseling and operational counseling. Besides, in relationship between buying and selling teams in a B2B environment, tie intensity and matching these ties are viewed as instruments that enhance the effectiveness of a solution (Murtha, Bharadwaj & Van den Bulte, 2007). Tie intensity refers to the frequency of the interaction between buying and selling teams. The concept of matching ties is the need to connect people with the same domain of expertise in two organizations. To illustrate, in a logistic company it is important to create strong ties between IT specialists in a buying and selling team in order to streamline the supply chain and the communication between these links.
This research focuses on the customer variables described above (customer adaptiveness, political counseling and operational counseling) that would probably affect the effectives of solutions. Furthermore, the relational constructs will be proposed as mediators of the relationship between solution effectiveness and the customer antecedents. These factors will be outlined in the next sections.
2.2 C
USTOMERA
NTECEDENTS OFS
OLUTIONE
FFECTIVENESS2.2.1 Customer adaptiveness
It is important to meet the needs of the customers when delivering a solution. Customers are also willing to adapt their internal routines and processes to accommodate supplier’s products (Tuli, Kohli & Bharadwaj 2007). In the research of Tuli and colleagues several customers noted the importance of adaptiveness in solution selling. The willingness of customers to adapt to their internal routines can enhance the effectiveness of a solution by stimulating the relational processes. A more adaptive customer makes it easier to customize and integrate the solution. Furthermore, during the deployment and postdeployment stage the solution can be more effective when the customer is willing to adapt. When customers are more willing to adjust and accommodate to supplier’s needs, deployment and postdeployment can be more effective.
flexibility on adaptability. When both buyer and supplier are more willing to adjust readily to changes, the adaptation to uncertainty for both parties will be higher. This flexibility in the different relational processes of a solution could provide a more effective solution for customers. Therefore, this research considers that the effectiveness of a solution can be improved by increasing the adaptiveness of customers.
H1: The greater the customer adaptiveness, the greater is the solution effectiveness.
2.2.2 Political Counseling
According to Tuli, Kohli and Bharadwaj (2007) political counseling refers to the extent to which a customer provides a supplier with information and guidance regarding the political landscape in the customer organization. Politics in organizations can influence critical processes in organizations. These critical processes (e.g. resource allocation or managerial decision making) influence the organizational effectiveness (Chang, Rosen & Levy, 2009). Political counseling helps a supplier to find a way in the political landscape of customers. In marketing literature it is analyzed as one form of the norm of information exchange (Heide and John, 1992) and can be defined as the extent to share information that may be relevant to both parties (Cannon and Perrault, 1999). Guidance in organizational politics of the customer is needed to gain a better understanding regarding the priorities of various stakeholders. It will take too long for a supplier to learn about internal politics of customer firms. Customers of a solution should be willing to navigate vendors through their own political landscape. For example, the success of a solution in case of a specific stakeholder can be critical or the introduction of a new solution can cause resistance because of organizational changes (Cummings & Worley, 2003). Therefore, political counselling can effectuate the relational processes of a solution. Customer’s requirements will be defined in a more nuanced manner, customization and integration address the sensitivities of various stakeholders and during deployment and postdeployment of a solution political counseling can be critical in the success of the solution (Tuli, Kohli & Bharadaj, 2007).
Handling with politics in organizations is a critical factor in the effectiveness of organizations and possibly will influence the effectiveness of a solution, therefore:
H2: The greater the customer’s political counseling, the greater is the solution
effectiveness.
2.2.3 Operational Counseling
Similar as with political counseling, operational counselling is a form of the information exchange norm. Heide and John (1992) define this norm of information exchange as the expectations of both parties to provide pieces of information that might help each other. According to Tuli, Kohli and Bharadwaj (2007), information regarding operations refers to information about technical systems, business processes, and firm policies that is important for implementing a solution. In order to create a suitable solution it is important for a supplier to know what is going on in the operational environment of the customer. Knowing how the business is working provides the vendor the opportunity to know what the solution would do for the customers. This will stimulate supplier’s learning about specific businesses which would make a completely and more accurately definition of the customer’s requirements possible. Besides, an open information exchange between firms would enhance the effect of trust in the vendor-customer relationship and stimulate a beneficial partnership between both organizations (Aulakh, Kotabe & Sahay, 1996). The other relational processes of the solution would also benefit from knowledge of customers’ operations. Customization and integration of a solution will suit with the business environment when having extensive knowledge about the operations of a customer. During deployment of a solution it may be essential to have correct knowledge about customers’ operations because of the unique aspects. Moreover, with operational counseling the supplier of solutions remains informed about changes in their operational environment when sharing information proactively. When operations change, new requirements would be identified which would increase the effectiveness of postdeployment support, therefore:
H3: The greater the customer’s operational counseling, the greater is the solution
effectiveness.
2.3 M
ODERATINGE
FFECTS OFR
ELATIONALV
ARIABLESsolutions could be influenced by the aspects of these relationships. Murtha, Bharadwaj and Van den Bulte (2009) describe the concept of relational ties between buying and selling team. This research will investigate what the effects of these relational ties are on the relationships between solution effectiveness and their customer antecedents. The following sections describe the concept of tie intensity and matching ties.
2.3.1 Tie Intensity
According to Murtha, Bharadwaj and Van den Bulte (2009), tie intensity is the frequency of interaction between two parties. Prior research investigated the importance of strong ties in the relationships between buyer and seller (Spekman & Johnston, 1986). Intense ties can provide coordination benefits, mobilize support and can transfer complex knowledge (Wuyts et al. 2004; Uzzi & Lancaster, 2003). Strong ties can enhance the effectiveness of solutions but it depends on the role of tie strength in the relational process of a solution. Strong ties can have benefits for both knowledge transfer as well as governance roles in the relationship between vendors and buyers (Murtha, Bharadwaj & Van den Bulte, 2009). However, in the development stage of a solution governance oriented activities are important when integrating and customizing a solution. Besides, in the deployment stage, especially, knowledge transfer, is important in buyer-seller dyads. These findings suggest that strong ties can complement the relational processes and consequently will increase the effectiveness of a solution (Murtha, Bharadwaj & Van den Bulte, 2009). Therefore, highly intensive ties could strengthen the relationship between solution effectiveness and their antecedents. Thus,
H4a: When tie intensity is high, the relationship between customer adaptiveness and
solution effectiveness will be strengthened.
H4b: When tie intensity is high, the relationship between political counseling and
solution effectiveness will be strengthened.
H4c: When tie intensity is high, the relationship between operational counseling will be
strengthened.
2.3.2 Matching ties
keeping track of the current status of the development and deployment of the solution (Murtha, Bharadwaj & Van den Bulte, 2009). For example, these ties can exist between an information technology (IT) specialist from the selling organization and the IT manager from the buying firm. Another example of a matching tie is the relationship between an account manager and purchasing manager from both firms, because of the fact that both have the same kind of expertise and specific knowledge in order to coordinate the relational processes of a solution. Coordinating expertise in teams is a crucial factor in enhancing team performance (Faraj & Lee, 2000). However, matching ties can be inefficient when maintaining a large group of contacts is demanding or mobilizing the network is hard. In these situations, going beyond matching ties and work indirectly through one’s own experts can be more efficient (Murtha, Bharadwaj & Van den Bulte, 2009). Nevertheless, an overall main effect of matching ties on the relationship between solution effectiveness and their customer antecedents is expected. Therefore,
H5a: When there is a high degree of matching ties in a buyer-seller relationship, the
relationship between customer adaptiveness and solution effectiveness will be strengthened.
H5b: When there is a high degree of matching ties in a buyer-seller relationship, the
relationship between political counseling and solution effectiveness will be strengthened.
H5c: When there is a high degree of matching ties in a buyer-seller relationship, the
relationship between operational counseling will be strengthened.
2.4 T
HEE
FFECT OFS
OLUTIONE
FFECTIVENESS ONM
EASURES OFC
USTOMERF
EEDBACK2.4.1 Satisfaction
satisfaction describes the “postchoice evaluative judgment of a specific purchase occasion”. In addition, cumulative satisfaction can be defined as “the overall evaluation based on the total experience with a good or service over time”.
When relating customer solutions to satisfaction, the cumulative satisfaction conceptualization of Anderson, Fornell and Lehmann (1994) will be used in this research. This definition of satisfaction as an “overall evaluation based on the total experience” is more extensive than the transaction specific view. This is consisting with the definition of Tuli, Kohli and Bharadwaj (2007) who argue that a solution is more than an integrated bundle of goods and services (transactions) and define it as a set of four relational processes. Tuli and his colleagues also state that suppliers’ inattention to these processes would result in dissatisfied customers. Moreover, customers have no interest in products and services alone, they want a solution for problems they face in their business (Sawhney, 2006). Therefore:
H6: Solution effectiveness has a positive relationship with satisfaction.
2.4.2 Loyalty
According to the service-profit chain, customers who are satisfied are more loyal to the company (Heskett et al., 1994). The relation between satisfaction and retention has received broad attention in marketing literature. Oliver (1999) demonstrates clearly that satisfaction and retention are two distinct concepts, however, satisfaction is an essential ingredient for the emergence of loyalty. Not every satisfied customer will retain at the company, but they also state that it is hard to develop loyalty without satisfaction. Satisfying customers in a service industry can enhance loyalty.
In literature loyalty is mostly distinguished in two kind of constructs: behavioural loyalty and attitudinal loyalty. Behavioural loyalty in a B2B context means the willingness of business customers to repurchase the offering and maintain the relationship with the supplier (retention). Attitudinal loyalty can be defined as the psychological attachments and attitudinal advocacy towards the service provider (Rauyruen & Miller, 2007).
The link between loyalty and profitability has been discussed extensively in marketing literature. The connection seems obvious, however, in literature there is some dissemination about this relation. In their extensive research Rust and Zahorik (1993) show the link between satisfaction, retention, market share and in the end profitability. They provide a framework for how to allocate budget in order to improve satisfaction and how it will have a positive effect on the profitability of customers. Nevertheless, several scientists also claim that the relationship between loyalty and profitability is not as strong as been argued by others. Reinartz and Kumar (2002) state in their research about the mismanagement of loyalty that some claims about loyalty are not generalizable in all situations. The costs to serve a loyal customer are not always lower. They argue that experienced customers know their value to the company and often try to get a premium service for the lowest price, this disproves the claim that loyal customers pay higher prices for the same bundle of services and/or products.
Although the profitability of loyalty is debatable, the retention of customers is an important element in improving the lifetime value of customers (Berger & Nasr, 1998; Gupta & Zeithaml, 2006). Because a solution is more than a bundle of products and can be seen as a set of four relational processes with the intention to enhance a strong relationship with the customer (Tuli, Kohli and Bharadwaj 2007), this research suppose the following hypothesis:
H7: Solution effectiveness has a positive relationship with loyalty.
2.4.3 Profitability
the specification of a profitability analysis will vary for different applications. The purpose of this profitability analysis is to analyze the effect of solution effectiveness on the profitability of business customers. Therefore, the key measurement components and issues involved in this analysis are specified to this purpose. The customer unit of this profitability analysis is the sample group used in this research with existing “active” customers. These are customers of predetermined business units, part of a large logistic service provider in the Netherlands. In this analysis the profit of the year 2010 will be measured. The cost allocation concentrates on the purpose of this analysis (the effect of solution effectiveness on customer profitability). Because future purchase and cost streams cannot be accurately forecast at an individual level, a historical profitability model (Mulhern, 1999) will be used:
(1) where,
CPi = the profit of customer i to a firm, T = time horizon
t = time index
Ji = total purchases J made by customer i j = purchase
Pij = price of purchase j made by customer i in period t, Cij = unit cost of purchase j made by customer i in period t, Ki = total marketing costs attributed to customer i
k = marketing costs
MCikt = variable marketing cost, k, for customer i, in period t, I = annual inflation rate
needs when improving the relational processes of a customer solution (Tuli, Kohli and Bharadwaj 2007) , this research supposes:
H8: Solution effectiveness has a positive relationship with customer profitability.
2.5 C
ONTROL VARIABLESIn this research I will use several control variables to test the relative impact of the independent variable in a relationship. The control variables are workforce, solution experience, dependence on seller, commitment and trust. Workforce is chosen because there are strong differences in workforce of the organizations of the respondents. It could be that the effectiveness of solutions has been influenced by the size of the organization of the customer. Furthermore, solution experience can be compared with product-norm experience and represents prior knowledge and information about how a solution would perform (Sharma & Patterson, 2000). A more experienced customer in the area of the solution can possibly increase or decrease the effectiveness of it. This research also controls the dependence on seller. Dependence on seller is the need of the buyer organization to maintain the relationship with a seller in a business relationship and it is mostly determined by how difficult it is to replace that seller organization (Fang, Palmatier & Evans, 2008). Therefore, dependence of the buyer organization on the seller firm can influence the effectiveness of a solution. Finally, the control variables commitment and trust will be used. Commitment is “an enduring desire to maintain a valued relationship” (Moorman, Zaltman, & Desphandé, 1992, p. 316), and trust is “confidence in an exchange partner’s reliability and integrity” (Morgan & Hunt, 1994, p. 23). In literature, there is little agreement about the strength of these constructs to capture the key aspects of a relationship (Palmatier et al., 2008). Therefore, these variables will be used as control variables in this research.
3
R
ESEARCH DESIGN
This section will describe the methodology of the research. The hypotheses of the previous chapter will be investigated with an empirical study. In this study the differences in the relationships of the conceptual model will be measured. For this survey customers of a logistic service provider in a business-to-business environment were asked to fill in a questionnaire.
3.1 S
AMPLE&
D
ATAC
OLLECTIONA total number of 475 business-to-business customers from a Dutch logistics service provider were invited to participate in the survey. These customers are picked randomly from a database of customers with a revenue of € 5000,- or more in year 2010 at Company X. The respondents were asked to fill in a digital questionnaire by e-mail. From these 475 customers, 182 (38%) handed in this questionnaire. Only 153 (32%) respondents filled in the questionnaire completely. Most of the respondents were firms with a workforce of 0 – 99. This group contains 111 companies (73%). Furthermore, 12% of the responded organisations had a workforce between 100 and 199; 4% between 200 and 299; 3% between 300 and 399; and 9% of the responded firms had a workforce of 400 employees or more. To determine to what extent the respondents have already received a solution, the customers that participated in the survey were asked to choose a situation which best fits the company’s current relationship with Company X. Respondents could choose between the following situations (ranging from a non-solution situation to a full solution situation).
Situation 1: We sometimes make use of logistic services of Company X.
Situation 2: We are regularly doing business with Company X. We make use of several services and collaborate with our contact at Company X, when needed.
Situation 3: We are having an intensive relationship with Company X. We regular contact a logistic specialist at Company X about our needs. Together we try to provide the most efficient supply chain.
Most of the respondents rank themselves in situation 2 (48,4%). 28,1% chose situation 3, 22,2% situation 1 and 1,3% chooses situation 4 (figure 3.1).
The sample is composed of small and large customers (in terms of relative contribution to gross profit of Company X). Figure 3.3 shows the relative contribution to the profit of Company X. Most of the respondents contribute a maximum of € 10,000 gross profit margin to the total profit of the organization.
3.2 C
AUSALITYTo examine whether there is a true causal relationship between the independent and dependent constructs, causal research is needed (Malhotra, 2007, 220). This causality analysis is divided into correlation tests and regression analyses.
With correlation analysis the strength of association between two metric variables will be measured (Malhotra, 2007). In this case, the product moment correlation r (Pearson correlation) is the commonly used method. It indicates the degree to which the variation of one variable is related to another variables. The correlation of the relationships in the hypotheses (H1 – H10) will be tested.
(Hair et al., 2010). Therefore, we examine our proposed relationships on linearity by scatter diagrams. Furthermore, because an orthogonal rotation method is used, multicollinearity between the factors (constructs) in multiple regression analysis is no issue. Moreover, to prove a causal relationship where the independent variable causes the dependent variable and not the other way around, regression analysis is applied by which these variables are turned around. In addition, some variables are included to control for other possible causal factors (Hair et al., 2010).
3.3 M
EASUREMENT SCALESThis paragraph will demonstrate which measurement scales have been used to measure the several constructs. To measure the constructs in the conceptual model the scales of solution effectiveness, political counseling, operational counseling and customer adaptiveness will be explained. Furthermore, the constructs satisfaction, loyalty and profitability will be outlined. Also the moderating relationship constructs will be described in this section.
Three customer-based antecedents are included in the conceptual model namely, political counseling, operational counseling and customer adaptiveness. Political counseling and operational counseling are based on the norm of information exchange from Jap en Ganesan (2000), specified to respectively political and operational issues. Customer adaptiveness is measured by a relational flexibility scale of Noordewier, John and Nevin (1990) and can be compared with customer adaptiveness (Tuli, Kohli & Bharadwaj, 2007). All these customer-based antecedents are measured on a 7-point Likert scale (1 = very strongly disagree, 7 = very strongly agree).
measure satisfaction and loyalty, scales from Lam et al. (2004) and Zeithaml, Berry and Parasuraman (1996) were used to asses these constructs, which are measured on a 7-point Likert scale, ranging from 1 = very strongly disagree, 7 = very strongly agree. In the survey respondents were asked to fill in the name of the organization. This was needed to match the survey with the customer profitability measure. For measuring customer profitability a model of Mulhern (1999) was used as described in paragraph 2.1.4. To measure solution effectiveness and the antecedents also a 7-point Likert scales were used (again, ranging from 1 = very strongly disagree, 7 = very strongly agree). These four constructs will be grouped together and measure the effectiveness of a solution offered to customers.
The measurement items of the described constructs above can be found in appendix 1 (table 3.1).
3.4 R
ELIABILITYTo assess the reliability and validity, the quality of measurement of several constructs will be tested. A reliability analysis will be used to test the internal consistency of the measurement items. Reliability of these scales is an assessment of the degree of consistency between the multiple measurements of several variables (Hair et al., 2010). The co-efficient used to assess the consistency of the entire scale is the Cronbach’s alpha. The Cronbach’s alpha is a widely used measure and in most of the cases the upon lower limit of this alpha is 0,7. A reliable scale with sufficient internal consistency will have a Cronbach’s alpha of 0,7 or higher.
sampling adequacy should be above 0,5, but one should aim for 0,6 (Hair et al., 2010). Because the primary objective is to identify the latent dimensions, a common factor analysis (Principal Axis Factoring) will be used in this research. To better interpret the factors a factor rotation will be performed. With factor rotation the reference axes of the factors will be turned to make better interpretation of the factors possible. The rotation method that will be used called, Varimax, which is an orthogonal rotation method. This method estimates uncorrelated factors which would be useful for conducting other multivariate analysis (e.g. Regression Analysis) techniques. Finally, the factors scores can be used for regression analysis to test the hypotheses described in the previous chapter (Hair et al., 2010).
3.5 R
EGRESSION ANALYSISThis research will use regression analysis, a dependence technique, to analyze the relationship between the dependent and independent variables in the conceptual model. The objective of regression analysis is to predict a single dependent variable from the knowledge of one or more independent variables (Hair et al., 2010). In the first part of the conceptual model multiple regression analysis will be used to test the relationship between grouped variable solution effectiveness as dependent variable and independent variables political counseling, operational counseling and customer adaptiveness. With these regression analyses the hypotheses 4, 5 and 6 (described in chapter 2) will be tested. The model to be estimated is:
SE = β0 + β1WOR + β2COM + β3DEP + β4TRU + β5EXP + β6TIE + β7MAT
+ β8POL + β9OPE + β10CAD + ε (2)
where,
SE = Solution Effectiveness;
WOR = Workforce;
COM = Commitment;
DEP = Dependence on Seller;
TRU = Trust;
EXP = Solution Experience;
TIE = Tie Intensity;
MAT = Matching Ties;
POL = Political Counseling;
OPE = Operational Counseling;
CAD = Customer Adaptiveness;
To test hypotheses 7 (a, b and c) and 8 (a, b and c) a new model will be estimated. This model will test for the moderating effects of relational variables tie intensity and matching ties on the relationships between solution effectiveness and their customer antecedents. Therefore, I included interaction effects. The models to be estimated are:
SE = β0 + β1WOR + β2COM + β3DEP + β4TRU + β5EXP + β6TIE + β7MAT
+ β8POL + β9OPE + β10CAD + β11(TIE * POL) + β12(TIE * OPE) +
β13(TIE * CAD) + ε (3)
SE = β0 + β1WOR + β2COM + β3DEP + β4TRU + β5EXP + β6TIE + β7MAT
+ β8POL + β9OPE + β10CAD + β11(MAT * POL) + β12(MAT * OPE) +
β13(MAT * CAD) + ε (4)
In the last part of the conceptual model three simple regression analysis will be performed individually to test the relationships between independent variable solution effectiveness and dependent variables satisfaction, loyalty and profitability. With these simple regression analyses, hypotheses 1, 2 and 3 will be tested. The models to be estimated are:
SAT = β0 + β1SE + β2WOR + β3COM + β4DEP + β5TRU + β6EXP + ε (5)
LOY = β0 + β1SE + β2WOR + β3COM + β4DEP + β5TRU + β6EXP + ε (6)
CPR = β0 + β1SE + β2WOR + β3COM + β4DEP + β5TRU + β6EXP + ε (7)
4
R
ESULTS
The results of the analysis will be outlined in this chapter. The first paragraph contains information about the population, descriptive statistics and reliability of the analysis. Furthermore, I will discuss the results of the factor analysis. Also, the results of the regression analysis will be discussed.
4.1 D
ESCRIPTIVE STATISTICSA total number of 475 business-to-business customers from a Dutch logistics service provider were invited to participate in the survey. 182 (38%) handed in the questionnaire. Only 153 (32%) respondents filled in the questionnaire completely. Most of the respondents were small firms with a workforce of 0 – 99. This group contains of 111 companies (73%). Furthermore, 12% of the responded organisations had a workforce between 100 and 199; 4% between 200 and 299; 3% between 300 and 399; and 9% of the responded firms had a workforce of 400 employees or more.
Most of the questions in the survey were answered on a Likert scale ranging from 1 to 7. In appendix II the means and standard deviations of the measurement items is presented. The mean of most of the measurement items is around 3 and 4. This suggests that none of the items scored very high or low. Most of the standard deviations are above one, which indicates that the respondents answered quite differently from each other.
4.2 R
ESULTSR
ELIABILITY ANALYSISIn order to perform factor and regression analysis, a reliability analysis have been conducted. To create reliable construct, the Cronbach’s Alpha of the these constructs was estimated and the relationships between the construct in the conceptual model are analyzed correlation. The Cronbach’s Alpha of the various constructs are shown in appendix III. SPSS (statistical software) also conducted the Cronbach’s Alpha if a specific item was deleted. With this statistic it is possible to improve a construct by deleting one or more items. Two items (CUI2 and EXP1) will be excluded in further analysis, because the Cronbach’s Alpha of the whole construct will be higher when deleting this item.
there is correlation between the customer antecedents of solution effectiveness. This theorizes a causal relationship between these variables and supports hypotheses 1, 2 and 3. A correlation analysis is also performed for the second part of the conceptual model. This analysis (Appendix IV, table 4.12) also shows significant correlation between solution effectiveness and the satisfaction, loyalty and profitability construct. These correlations support hypotheses 6, 7 and 8.
4.3 R
ESULTSF
ACTOR ANALYSISfor the control variables (table 4.15, appendix V). In this analysis only the eigenvalue criterion did not meet (0,769 at 4 components). The variance explained per factor is at least 7% and the total variance explained is 82%. The computed factor scores will be used in regression analysis.
4.4 M
ODEL ASSUMPTIONSIn order to obtain reliable results, the assumptions for regression analysis are examined. First, it is assumed that the variables analysed in a regression model are correlated with each other. Table 4.11 (Appendix IV) shows the outcomes of the correlation analysis of the first part of the conceptual model. Solution effectiveness, which will be used as dependent variables in a multiple regression analysis, correlates with all the customer antecedents. Customer adaptiveness has the highest correlation (0,367), but it is almost equal to Political Counseling (0,345) and Operational Counseling (0,367). All these correlations are significant at the 0,01 level (2-tailed). In the second part of the conceptual model, the relationship between solution effectiveness and satisfaction, loyalty and customer profitability will be tested. Table 4.12 (Appendix IV) demonstrates the correlations between these variables. All these variables are correlated with solution effectiveness. Satisfaction has the highest correlation (0,493), but loyalty also shows a high correlation (0,449). These correlations are significant at the 0,01 level (2-tailed). There is also a correlation (0,202) between customer profitability and solution effectiveness at the 0,05 level significance level (2-tailed).
Secondly, a regression analysis expects linear relationships. Therefore, several scatter plots are analysed. First, a residual plot was conducted to analyse the relationship between the multiple customer antecedents and solution effectiveness. This research assumes a linear relationship between these antecedents and solution effectiveness when analyzing the residual plot and partial regression plots (Appendix VI, figures 4.1 until 4.4). For the second part of the conceptual model, three scatter plots are conducted to analyze the relationship between solution effectiveness as independent variable and satisfaction, loyalty and customer profitability as dependent variables. Also in these cases linear relationships are assumed when analyzing the plots (Appendix VI, figures 4.5 untill 4.7).
4.5 H
YPOTHESIS TESTINGTo analyze the relationships between constructs in the conceptual model, regression analysis is used to compute the relative contribution of the independent variable(s) in predicting the dependent variable and trial the hypotheses. First, a multiple regression analysis is computed to check the effect of customer antecedents of solution effectiveness and the moderation effect of relational variables. Thereafter, three simple regression analysis are conducted separately. In both analyses a hierarchical regression method will be used. This means that the relationship is examined after control for several variables (trust, commitment, workforce, dependence on seller and solution experience).
4.6 R
ESULTS MULTIPLE REGRESSION ANALYSISTable 4.6: Results multiple regression analysis Model 1, standardized beta Model 2, standardized beta Model 3, standardized Beta (constant) - - - Trust 0,439* 0,402* 0,436* Dependence on Seller 0,310* 0,227* 0,195* Commitment 0,413* 0,301* 0,289* Solution Experience 0,280* 0,229* 0,258* Workforce 0,039 0,021 0,037 Operational Counseling - 0,118** 0,125** Political Counseling - 0,229* 0,226* Customer Adaptiveness - 0,108*** 0,132** TIE * POL - - -0,064 TIE * OPE - - 0,020 TIE * CAD - - 0,030 MAT * POL - - -0,095 MAT * OPE - - 0,063 MAT * CAD - - 0,095 R² 0,532* 0,589* 0,610* Adj. R² 0,516* 0,567* 0,571*
*p-value < 0,01; **p-value < 0,05; ***p-value < 0,10
NOTE: control variables: Workforce, Solution Experience, Dependence on Seller, Commitment and Trust.
proposed as independent variable and the customer antecedents as dependent variables. Solution effectiveness has an effect on political counseling and count for 0,69 R Square change when added to the model (p < 0,01). Nevertheless, Solution effectiveness has no significant effect on customer adaptiveness and operational counseling.
4.6.1 Modering effects of relational variables
As outlined in the conceptual model (figure 2.1), the relationship between solution effectiveness and the customer antecedents is possible moderated by relational constructs tie intensity and matching ties (paragraph 2.3). In order to examine hypotheses 4a, 4b, 4c and 5a, 5b, 5c, a new model will be estimated. To examine the moderating effects of the relational constructs tie intensity and matching ties on the relations between solution effectiveness and their customer antecedents, interaction effects will be included.
In model 3 (table 4.6) the customer antecedents and the moderators tie intensity and matching ties were added. The model, as a whole, is significant. However, none of the interaction effects between relational variables tie intensity and matching ties and the customer variables (political counseling, operational counseling and customer adaptiveness) have a significant effect on solution effectiveness. Therefore, hypotheses 4a, 4b, 4c and 5a, 5b and 5c are not supported. The relationship between solution effectiveness and their customer antecedents is not moderated by relational variables tie intensity and matching ties.
4.7 T
HE EFFECT OF SOLUTION EFFECTIVENESS ON MEASURE OF CUSTOMER FEEDBACKTable 4.7: Regression analysis, satisfaction Model 1, standardized beta Model 2, standardized beta (constant) - - Trust 0,832* 0,847* Dependence on Seller 0,017 0,028 Commitment 0,156* 0,170* Solution Experience 0,269* 0,279* Workforce -0,076** -0,075** Solution Effectiveness - -0,034 R² 0,823* 0,823* Adj. R² 0,817* 0,816*
*p-value < 0,01; **p-value < 0,05; ***p-value < 0,10 NOTE: Dependent variable: Satisfaction.
The R Square (R²) of the regression model is 0,823. This implies that 82,3% of the variance is explained by the model. Furthermore, the overall regression model is significant (p < 0,01, F=113,377). However, most of the variance is explained by the control variables and the effect of solution effectiveness on satisfaction (p=0,509) is not significant. Therefore, H6 is
not supported. The control variables of trust, commitment, solution experience and workforce are significant and describe most of the variance in the model. From all the variables in this regression model, trust has the largest effect on satisfaction.
Table 4.8: Regression analysis, loyalty
Model 1, standardized beta Model 2, standardized beta (constant) - - Trust 0,111*** 0,084 Dependence on Seller 0,403* 0,384* Commitment 0,585* 0,560* Solution Experience 0,015 -0,002 Workforce -0,031 -0,033 Solution Effectiveness - 0,061 R² 0,521* 0,523* Adj. R² 0,505* 0,503*
The R² of the regression model with loyalty as dependent variables is 0,523. The model is significant (p < 0,01). Again, the control variables explain most of the variance of the model. The effect of solution effectiveness is not significant (p=0,464). Thus, H7 is not supported.
Within the model, the control variables of dependence on seller (st. beta 0,384) and commitment (st. beta 0,560) are the only significant constructs and have the largest effect on loyalty.
Table 4.9: Regression analysis, customer profitability
Model 1, standardized beta Model 2, standardized beta (constant) - - Trust 0,063 0,021 Dependence on Seller 0,225* 0,196** Commitment 0,048 0,009 Solution Experience 0,157** 0,130 Workforce 0,058 0,054 Solution Effectiveness - 0,095 R² 0,084** 0,053** Adj. R² 0,088** 0,051**
*p-value < 0,01; **p-value < 0,05; ***p-value < 0,10 NOTE: Dependent variable, customer profitability.
In the third model customer profitability is taken into account as dependent variable in the regression analysis. The R² of the regression model is 0,088 and is significant (p < 0,05). Solution effectiveness is also not significant in this model (p=0,411) and therefore, H8 is not
supported. Dependence on seller is the only variable that is significant in this model (p < 0,05) and has a small effect on customer profitability (st. beta 0,196).
Only dependence on seller has a small but significant impact on the profitability.
Also the variables of the regression analyses in the second part of the conceptual model are turned around to support causality between the independent and dependent variables. Table 4.19 untill 4.21 (appendix VIII) shows the outcome of these analyses. In these analyses, loyalty, satisfaction and customer profitability are taken as independent variables. The simple regression analyses demonstrate that there is no significant effect of loyalty, satisfaction and customer profitability in explaining variation of solution effectiveness.
5
D
ISCUSSION
In highly competitive markets, general and marketing managers in a business-to-business environment constantly ask themselves how to generate customer value in order to maximize firm value and ultimately shareholder value. One of the possibilities to create this customer value in a B2B environment is to offer customers a solution. The aim of this research is to investigate the effectiveness of these solutions and to pinpoint factors to improve the effectiveness. A study in the logistic services industry has been conducted. 475 customers of a logistic service provider in the Netherlands have been asked to participate in this survey, 153 filled in the questionnaire completely. All the questions put forward at the introduction have been answered and will be discussed into detail in this chapter.
First, I will discuss the antecedents of solution effectiveness which have been analysed. Second, the relational variables that could mediate this relationship will be discussed. Finally, this discussion will end up with the effect of solution effectiveness on measures of customer feedback.
5.1 A
NTECEDENTS OF SOLUTION EFFECTIVENESS1994). The causality of the relationship between the customer antecedent operational counseling is supported by the fact that there is no significant effect when turning around this relationship (solution effectiveness as independent variable). This supports H3 which
states that the greater the operational counseling, the greater solution effectiveness and not the other way around. Also no effect of solution effectiveness on customer adaptiveness has been found in this research (rotated IV and DV). This supports H1 which states that the
greater the customer adaptiveness, the greater solution effectiveness. Although, political counseling has an effect on solution effectiveness, solution effectiveness also has a positive effect on political counseling. This undermines the causality of this relationship. A possible explanation for this fact is that when a solution is effective, a highly qualitative relationship will be established, which would increase the knowledge of political issues of Company X about the customer. All in all, a causal positive relationship between operational counseling and customer adaptiveness with solution effectiveness in a B2B relationship between buyer and seller has been detected in this research.
5.2 M
ODERATING EFFECTS OF RELATIONAL VARIABLES5.3 S
OLUTION EFFECTIVENESS AND THE EFFECT ON MEASURES OF CUSTOMER FEEDBACKThe second part of the conceptual model analyses the effect of solutions in term of measuring the customer feedback (satisfaction, loyalty and profitability). As it has been discussed earlier in the theoretical framework of this research, these measures are important in evaluating the effectiveness of solutions.
First, the effectiveness of solution effectiveness on satisfaction has been analysed. After control for the variables trust, commitment, solution experience and workforce, no evidence was found that solution effectiveness has an effect on the satisfaction of business customer in this context. Trust in a B2B relationship with Company X describes most of the variance of satisfaction. A possible explanation is that B2B customers expect an effective solution as general practice and are not less or more satisfied than customers with ineffective solutions. Another explanation is that selling an effective solution to customers is an ineffective marketing tool to make B2B customers more satisfied. This research suggests that trust is more important in satisfying B2B customers.
Secondly, this research suggests that the effectiveness of solutions has an effect on the loyalty of the customers. However, there is no evidence that this assumption is correct. Although loyalty is an important element in improving the lifetime value of customers (Berger & Nasr, 1998; Gupta & Zeithaml, 2006), solution effectiveness has no effect on loyalty of the customers of Company X. A possible explanation could be that the dependence on seller of customers of Company X is low (table 4.5, appendix II). Customers who are not dependent on one company in their need for logistic services can possibly switch easily between logistic suppliers, even when a solution is highly effective. Nevertheless, this study indicates that an effective solution is a weak marketing concept to increase loyalty of B2B customers.
on seller. A possible explanation is that only when customers are highly dependent on Company X, a high profit margin can be realised. In any other scenarios, competitiveness is too high to obtain profitable customers.
Finally, this research indicates that there is no significant relationship between solution effectiveness and measures of customer feedback. Therefore, in this context there is no evidence that an effective solution is an effective marketing effort.
6
M
ANAGERIAL
I
MPLICATIONS
It is important for general managers and marketing managers to meet the needs of the customer in order to generate customer value and finally firm value. An alternative to deliver this customer value in B2B markets is to provide a solution for the problems that business customers are facing. But what is actually a solution? This research offers a solution in four relational processes, as described by Tuli, Kohli and Bharadwaj (2007), which all aim to meet customers’ business needs. The effectiveness of a solution is determined by the extent to which these relational processes are fulfilled. This research shows to what extent these relational processes are fulfilled at Company X and it provides a guideline to improve the solutions they are delivering at the moment. This research explains that the relational processes, particularly, requirements definition and customization and integration, require improvements.
Furthermore, this research does not offer any significant evidence (after inspecting the major general relationship variables) that political, counseling, operational counseling and customer adapativeness have positive effects on the effectiveness of a solution. The exchange of information about political and operational issues between buying and selling team will stimulate the effectiveness of a solution. Since politics in organization can influence critical processes in the organizations, it is important to share this information between both parties. It is tough for both parties to identify the political manners in an organization immediately. Besides political counseling, operational counseling can also enhance solution effectiveness. Buying teams must inform the selling team about the operations of the organization. Knowing how the businesses works gives them the opportunity to fit the solution to their needs and will effectuate the solution. Besides, increasing the adaptiveness of customers will also enhance solution effectiveness. Solutions in a B2B environment are complex processes. A more adaptive customer in this complex situation will have a positive influence on the effectiveness of solutions.
7
L
IMITATIONS AND
D
IRECTIONS FOR
F
URTHER
R
ESEARCH
This section discusses the limitations of this research and provides some directions for further study. The primary limitation is the sample used in this research, which contains only the customers of Company X. The outcomes of this research would have been more universal if the customers of more than one service provider were used. The respondents that were used in this research, didn’t have the possibility to compare the services of Company X with other service providers. Besides, these customers could be just more satisfied because they are the customers of this company. Moreover, most of the customers are active in two branches (Machinery/equipment and Construction). For these reasons, the results of this research cannot be not cover all the business customers. Furthermore, to relate the outcomes of the questionnaire with other data (e.g. customer profitability), respondents were asked to fill in their company names. In other words, they could not answer the questions anonymously. It must also be noted that this research only analyzed three customer antecedents and two relational variables as possible manipulators of solution effectiveness. There could be more antecedents of solution effectiveness, for instance, supplier antecedents (contingent hierarchy, documentation emphasis, incentive externality, customer interactor stability and process articulation) as described by Tuli, Kohli and Bharadwaj (2007). Furthermore, the definition of a customer solution in this study is debatable. This research used the definition of Tuli, Kohli and Bharadwaj (2007). Tuli and their colleagues defined a solution as four relational processes, however, there is not enough evidence that this definition of a solution is sufficient to be used in all situations.
