Eindhoven, May 2015 Eindhoven University of technology

Antecedents of sales force new product adoption and the effect of adoption on sales performance.

By T.T. (Thomas) Janssens BSc Industrial Engineering for Healthcare – 2012 Student identity number 0650510

In partial fulfillment of the requirements for the degree of Master of Science In Innovation Management

Supervisors: Dr. J.J.L. Schepers, TU/e, ITEM Dr. Ir. W. van der Borgh, TU/e, ITEM Ir. J. Revet, AkzoNobel Department:

Innovation,

Technology,

Entrepreneurship

&

Marketing

-

TU/e.

- New products do not sell themselves – M. van der Borgh (2012)

1

Abstract

Purpose: Empirically investigate what factors influence new product adoption by the sales force. Design / Methodology / Approach: This research is conducted in collaboration with AkzoNobel. In total, 113 people participated, of which 97 yielded useful answers. These answers were analyzed using Partial Least Squares – Structural Equation Modeling (PLS-SEM) in SmartPLS. Findings: In the final model, 31.2% of the variance in sales performance, and 52.4% of the variance in product adoption by the sales force is explained. Analyses show that, next to product adoption, also perceived involvement and product newness directly influence the sales performance. Product perception and the control system most strongly influence the degree to which a new product is adopted. Practical implications: The findings of this research indicate that it is essential to establish a positive perception of the product amongst those who are in touch with an organization’s customers. This can be accomplished by understanding which social groups most strongly influence the product perception, or by involving salespeople in the development process of the product. Furthermore the findings implicate that, to increase product adoption, salespeople should be controlled based on their behavior. Value: This is the first study to empirically investigate factors influencing product adoption by the sales force in a selling- rather than usage situation. Subject headings: Sales performance, sales force, antecedents new product adoption, empirical, perceived involvement, product advantage, product newness, behavioral control

2

Management summary

Introduction For firms, product innovations and successfully introducing these innovations has become increasingly important to realize growth and survival (Pauwels, Silva-Risso, Srinivasan, & Hanssens, 2004). A stakeholder playing a significant role in new product success, is the sales force (e.g. Ahearne, Rapp, Hughes, & Jindal, 2010; Di Benedetto, 1999) amongst others because the sales function is a boundary spanning endeavor (Singh, 1998). Wieseke, Homburg, & Lee (2008) claim that a salesperson’s direct contact with customers makes them able to strongly affect market success of innovations. Because the sales force is so pivotal for market success, it is essential to understanding how they can affect market success. A concept that positively influences sales performance and is related to the sales force, is the extent to which a salesperson adopts a new product (Hultink & Atuahene-Gima, 2000). Although the effect of adoption on sales performance has been investigated before (both conceptually and empirically), empirical research on factors influencing product adoption by the sales force does not exist. The purpose of the current study therefore is to understand what factors influence product adoption by the sales force, and empirically investigate what factors are considered most influential. Literature review Fundamental for the current research are the relationships between attitudes, intentions and actual behavior. Attitudes influence intentions, and indicative of actual behavior intentions in turn capture the motivational factors that influence behavior (see Theory of Reasoned Action (TRA) by Fishbein & Ajzen (1975) and the Theory of Planned Behavior (TPB) by Ajzen (1991)). Two of the most important adoption theories are the Technology Acceptance Model (TAM, Davis (1985)), which builds on the TRA and TPB, and the Innovation Diffusion Theory (IDT, Rogers (1962)). These adoption theories both investigate what factors influence adoption. While the TAM approaches adoption from a psychological point of view and investigates how perceptions of a technology influence attitudes towards using the technology, the IDT focusses on product characteristics and investigates how product characteristics influence the degree to which an innovation is adopted. Although the TAM and IDT both investigate factors influencing adoption, a fundamental difference exists between their perspective, and the perspective of the current research. Most importantly, a review of literature on product adoption suggests that generally, adoption is investigated in a situation in which multiple roles were united in one individual (most importantly the decision maker and the user). For the TAM, this for example means that the ‘unit of adoption’ is an individual who is both a decision maker and a user. For the current research however, we are particularly interested in adoption by an individual (a salesperson) who has to assess the value of a new product for someone else, namely his/her customers (the user). This means that this salesperson adopts a new product being only an influencer in a customer’s decision making process. According to Hultink & Atuahene-Gima (2000), this perspective did receive little attention.

3

Because this fundamental difference exists, findings of the ‘traditional’ adoption literature might not apply. We therefore define sales force new product adoption (SFNP adoption) as the interaction between the degree to which salespeople accept and internalize the goals of a new product (i.e. commitment), and the extent to which they work smart and hard (i.e. effort) to achieve these goals’ (Atuahene-Gima, 1997, p.500). This definition includes an attitudinal (commitment), and behavioral (effort) dimension. But what factors influence SFNP adoption? Theory & Model To understand what factors might influence SFNP adoption, we have to investigate the choices and decisions individuals make to accept or reject an innovation, and how these choices and decisions are made. In line with the adoption theories mentioned above, we will therefore investigate factors influencing a salesperson’s attitude towards an innovation, and how characteristics of the new product influence adoption. We expect that different categories of antecedents influence the degree to which salespeople adopt a new product. To determine these antecedents, research streams, focusing on the subjects below, are reviewed and effect sizes are compared: (i) The link between attitudes, intentions, and behavior (e.g. Technology Acceptance Model (Davis, 1985; Venkatesh & Davis, 2000) and Unified Theory of Acceptance and Use of Technology (Venkatesh, Morris, Davis, & Davis, 2003)), (ii) Characteristics of the product (e.g. Innovation Diffusion Theory (IDT) by Rogers, 1962), (iii) New product selling (e.g. e.g. Sujan, Kumar & Weisz, 1994; Anderson & Oliver, 1987). Based on this review, we selected five antecedents: ‘product advantage’ and ‘product newness’ (both product related); ‘subjective norms’ as a social factor; ‘goal orientation’ as a personal trait of the salesperson; and ‘control system’ as a process-related factor that influences the adoption process. Additionally, when people face events that bring potential change to their routines -of which a new product introduction is a perfect example-, an individual’s resistance to change dictates how they will respond (Oreg, 2003). A concept that shows to decrease resistance is the degree to which people perceive to be involved in the decision making process. Additionally, another stream of literature elaborates on the importance of a salesperson’s involvement during a phase in which important decisions are made concerning product specifications: the development phase (Avlonitis & Gounaris, 1997; Joshi, 2010; Judson et al., 2006). Because involvement in the decision making process has been linked to resistance, and involvement during the development phase has a profound impact on product performance, a sixth antecedent is added to this research: ‘perceived involvement’. This leads to the conceptual model represented on the next page. Methodology To investigate the extent to which these antecedents influence the adoption of a new product, a survey was conducted. Self-reported measures for all concepts were used, including the sales performance. To gather the data necessary to conduct the analysis, AkzoNobel’s customer-contact employees in English and Dutch-speaking countries were 4

invited to participate. After collection, the data was analyzed using Partial Least SquaresStructural Equation Modeling (PLS-SEM).

Results & Discussion In total, 113 people participated, which yielded 97 useful surveys. The results show that, with the selected antecedents, the final model is able to predict 31,2% of the variance in sales performance, while 52,4% of the variance in SFNP adoption is explained with the selected antecedents (see model below). In total, six out of ten hypotheses are supported by these data, while one hypothesis is marginally supported. Additional analysis show that one significant moderation effect occurred: the relationship between product advantage and adoption is moderated by the age of a salesperson such that a higher age weakens the relationship between product advantage and product adoption.

Note: significance levels ***p1.65)

5

Last, based on apparent differences between commitment and effort in the correlation matrix, it was investigated how exactly these constructs differ. First, commitment shows the strongest similarity with adoption regarding the amount of variance explained. Another interesting finding is that learning orientation relates only significant to commitment and does not influence effort while performance orientation has no significant effect on either one of the two dimensions of adoption. This conflicts with previous research of for example Sujan et al. (1994). We furthermore conclude that, from a theoretical perspective, and in line with Fu et al. (2010), it might be hypothesized that including intentions would lead to a better conceptualization of what ‘SFNP adoption’ intends to represent. Including intentions would lead to conceptualization that is more in line with behavioral models such as the TRA and the TAM. A comparison between the findings of the current research with the findings of the TRA and an extension of the TAM, TAM2 (Venkatesh & Davis, 2000) show strong similarities between adoption, the TRA, and the TAM. However, a comparison between the commitment and effort show that commitment most strongly relates to adoption, the findings of the TRA, and findings of the TAM. This suggests that a proper conceptualization of product adoption by the sales force has yet to be found. Implications To our knowledge, this research is the first to empirically investigate the influence of these antecedents of product adoption. It therefore deepens our understanding of the new product selling process and provides valuable theoretical insights concerning this process. In a selling- rather than usage context, findings from the current research are similar to the findings of the TAM. In line with the agency theory (e.g. Anderson & Oliver, 1987), behavioral control positively influences product adoption. These findings indicate that the type of control system not only has a profound impact once a product is adopted (e.g. Hultink & Atuahene-Gima, 2000; Hultink et al., 2000), but control systems also directly influence adoption. The current research furthermore adds to existing literature the introduction of involvement. While adoption is conceptualized as pro-change behavior and involvement has been shown to foster pro-change behavior, no direct effect is found between involvement and adoption in the current research. However, involvement influences sales performance both directly and indirectly via product newness, which is an important theoretical implication. Last, we investigated how our conceptualization of adoption relates to behavioral theories such as the TRA, and adoption theories such as the TAM. From a theoretical perspective, and in line with Fu et al. (2010), it might be hypothesized that including intentions would lead to a better conceptualization of what ‘SFNP adoption’ intends to represent. A comparison between commitment and effort lead to the same conclusion, which suggests that a proper conceptualization of product adoption by the sales force has yet to be found. For managers, several findings are worth mentioning. It is essential to establish a positive product perception amongst those who are in touch with an organization’s customers. Combining the influence of being involved and the impact of the social environment, managers should do several things. Because most organizations are too big to involve everyone, managers should find out which individuals are decision makers and/or opinion leaders. These opinion leaders (within the different reference groups) should be involved 6

such that they influence the product perception of others. These opinion leaders themselves thus are the ‘social influences’ described under subjective norms. How can the product perception of these opinion leaders be shaped? Importantly, managers and product developers should emphasize the importance of the opinion leaders’ input in the development process. Additionally, feedback should be provided on what has been achieved using their input. Last, once a product has been developed and is being tested, these opinion leaders should again be involved in this testing phase. Importantly, employees should be trained in how to gather customer information, guidelines for the collection process- and reporting methods for the feedback provided by customers should be established (Judson et al., 2006). Additionally, this research shows that focusing on procedures positively influences the degree to which salespeople adopt a new product, which means that managers should actively guide a salesperson, evaluate their behavior, and provide feedback on how to accomplish performance goals, while appropriate flexibility should be provided (Ahearne et al., 2010). Last, because perceived involvement is negatively influenced by learning orientation, managers should make a conscious trade-off between demanding a salesperson’s time on learning to improve their skills or being involved in the development process.

7

Preface

This master thesis project consists of two distinct parts. In order to theoretically understand the research problem, the first part elaborated on literature about product adoption and new product selling. Based on this review, we concluded that although the importance of the sales force is acknowledged, empirical research investigating antecedents of product adoption, and the effect of product adoption on sales performance, is scarce. The current report elaborates on the second part of the graduation project. It describes a research conducted in collaboration with AkzoNobel, in particular the Global Color Marketing (GCM) department, part of the Vehicle Refinishes Business Unit. Throughout the organization, employees agree that, when the internal organization -from the global department to the people that actually have to sell a product to customers- is convinced of a new product and ‘believes in it’, this product will be successful in the market. Because GCM experiences some difficulties with engaging different layers of the organization, they are looking for ways to improve product launches by increasing the degree to which salespeople adopt a new product. This second part of this graduation project implements the general descriptive findings of the first part and empirically investigates antecedents of new product adoption by the sales force. Furthermore the link between adoption and selling performance is investigated.

8

Acknowledgement

Eindhoven, May 2015 Nine months of dedication were needed to deliver the report you are currently reading. Nine months that, when looking back, turned out to be very different than I expected up front, both from a scientific - as from a personal point of view. Finally, being able to say: I’m done, this report marks the end of a very interesting chapter of my life. Completion of this thesis means that I will finish my Master Innovation Management, that I will finish my time as a student, and that I will start with the next chapter: the working life. Two things influenced the choice to focus this graduation thesis on product adoption. Besides AkzoNobel’s desire to increase the sales performance by increasing the degree to which selling personnel is ‘engaged’, a strong fit exists between this project, my personal interest in new product selling processes, and the areas of expertise of my supervisors from the university. During this project, two supervisors from the Technical University of Eindhoven were involved to maintain a scientific level: Dr. J.J.L. Schepers, and Dr. Ir. W. van der Borgh. Jeroen, I want to thank you for your attitude towards me and my work. I did learn a lot about myself due to your questions, answers, and other forms of guidance. Furthermore, I am thankful for your flexibility concerning my planning. Michel, I would like to thank you for answering my questions and the substantive feedback on my work. You too have helped me learn a lot about myself. Of course, I want to thank AkzoNobel as well. I am very happy that I had the possibility to do my graduation project there. I am very thankful for the support this project received within the organization, the continuous enthusiasm of all the people I have worked with, and the support of those who participated in my research. Furthermore, I am thankful for the flexibility of those around me when my personal life had an impact on the project. Most importantly, I want to thank Ir. Jacqueline Revet. Thank you for introducing me to the Global Color Marketing department, your questions, the brainstorm sessions we have had, your support, and your trust in me. Not only on a professional and academic level you have helped me a lot, also on a personal level I have learned a lot from you. I too want to thank all my friends. This includes colleagues from the university, from Dispuut Jupiter, Jaarclub NoLimit, Vienna, and anywhere else. You guys made my time as a student an awesome experience. Finally, I want to express my gratitude to my family. Papa, Mama, Hanne, en Jante. Bedankt voor de steun die ik van jullie heb mogen ontvangen tijdens deze lastige maanden, en alle jaren daarvoor. Zonder jullie ondersteuning was dit allemaal niet mogelijk geweest en mijn dankbaarheid is groot. Ik ben enorm blij en gelukkig met een gezin zoals het onze!

9

Contents

Abstract ................................................................................................................................. 2 Management summary ......................................................................................................... 3 Preface................................................................................................................................... 8 Acknowledgement ................................................................................................................ 9 List of figures and tables ..................................................................................................... 13 List of figures ................................................................................................................... 13 List of tables..................................................................................................................... 13 1

2

Introduction .................................................................................................................. 14 1.1

Research context ................................................................................................... 14

1.2

Scope ..................................................................................................................... 15

1.3

Purpose of this thesis ............................................................................................ 15

1.4

Research question ................................................................................................. 16

1.5

Outline ................................................................................................................... 16

Theoretical background & model ................................................................................. 17 2.1

Defining Sales Force New Product Adoption (SFNP adoption) ............................. 17

2.2

Theory .................................................................................................................... 18

2.2.1

Overview of relevant literature ...................................................................... 18

2.2.1.1

Literature on intentions and behavior, and adoption ................................ 18

2.2.1.2

Literature on new product selling .............................................................. 19

2.2.1.3

Fundamental difference ............................................................................. 19

2.2.2 2.3

Determining antecedents............................................................................... 20

Hypothesis ............................................................................................................. 21

2.3.1

SFNP Adoption on Sales Performance ........................................................... 21

2.3.2

Product Advantage ......................................................................................... 21

2.3.3

Product Newness ........................................................................................... 22

2.3.4

Social Influence .............................................................................................. 23

2.3.5

Goal Orientation ............................................................................................. 24

2.3.6

Control system ............................................................................................... 25

2.3.7

Perceived Involvement................................................................................... 26 10

3

2.4

Conceptual model.................................................................................................. 27

2.5

Conclusion ............................................................................................................. 28

Methodology................................................................................................................. 29 3.1

Method .................................................................................................................. 29

3.1.1

4

3.2

Sample ................................................................................................................... 30

3.3

Measures ............................................................................................................... 30

3.4

Data Analysis.......................................................................................................... 32

Data Analysis and Results ............................................................................................. 33 4.1

Examination of data............................................................................................... 33

4.1.1

Missing values ................................................................................................ 33

4.1.2

Outliers ........................................................................................................... 33

4.1.3

Assumption testing ........................................................................................ 34

4.1.4

General information about participants ........................................................ 34

4.2

5

Increasing response rates and overcoming biases ........................................ 29

Factor analyses ...................................................................................................... 34

4.2.1

Exploratory factor analysis on product perception ....................................... 35

4.2.2

Confirmatory factor analysis .......................................................................... 36

4.3

Correlations, mean and standard deviation .......................................................... 39

4.4

Model results ......................................................................................................... 40

4.5

Additional analysis ................................................................................................. 41

4.5.1

Moderating effects ......................................................................................... 41

4.5.2

Additional direct paths ................................................................................... 42

4.5.3

Difference between Commitment and Effort ............................................... 42

Discussion and conclusion ............................................................................................ 44 5.1

Discussion of results .............................................................................................. 44

5.2

Implications ........................................................................................................... 48

5.2.1

Theoretical Implications ................................................................................. 48

5.2.2

Managerial Implications ................................................................................. 49

5.3

Limitations and future research ............................................................................ 50

References ........................................................................................................................... 53 Appendices .......................................................................................................................... 60 A.

Measures ............................................................................................................... 61 11

B.

Test for normality – Skewness & Kurtosis ............................................................. 64

C.

Exploratory factor analysis .................................................................................... 65

D.

Discriminant validity .............................................................................................. 66 1.

Fornell – Larcker criterion .................................................................................. 66

2.

Cross-loadings .................................................................................................... 66

E.

Analysis for moderation ........................................................................................ 68

F.

Additional paths investigated ................................................................................ 70

G.

Early versus late response ..................................................................................... 71

H.

Notifications .......................................................................................................... 72

I.

1.

Pre-notification to customer-contact employee ............................................... 72

2.

Pre-notification to sales manager ...................................................................... 73

3.

Invitation to customer-contact employee ......................................................... 74

4.

Notification invitation to sales manager ........................................................... 75

5.

First reminder to customer-contact employee ................................................. 75

6.

First reminder to sales manager ........................................................................ 76

7.

Second reminder to customer-contact employee............................................. 76

8.

Second reminder to sales manager ................................................................... 77

List of antecedents mentioned in literature ............................................................. 78

12

List of figures and tables

List of figures Figure 1: Conceptual model – antecedents of sales force new product adoption............. 27 Figure 2: Model results for testing hypothesis ................................................................... 40 Figure 3: Moderation Learning Orientation ........................................................................ 41 Figure 4: Moderation Age ................................................................................................... 41 Figure 5: Additional direct paths ......................................................................................... 42

List of tables Table 1: Overview of hypothesis ......................................................................................... 27 Table 2: General information about participants ............................................................... 34 Table 3: Factor matrix product perception ......................................................................... 35 Table 4: Factor matrix product perception after case deletions (N=97) ............................ 36 Table 5: Confirmatory factor analysis, reliability- and validity tests................................... 37 Table 6: Correlations, Mean and Standard Deviation ......................................................... 39 Table 7: Overview models for testing hypotheses.............................................................. 40 Table 8: Moderation effects ................................................................................................ 41 Table 9: Difference between Commitment and Effort ....................................................... 43

13

1

Introduction

For firms, product innovations and successfully introducing these innovations has become increasingly important to realize growth and survival (Pauwels, Silva-Risso, Srinivasan, & Hanssens, 2004). A stakeholder playing a significant role in new product success is the sales force (e.g. Ahearne, Rapp, Hughes, & Jindal, 2010; Di Benedetto, 1999). The sales function is a boundary spanning endeavor (Singh, 1998). This means that a salesperson is the last actor within the boundaries of a firm and will sell new products to customers or end-user, actors that find themselves outside the boundaries of the firm. Because of this, Wieseke, Homburg, & Lee (2008) claim that a salesperson’s direct contact with customers makes them able to strongly affect market success of innovations, particularly in a business-to-business market (Fu, Richards, Hughes, & Jones, 2010). In line with this, some claim the sales force to be the most important communication vehicle for launching new products (Abramovici & Bancel-Charensol, 2004; Moriarty & Kosnik, 1989), which particularly holds in a market that can be characterized as high-tech (Samli, Wirth, & Wills, 1994). Salespeople facilitate customer adoption of innovations (Anderson & Robertson, 1995), amongst others because they influence a customer’s perception (Ahearne et al., 2010). Because the sales force is so pivotal for new product success, understanding what influences their selling performance is essential. A concept related to a salesperson which positively influences the sales performance, is the degree to which a salesperson adopts a new product (Hultink & Atuahene-Gima, 2000). Although adoption of new products by salespeople has been investigated extensively, an important distinction should be made. It is essential to distinguish between new product adoption by salespeople as end-users, and new product adoption by salespeople who then have to sell the innovation. We are particularly interest in adoption by an individual (a salesperson) who has to assess the value of a new product for someone else, namely his/her customers (the user). This means that this salesperson adopts a new product being only an influencer in a customer’s decision making process. According to Hultink & Atuahene-Gima (2000), this perspective did receive little attention. Although Atuahene-Gima (1997) conceptually investigated factors that might influence product adoption by a salesperson, to our knowledge, we are the first to empirically investigate this.

1.1 Research context This research is conducted in collaboration with AkzoNobel. AkzoNobel is a Dutch multinational specialized in paints, coatings and specialized chemicals. The business unit at which the research is held, the Strategic Business Unit Vehicle Refinishes (SMU-VR) is located in Sassenheim. At the site of this business unit, departments such as R&D, production and marketing for both the automotive and aerospace market are situated. The department at which the research is conducted –Global Color Marketing (GCM)- is responsible for the organization of the process that goes from finding new colors at car manufacturers until the actual application of the right color at body shops. GCM’s activities therefore include the development and launch of new color tools and color selection 14

software, as well as making sure that (information about) new products, tools, and software packages (hereafter referred to as ‘new product’) diffuse through the organization successfully. For GCM, this means they predominantly deal with internal customers. In all layers of the organization, employees acknowledge the importance of stakeholders adopting a new product. Throughout the organization, employees agree that, when the internal organization -from the global department to the people that actually have to sell a product to customers- is convinced of a new product and ‘believes in it’, this product will be successful in the market. During the launch of new products, one group in particular is perceived to be crucial: those who are actually in touch with the customer: technical consultants (TCs) and salespeople, hereafter all referred to as ‘sales force’. Currently, the sales force is not fully convinced of a product that is soon to be launched, and has a hard time selling a product that is ‘not so good’. Adoption of the new products by the sales force can thus be considered a crucial first step in successfully selling these products. For this reason, GCM is looking for ways to improve product launches by increasing the degree to which salespeople adopt a new product. This study presents a first attempt to identify what specifically influences adoption by the sales force and offers practical guidance for people responsible for new product launches and sales managers.

1.2 Scope Because the research is conducted at AkzoNobel, the results will be applicable for a hightech market. Furthermore, this research will focus on new products, not existing ones. Next to that, a business-to-business market is investigated. Importantly, we will investigate the link between product adoption and sales performance. This will be done at the level of an individual, rather than a team-level, all the more because a salesperson has to sell new products individually. Because we are investigating the link between product adoption and sales performance, we will specifically look at employees that are able to influence a customer’s perception of a new product.

1.3 Purpose of this thesis Since product adoption by the sales force is positively related to the sales performance, it is essential to know how to increase the degree to which a salesperson adopts a new product. Additionally, an important finding of the first part of this thesis is that research on factors influencing Sales Force New Product Adoption (SFNP adoption) is scarce (e.g. Atuahene-Gima, 1997; Hultink & Atuahene-Gima, 2000). Furthermore, since GCM is looking for ways to improve product launches by increasing the degree to which salespeople adopt a new product the purpose of the current study from a theoretical and practical point of view is to understand what factors influence new product adoption by the sales force, and to empirically investigate the influence of these antecedents.

15

1.4 Research question To better understand product adoption and its influence on sales performance, the following research question is set: How can the sales performance be increased through product adoption by the sales force? In order to answer this main research question, a set of sub-questions is formulated. The following sub-questions will be examined: 1) How does sales force new product adoption differ from product adoption by endusers and how can it be defined? 2) From a theoretical point of view, what are the most important antecedents of adoption? 3) How do these antecedents influence adoption? The remainder of this report describes the endeavor to answer these questions.

1.5 Outline The remainder of this report covers four chapters. Chapter 2 will first provide a brief overview of literature relevant for the current research. After that, the hypothesized relationships are explained and the research model is presented. Chapter 3 then describes the methodology that was used to conduct this research. Most importantly, a survey was conducted to gather data on product adoption by salespeople. Chapter 4 describes the analysis of the data, and displays the basic and the extended model. Last, Chapter 5 discusses the results of the analysis, provides theoretical and practical implications, discusses the limitations of this research, and provides suggestions for future research.

16

2

Theoretical background & model

In this chapter, we will elaborate on the theoretical background that supports this research. First the focal concept of the current research, Sales Force New Product Adoption (SFNP adoption), is defined. Then, it is described how antecedents that are included in the research model are selected. Thereafter, the hypothesized relationships will be elaborated on. Finally, the research model will be presented.

2.1 Defining Sales Force New Product Adoption (SFNP adoption) From a company’s perspective, the selling performance and the profit generated from this selling activity is the point of interest. In an empirical study, Hultink & Atuahene-Gima (2000) show product adoption by the sales force to be positively related to selling performance. Therefore, the focal concept for this research is Sales Force New Product Adoption (SFNP adoption). Although the name implies adoption at a team level, this concept concerns new product adoption at the level of an individual salesperson. The current research will not focus on adoption on a team level. Atuahene-Gima (1997) defines SFNP Adoption as “the interaction between the degree to which salespeople accept and internalize the goals of a new product (i.e. commitment), and the extent to which they work smart and hard (i.e. effort) to achieve these goals” (p.500). According to this author, commitment occurs when a person accepts the new product and by that is emotionally committed to make it a success. It concerns an attitude towards the new product. The more behavioral utterance of adoption, effort, covers the activities to achieve desired results. Different levels of effort occur due to differences in a salesperson’s force, energy, persistence, and the intensity with which these activities are executed. Note that effort in this way has a positive connotation. It covers a more motivational, rather than a demanding force. Innovation adoption literature conceptualizes adoption as functional, pro-change behavior (Atuahene-Gima, 1997). This scholar argues that if a salesperson identifies himself with - and internalizes the goals of- an innovation, this empowers adoption. In line with previous research, adoption thus is a sign that goals of the new product are compatible with the goals, needs and past experiences of the adopter (Atuahene-Gima, 1997; Rogers, 1962). When a new product is introduced, the goals of this new product thus might or might not be compatible with a salesperson’s goals. If no congruence exists, most likely a salesperson is still required to sell a new product because of his/her job requirements. In this case, effort is devoted to selling a new product, but this salesperson is likely to be less successful due to a lack of commitment. For this reason, Atuahene-Gima (1997) argues that adoption is represented by both commitment and effort because the separate constructs do not constitute adoption. While both an interaction effect and a reflective-formative construct could conceptualize adoption, which representation of adoption covers our interest best? Hultink & AtuaheneGima (2000) state that, from a mathematical point of view, a multiplication of commitment and effort (interaction) emphasizes the difference between those who adopt (thus high on commitment and effort), and those who do not adopt (low on either one of the scales), because this multiplication introduces desired variance into the data set. “Therefore, the 17

multiplicative rather than additive aggregation [reflective-formative construct] reflects the theoretical assertion that greater than linear returns in terms of performance in new product selling accrue from salespersons who show both effort and commitment to a new product” (Hultink & Atuahene-Gima, 2000, p.442).

2.2 Theory Now we have defined product adoption by the sales force, we can investigate what factors might influence it. 2.2.1 Overview of relevant literature Innovation adoption literature investigates the choices and decisions individuals make to accept or reject an innovation. Factors that might influence the decision are formed over time (Straub, 2009) and can relate to the process and to beliefs and attitudes regarding the product. Furthermore, cognitive, emotional, social, and contextual factors can influence adoption (Burkhardt, 1994; Kraut, Rice, Cool, & Fish, 1998; Straub, 2009; Zablah, Chonko, Bettencourt, Allen, & Haas, 2012). The above suggests that many different factors do influence the decision to (not) adopt. The research model developed in this chapter follows this line of reasoning, and is composed of several ‘categories of antecedents’. These categories and antecedents are inspired by research streams focusing on: (iv)

(v) (vi)

The link between attitudes, intentions, and behavior, e.g. Technology Acceptance Model (Davis, 1985; Venkatesh & Davis, 2000), and Unified Theory of Acceptance and Use of Technology (Venkatesh, Morris, Davis, & Davis, 2003), Characteristics of the product, e.g. Innovation Diffusion Theory (IDT) by Rogers (1962), New product selling (e.g. Anderson & Oliver, 1987; Atuahene-Gima & Li, 2002; Sujan et al., 1994).

2.2.1.1 Literature on intentions and behavior, and adoption Fundamental for the current research are the relationships between attitudes and intentions, and intentions and actual behavior. Two widely accepted and extensively tested theories describe these relationships: the Theory of Reasoned Action (TRA) by Fishbein & Ajzen (1975) and the Theory of Planned Behavior (TPB) by Ajzen (1991). These theories state that intentions are influenced by attitudes. Indicative of actual behavior, intentions in turn capture the motivational factors that influence behavior. Building on these theories, the Technology Acceptance Model (TAM) by Davis (1985) and its successors such as the Unified Theory of Acceptance and Use of Technology (UTAUT) by Venkatesh et al. (2003), investigated how perceptions of a technology influence attitudes towards using the technology and how these perceptions are influenced by, for example, social factors. Interestingly, because SFNP adoption comprises an attitudinal and a behavioral dimension, strong similarities exist between the TRA and TBP, and SFNP adoption. Later on in this thesis, we will elaborate on how adoption relates to these two theories. Another relevant stream of literature is grounded in the Innovation Diffusion Theory (IDT) by Rogers (1962). This widely accepted theory states that certain characteristics of an innovation influence the rate of adoption. Although TAM finds its roots in psychology while the IDT originated from a sociological background, strong similarities do exist between 18

fundamental conceptions of these streams. Most importantly, both conclude that adoption of an innovation (or acceptance of a technology) is particularly influenced by the degree to which it is perceived as beneficial and easy to use. 2.2.1.2 Literature on new product selling Because we investigate the link between adoption and performance in the context of selling new products, new product selling literature is investigated as well. In the personal selling domain, this stream finds that a salesperson’s characteristic, related to goals that people pursue in achievement situations, influences the way in which new products are sold (e.g. Sujan et al., 1994). Furthermore, this stream emphasizes the importance of incentivized salespeople. It argues that, based on factors related to perceptions of uncertainty and risk (e.g. Anderson & Oliver, 1987; Atuahene-Gima & Li, 2002), different types of control should be applied to monitor, direct, and evaluate the behavior of salespeople. Furthermore, it provides insights on how to incentivize salespeople. 2.2.1.3 Fundamental difference It is however of paramount importance to acknowledge that there is a fundamental difference between the perspective of the current research and the literature streams described above. When investigating different roles within a decision-making process, this difference becomes clear. Within a decision-making process, Engel, Blackwell, & Miniard (1995) proposed five different roles: (i) the gatekeeper; the individual who recognizes a need and is typically the most influential, (ii) influencers; individuals whose opinion is sought and may affect the decision criteria, (iii) the decision maker; the individual that makes the ultimate decision, (iv) the buyer; individual that physically purchases the product, and (v) the user; individuals that consume the product. A review of literature on product adoption suggests that, generally, adoption was investigated in a situation in which multiple roles were united in one individual (most importantly the decision maker and the user). In the case of the TAM, this for example means that the ‘unit of adoption’ is an individual who is both a decision maker and a user. This individual therefore decides for him-/herself whether or not he/she will like the product, adopts it, and is ultimately going to use it. For the current research however, we are particularly interested in adoption by an individual (a salesperson) who has to assess the value of a new product for someone else, namely his/her customers. This means that this salesperson adopts a new product being only an influencer in a customer’s decision making process. This research therefore distinguishes itself by specifically focusing on adoption by individuals who have to sell a new product rather than to use it. To our knowledge, little research has been devoted to this perspective. Although some scholars investigated outcomes of SFNP adoption (e.g. Hultink & Atuahene-Gime, 2000; Hultink, Atuahene-Gima, & Lebbink, 2000), empirical evidence on antecedents of SFNP adoption does not exist. Although the perspective we take fundamentally differs from the perspective of, for example, the TAM and IDT, the literature streams, theories, and models described above still provide a good starting point for investigating antecedents of SFNP adoption. Note that together with paragraph 2.1, this answers sub-question 1. The next paragraph explains how the concepts of the research model are selected.

19

2.2.2 Determining antecedents While the streams mentioned above discuss a plethora of antecedents that (might) influence adoption, we are particularly interested the most influential ones. To determine influential antecedents, a list was created of antecedents mentioned in innovation adoption-, and new product selling literature (see Appendix I). Thereafter, the effect sizes of these concepts are reviewed. Based on this review, five concepts were selected: (i) (ii) (iii) (iv) (v)

Related to the benefits of the product, hereafter ‘product advantage’. Related to the compatibility of a new product, hereafter ‘product newness’. Related to the social environment in which salespeople operate, hereafter ‘subjective norms’. Related to the underlying goals people pursue in achievement situations, hereafter ‘goal orientation’. Related to the organization’s set of procedures for monitoring, directing, evaluating, and compensating its employees, hereafter ‘control system’.

As mentioned before, adoption is conceptualized as functional, pro-change behavior by the adopter (Atuahene-Gima, 1997). It is known that individuals differ in their reaction to change (Van Dam, Oreg, & Schyns, 2008). According to social psychology for example, when people face events that bring potential change to their routines -of which a new product introduction is a perfect example-, an individual’s resistance to change dictates how they will respond (Oreg, 2003). Furthermore, a lack of commitment implies innovation resistance (Gatignon & Roberston, 1989; Rogers, 1962). It therefore seems important to incorporate a concept that influences the degree to which people resist to change. A concept that shows to decrease change resistance is the degree to which people perceive to be involved in the decision making process. Another stream of literature elaborates on the importance of a salesperson’s involvement during a phase in which important decisions are made concerning product specifications: the development phase (Avlonitis & Gounaris, 1997; Joshi, 2010; Judson et al., 2006). Avlonitis & Gounaris (1997) state that, because a salesperson has a thorough understanding of customer needs, companies should use this to develop a competitive advantage. In line with this, Judson et al. (2006) claim that “salespeople are preeminent among the individual-level drivers of product modifications within organizations (p.94)”, and should therefore be involved in the development process, when these modifications are made. Joshi (2010) investigated the effect of different influence strategies adopted by a salesperson, how these strategies foster compliance by decision makers, and empirically show that involvement in the development phase positively influences the performance of a new product in the market. Although these articles all emphasize the importance of involving salespeople in the development process, the link between involvement and product adoption has not been investigated yet. Because involvement in the decision making process has been linked to resistance, and involvement during the development phase has a profound impact on product performance, this research will therefore incorporate involvement, and a sixth antecedent in the research model is: (vi)

Related to the perception of being involved in the development process, hereafter ‘perceived involvement’.

20

Having decided which antecedents might influence SFNP adoption, paragraph 2.3 will discuss how adoption influences sales performance, and how the various antecedents are expected to influence adoption.

2.3 Hypothesis 2.3.1 SFNP Adoption on Sales Performance Atuahene-Gima (1997) defined sales performance as “the degree to which a salesperson meets the sales targets, customer use targets, and other objectives set for the product” (p.510). Paragraph 2.1 shows SFNP adoption to be conceptualized by two dimensions: commitment and effort. Atuahene-Gima (1997) argues that a combination of these two dimensions yields greater impact. However, the question rises how adoption might influence sales performance. Because commitment and effort together conceptualize SFNP Adoption, it is very likely these two dimensions both influence sales performance, whether dependently or independently. Many studies find positive relationships between commitment or effort, and performance (e.g. Ahearne et al., 2010; Atuahene-Gima & Micheal, 1998; Brown & Peterson, 1994; Fu, Richards, & Jones, 2009; Jaramillo, Mulki, & Marshall, 2005; Krishnan, Netemeyer, & Boles, 2002; Leong, Randall, & Cote, 1994; Mathieu & Zajac, 1990; Sujan, Weitz, & Kumar, 1994). Others show that selling intentions relate positively to the performance of a new product (Fu, Jones, & Bolander, 2008), and the growth rate of product sales (Fu et al., 2010). Most importantly we build on previous findings of Hultink & Atuahene-Gima (2000). Following their line of reasoning, the first hypothesis is: H1: SFNP Adoption is positively related to Sales Performance. 2.3.2 Product Advantage Many researchers have investigated the influence of certain product characteristics on the rate of adoption. According to the IDT by Rogers (1962), the degree to which an innovation is perceived as better than the idea it supersedes is one of the most important predictors of adoption and diffusion. In line with this, a central concept in the TAM (Davis, 1985) is perceived usefulness (PU), defined as "the degree to which an individual believes that using a particular system would enhance his or her job performance" (p.26). This author found a strong positive relationship (0.79) between the PU of a technology and the intention to use this technology. Later, scholars acknowledged strong similarities between relative advantage and PU (e.g. Davis, Bagozzi, & Warshaw, 1989; Moore & Benbasat, 1991). Importantly, these literature streams agree that the advantage of a product strongly influences the degree to which people intend to use a product and thus adopt it. Ample empirical evidence confirms the positive link between a product’s advantage and behavioral intentions (e.g. Lu, Yao, & Yu, 2005; Schepers & Wetzels, 2007; Venkatesh et al., 2003). In a sales context, many support the general view of a positive relationship between PU and usage intentions (e.g. Robinson Jr, Marshall, & Stamps, 2005) or adoption (e.g. Homburg, Wieseke, & Kuehnl, 2010; Schillewaert, Ahearne, Frambach, & Moenaert, 2005). In new product selling literature, the term ‘product advantage’ is used, defined by Atuahene-Gima (1996) as ‘the benefits that customers get from the commercial entity that is the outcome of the innovation process’ (p.95). In the remainder of this research, this term will be used.

21

The current research however does not investigate the link between adoption and use. As explained in paragraph 2.2.1.3, we are interested in a situation in which the salesperson (the adopter) has to assess the advantage of a new product for someone else, namely his customer (the user). Different from the perspective of the TAM and IDT, the adopter and the user of a new product therefore are not the same individual. Due to this subtle difference, the findings of the TAM and IDT might not apply here. Apart from Ahearne et al. (2010), who investigated the influence of product perception on the effort devoted to selling it, to our knowledge the influence of product advantage, and its relationship with adoption has not been investigated yet. How can the advantage of a product lead to adoption by a salesperson? Because believes, attitudes and behavior are related, it seems likely that if a salesperson believes a product offers substantial benefits for a customer (enhance effectiveness, increase performance, more cost effective, etcetera), this salesperson better understands the added value of the new product, therefore believes it is easier to sell a product, and his/her attitude towards that product will be more positive. Attitudes in turn influence behavioral intentions (Fishbein & Ajzen, 1975). Approaching adoption from a TRA and TPB perspective, a positive link would thus be expected between product advantage (belief about the product) and adoption (attitude and behavior). Building on institutional economics and psychological literature however, Ahearne et al. (2010) state that salespeople have to allocate their effort across a portfolio of (new) products. They argue that, if salespeople perceive a new product to offer substantial benefits, they “may believe that such a new product can ‘sell itself’ with little effort on their part (p.766)”. Indeed, their findings show a negative link between product perception and the behavioral dimension of adoption: effort devoted to selling a new product. Although the effort devoted to selling the new product might decrease, the salesperson is still obliged to sell the new product. We however expect that the product advantage will positively influence the commitment towards the new product to a greater extent than the effort devoted to selling it will decrease. Since adoption is conceptualized as the interaction between these two, a positive effect is expected from a salesperson’s perception of the product’s advantage and adoption. Our second hypothesis therefore is: H2: Product Advantage is positively related to SFNP Adoption. 2.3.3 Product Newness A central construct in the TAM is perceived ease of use (PEoU). Davis (1985) defined PEoU as "the degree to which an individual believes that using a particular system would be free of physical and mental effort" (p.26). The effect of this construct on behavioral intention and adoption too has been investigated extensively and tested empirically (e.g. Davis, 1985; Lu et al., 2005; Schepers & Wetzels, 2007; Venkatesh & Davis, 2000; Venkatesh et al., 2003). Most articles test PEoU in combination with the construct covered in the previous paragraph. In a sales context, empirical support for the TAM can be found as well. Robinson Jr et al. (2005) find support for the positive relationship between PEoU and behavioral intentions, while others (e.g. Homburg et al., 2010; Schillewaert et al., 2005) show that PEoU relates positively to adoption by the sales force. In line with new product selling literature, we will use the term ‘product newness’. Two perspectives on newness can be distinguished: newness to the firm and newness to the market. In the current research, 22

we will investigate the latter. Atuahene-Gima (1996) defined product newness as ‘the extent to which an innovation is compatible with the experiences and consumption patterns of customers’ (p.94). It should be noted that high levels of newness are represented by lower levels of compatibility (and thus lower levels of PEoU). Once again however, our perspective fundamentally differs from existing literature such as the TAM and IDT. So how might the newness of a product influence product adoption by the salesforce? According to Rogers (1962), people are more responsive to innovations that require less mental or physical effort to adopt. It therefore seems likely that, if a salesperson believes a new product is less compatible with experiences of his customers (e.g. not selfexplanatory, prior training needed), this salesperson is less eager to adopt the new product, all the more because it is harder to sell this product. Additionally, as novelty of the product increases, selling situations might become more uncertain, complex, and risky. As a result, higher levels of newness are associated with higher levels of change. The above suggests that greater uncertainty about the effort-performance linkage is the result of higher levels of product newness (Atuahene-Gima, 1997). As a result, less adoption might occur. In line with the TAM, we furthermore expect a direct link between product newness and product advantage. We therefore hypothesize: H3: Product Newness is negatively related to (a) SFNP Adoption and (b) Product Advantage. 2.3.4 Social Influence According to Hultink & Atuahene-Gima (2000), SFNP adoption is a motivational force that energizes the sales force to work towards the short and long-term success of the new product. Since social-contextual factors are a dominant perspective on the determinants of work motivation among employees (Haslam, 2004; Latham & Pinder, 2005), the social context is therefore involved in the current research, using the term ‘subjective norms’. The TRA and TPB defined subjective norms as "the person's perception that most people who are important to him think he should or should not perform the behavior in question" (Fishbein & Ajzen, 1975, p. 302). More focusing on changes in attitudes, Kelman (1958) distinguished three different processes of influence: compliance, identification, and internalization, where specifically compliance and identification focus on social interactions. When an individual is prone to a favorable reaction from another person or group and therefore accepts influences of others, compliance occurs. According to Kelman (1958), identification can occur when “an individual accepts influences because he wants to establish or maintain a satisfying relationship to another person or a group (p.53).” When investigating the link between product perceptions and behavioral intentions, TAM, its successors, and models based on them, find strong positive links between social influences and the product characteristics discussed in section 2.3.2 and 2.3.3 (e.g. Lu et al., 2005; Venkatesh & Davis, 2000). In a sales context, Schillewaert et al. (2005) for example found a positive link between social influences and both product-related constructs. Additionally, strong positive effects have been found between social influences and behavioral intentions (Venkatesh et al., 2003). In a sales context, Agarwal & Prasad (1998) show that salespeople will accept sales technology when their peers do so, while Homburg et al. (2010) demonstrated that coworkers’ and superiors’ adoption positively influence the

23

degree to which a subordinate adopts. So, subjective norms show to influence the product perception as well as actual usage behavior. We however are interested in a situation in which the adopter and the user are not the same individual. Two articles show that in a context related to selling -rather than usinginnovations, social influences too play an important role. First, Wieseke et al. (2008) hypothesize that brand adoption by sales managers is associated with the brand adoption of salespeople in a positive way, and find empirical support for this hypothesis. Second, Fu et al. (2010) find a positive relationship between “perceived organizational, managerial and social pressures”, and a salesperson’s selling intention. In line with Kelman (1958), the TAM and its successors, and articles by Wieseke et al. (2008), Fu et al. (2010), we expect subjective norms to influence product perceptions (product advantage and – newness), and product adoption in similar ways as described above. In sum, this leads to the following hypotheses: H4: Subjective norms are positively related to (a) Product Advantage, (b) negatively related to Product Newness (c), and positively related to SFNP Adoption. 2.3.5 Goal Orientation The process of selling new products is an endeavor in which salespeople try to achieve positive results related to that new product. Psychologists have identified two different underlying goals that people pursue in achievement situations. In the current research, effort covers ‘the salesperson’s force, energy, persistence, and the intensity of his/her activities to achieve desired results’ (Atuahene-Gima, 1997, p.500). In a sales context, exactly what influences the extent to which salespeople work hard and smart to achieve goals has been investigated by Sujan et al. (1994). In line with other scholars, they distinguish two types of goal orientation –learning- and performance orientation–, that in turn influence the way in which salespeople work. Someone who is more learning orientated desires to increase one's task competence, whereas a performance orientation reflects a desire to be positively evaluated by others and to do well (Farr, Hofmann, & Ringenbach, 1993). Empirical research suggests a positive link between both types of goal orientation and the effort devoted to selling a product (e.g. Sujan et al., 1994; VandeWalle, Brown, Cron, & Slocum Jr, 1999). These articles find a direct link between goal orientation and effort (e.g. working hard and – smart). In line with the findings of Sujan et al. (1994) and VandeWalle et al. (1999), the current research expects salespeople who tend to favor a learning orientation will be more eager to adopt a new product. Because they are innovative and risk taking (Atuahene-Gima, 1997), they are not afraid to face challenging selling situations in which they can learn from their mistakes, they will perceive the selling of a new product as a challenging new opportunity. Therefore, they are likely to adopt a new product. In contrast, their performance oriented colleagues are expected to be less eager to adopt a new product. Mainly because the selling of new products involves uncertainties and risks that may lead to mistakes, they fear negative evaluations. Furthermore, performance oriented people tend to avoid challenges (Button, Mathieu, & Zajac, 1996) such as the development of new selling skills, and thus feel threatened by the introduction of a new product. This leads to the following hypothesis: H5: Learning Orientation is positively related to SFNP Adoption while Performance Orientation is negatively related to SFNP Adoption. 24

2.3.6 Control system An important factor for controlling a salesperson’s behavior and aligning a salesperson’s behavior with priorities of the company, is the way in which their performance is measured, the so called control system (Ahearne et al., 2010). Although both formal and informal systems exist, we will focus on formal control systems because these can be designed by the company and by management (Atuahene-Gima, 1997). Following the definition of Anderson & Oliver (1987), a control system is ‘an organization’s set of procedures for monitoring, directing, evaluating, and compensating its employees’ (p. 76). In general, a control system can be outcome based, behavior based, or somewhere on the continuum between these two. In an outcome-based system, a salesperson’s performance is measured using objective, tangible results with little guidance or monitoring from management. In contrast, behaviorbased systems are characterized by high levels of guidance and monitoring by the management. In a new product selling context, the effect of both control systems have been investigated empirically. Hultink et al. (2000) for example state that commitment to goals is strongly affected by rewards, and empirically show that output-based compensation is positively related to performance while compensation based on behavior negatively influence the new product selling performance. Ahearne et al. (2010) find a negative relationship between a salesperson’s product perception and the effort devoted to it. However, based on the type of control system, the intensity of this relationship is increased (outcome-based) or reduced (behavior-based). Furthermore, path-goal theory suggests that commitment towards goals is influenced by the type of compensation used, while agency theory suggests that effort and commitment are increased by a behavior- rather than an outcome-based control system (Anderson & Oliver, 1987). But how exactly might it relate to SFNP adoption? Atuahene-Gima (1997) states that under a behavior-based system, a sense of security and steadiness is created to salespeople. This specifically is important in situations with high degrees of uncertainty, such as the process of selling new products. Due to this security, salespeople can learn about the new product and how to sell it without having to fear direct (negative) consequences, which should lead to greater commitment. Contrary, because an output-based system focuses on short-term goals without providing a sense of security, it may lead to little loyalty to the organization or product. In line with what has been written above we hypothesize: H6: The greater the degree to which a control system is behavior-based (outcome-based), the higher (lower) the degree to which the salesperson adopts a new product.

25

2.3.7 Perceived Involvement As discussed in paragraph 2.3.3, the introduction of new products involves change. According to social psychology, when people face events that bring potential change to their routines, an individual’s resistance to change dictates how they will respond (Oreg, 2003), and is therefore regarded as a determinant of successful change initiatives (Herold, Fedor, & Caldwell, 2007). Importantly, Lines (2004) states that ‘a person’s attitude towards the change and subsequent behavior stem from a process by which the perceived outcomes of a change are compared with the individual’s goals and values (p.198)”. Since adoption is conceptualized as functional, pro-change behavior by the adopter (Atuahene-Gima, 1997) and is composed of an attitudinal (commitment) and behavioral (effort) dimension, it seems important to include a concept that fosters pro-change rather than change resisting behavior. This supposition is fed by the fact that Jaramillo et al. (2012) state that salespeople in particular are subject to change. According to Giangreco & Peccei (2005), amongst others, two factors might influence the degree to which people resist to changes. First, the extent to which they see change as a threat to their interest influences their resistance. Acceptance or resistance to change could result from employees’ assessment of the costs and benefits associated with the change (Shum, Bove, & Auh, 2008), and the compatibility of the change with their values and norms (Dobosz-Bourne & Jankowicz, 2006). Second, the degree of their involvement in the change process influence whether they will perform pro-change or resistance to change behavior. In the current research, we will therefore include involvement. According to Glew, O’Leary-Kelly, Griffin, & Van Fleet (1995), involvement includes ‘opportunities for individuals or groups at a lower level in the organization to have a greater voice in one or more areas of organizational performance (p. 402)’. Ample empirical research finds evidence for a positive relationship between involvement and organizational commitment (Lines ,2004; Mayer & Schoorman, 1998; Savery, 1994), a positive link between involvement and a person’s attitude towards the change (Giangreco & Peccei, 2005), and negative influences of involvement on the degree to which people resist to change (e.g. Lines, 2004; Sagie & Koslowsky, 2000). Furthermore, in line with the definition of commitment adopted in the current research, involvement has been shown to lower one's resistance to accepting an externally imposed goal (Erez & Arad, 1986), while it relates positively to goal acceptance (Renn, 1998). The general mechanism behind this is that involvement can generate greater motivation and commitment to the change (Pugh, 1993), and because it helps motivate people to follow the path of change (Giangreco & Peccei, 2005). Based on the above, we hypothesize that: H7: Perceived involvement is positively related to SFNP Adoption.

26

2.4 Conceptual model Having decided which antecedents are considered most influential on SFNP adoption and how these concepts might relate to adoption, the conceptual model is developed. Figure 1 displays this model. Table 1 in turn, displays the relationships hypothesized in this research. Note that the creation of this conceptual model answers sub-question 2. Figure 1: Conceptual model – antecedents of sales force new product adoption.

Table 1: Overview of hypothesis

Hypothesis H1 H2 H3a H3b H4a H4b H4c H5 H6 H7

Statement

SFNP Adoption is positively related to Sales Performance. Product advantage is positively related to SFNP Adoption. Product newness is negatively related to SFNP Adoption. Product newness is negatively related to product advantage. Subjective norms are positively related to product advantage. Subjective norms are negatively related to product newness. Subjective norms are positively related to SFNP Adoption. Learning Orientation is positively related to SFNP Adoption while Performance Orientation is negatively related to SFNP Adoption. The greater the degree to which a control system is behavior-based (outcomebased), the higher (lower) the degree to which the salesperson adopts a new product. Perceived involvement is positively related to SFNP Adoption.

27

2.5 Conclusion This study attempts to extend our understanding of the new product selling process. Particularly, because a salesperson has to assess the value of a new product for someone else (namely his/her customers) this research distinguishes itself. While previous research focused on the effects of adoption in a situation in which the decision maker, the adopter and the user were the same individual, we are interested in a situation in which these different roles are not embodied in one individual: the salesperson is only an influencer in the customer’s (the user) decision process. To our knowledge, this perspective received little attention. For the company involved, this research presents a first attempt to identify what specifically influences adoption by the sales force and offers practical guidance for productand sales managers.

28

3

Methodology

In the previous chapter, the conceptual model and the hypothesis were described. To test this conceptual framework, an explanatory research will be conducted within the coating manufacturer. In this chapter of the research proposal, four elements of the research design are elaborated on: the method, the sample, measures used, and the data analysis.

3.1 Method Concepts covered in the current research all stem from a review of literature on product adoption and new product selling. It is noteworthy to mention however that, in order to establish a high level of practical relevance, two brainstorm sessions were arranged with multiple people from different layers of the cooperating organization. Ideally, both salespeople and product managers would attribute to one single session but due to practical reasons, only one meeting with product managers was held. Keeping the antecedents derived from literature in mind, the most important antecedents of SFNP adoption for the company were selected during this session. Interestingly, product managers mentioned most important antecedents, including both product-related concepts, involvement, social influences and ways to incentivize salespeople. To test the hypothesized relationships of the conceptual model and to better understand the new product selling process, an online survey was conducted, using the online survey tool ‘www.surveymonkey.nl’. For multiple reasons, this method is chosen: it saves costs, large sample sizes are possible, it is easy to use, responses can be very quick, it is available all the time, and it eliminates the interviewer bias (Singleton & Straits, 2005). Downsides however are the low response rate and non-response errors. 3.1.1 Increasing response rates and overcoming biases As mentioned, online surveys tend to be subject to low response rates. To increase the response rate, we followed the “mail questionnaire” procedure elaborated on by Dillman et al. (2009), and furthermore included the sales managers in this procedure. Both sales managers and actual customer-contact employees received 4 notifications. First, both groups received a pre-notification (see Appendix H.1 and H.2). A week later, the customercontact employees received an invitation (see Appendix H.3). A day after this invitation, their sales managers received a notification that the invitations had been sent (see Appendix H.4). In the last week before starting the actual analysis, this same procedure was repeated with a first reminder (see Appendix H.5 and H.6), and a second reminder (see Appendix H.7 and H.8). Furthermore, two biases are accounted for: the common method-, and the non-response bias. Because participants are responsible for the dependent and independent variable, common method bias might occur. It is “the variance that is attributable to the measurement method rather than to the constructs the measures represent” (Podsakoff, MacKenzie & Lee, 2003, p. 879). Podsakoff, MacKenzie, & Podsakoff (2012) mention several procedural remedies to overcome this type of bias, including: obtaining data from different sources, separation between predictor and criterion, elimination of common scale 29

properties, scale item improvement to eliminate ambiguity, reducing social desirability, and balancing positive and negative items. A priori, we intended to take these remedies into account. Unfortunately, objective data regarding the sales performance were not available due to confidentiality of data. Although all other procedural remedies were accounted for as best as possible, the common method bias might still be present. Second, the non-response bias is accounted for. This bias occurs if answers of those who responded differ from potential answers of those who did not respond to the survey. Findings cannot be generalized if significant differences occur between these groups. Although we cannot analyze the answers of those who did not respond, Armstrong & Overton (1977) suggest that a subjective estimation of the bias can be done by comparing late respondents to early respondents. A t-test was then conducted to measure the difference between these two groups. One significant effect arises (see Appendix G): those who participated before a reminder was sent rate their performance significantly lower than those who respond only after a reminder was sent (t(66,46) = −3,00, p < .01).

3.2 Sample The target population of this research consists of customer-contact employees in a business-to-business context, employees that are able to influence a customer’s perception of a product. Because the target population is too big to target entirely, a sample population will be used to test the hypothesized relationships and draw conclusions on the population. Within the cooperating company however, two types of people exist that can be considered customer-contact employees (and thus are considered influencers in a customer’s decision making process). First, ‘traditional’ sales are responsible for selling new products to customers. Second, technical consultants (TCs) have a technical background and are responsible for solving product related problems. These TCs however, too are aware of the latest products and can shape a customer’s perception of a new product. For practical reasons, specifically countries where either Dutch or English is an official language are invited to participate. Although the method to analyze the results (PLS-SEM) is able to deliver reliable results with relatively small sample sizes (Hair, Sarstedt, Ringle, & Mena, 2012), when using PLSSEM, Hair, Ringle, & Sarstedt (2011) recommend a sample size of (i) at least ten times the largest number of indicator variables measuring one construct, or (ii) ten times the largest amount of paths pointing at one latent construct in the structural model. Since the largest amount of paths pointing at one construct (SFNP Adoption) is eight (goal orientation and control system both include two constructs), the aim is to collect the answers of at least 80 participants. Furthermore, the sample size should be representative in order to secure the validity and reliability of the constructs. Multiple statistical tests can be conducted to test for this validity and reliability of the findings (see Chapter 4).

3.3 Measures Reliable and tested perceptual measures, derived from previous research, are used, all being self-reported by the participant. The constructs in the model are all measured using multi-item (Likert) scales. To prevent repetitive answering and other response biases, five point- and seven point scales are used interchangeably. Furthermore, reversed items were added which should balance the questionnaire and help to overcome the common method bias. 30

In general, we use the following statements: (1) “To a little extent” to “To a great extent”, (2) “Strongly disagree” to “Strongly agree”, (3) “Much less” to “Much more”, and (4) “Very unimportant” to “Very important”. Participants were given five products to choose from. If a question in the survey referred to ‘product’ or ‘new product’, the chosen product was kept in mind. On average, these products were available for 11 months (varying from 21 to 2 months). First, the measure for sales performance is adopted from (Sujan et al., 1994) and measured using the first statement. In contrast to other measures, this construct is measures on a ten point scale. Participants are asked the degree to which he or she has been successful in achieving certain sales objectives. Although objective performance data are preferred, the company involved did not allow us to use these data. For this reason, selfreport measures are used. Second, SFNP Adoption is conceptualized by the interaction of two measures: commitment and effort. Both measures are adapted from Hultink & Atuahene-Gima (2000), where commitment is measured using the second statement while effort uses the third statement. Using statement two, the measures for product advantage and product newness are derived from different articles (see Appendix A) while two items are added. These seven product-related items are presented together, and it is expected that, in a factor analysis, item 1,2 and 3 will load on product advantage, while all other items load on product newness. Furthermore, two items were reverse-scored. Using the first statement, the measurement for Subjective Norms is derived from Fu et al. (2010). It is composed of the degree to which salespeople think others in their environment consider selling the new product to be important (Others) and the degree to which people are motivated to comply with the wishes of these others (Motivated). Four reference groups are used: sales manager, product manager, fellow sales reps, and technical consultant. To calculate the value for subjective norms, the following formula is used: 4

𝑆𝑁 = ∑ 𝑂𝑡ℎ𝑒𝑟𝑠𝑖 ∗ 𝑀𝑜𝑡𝑖𝑣𝑎𝑡𝑒𝑑𝑖 𝑖=1

Goal Orientation uses the second statement and is adopted from VandeWalle et al. (1999), while the control system is measured using the first statement, and is adopted from Jaworski & MacInnis (1989). Focusing on the development process but from a salesperson’s perspective, the measure for Perceived Involvement is partially adopted from Searfoss & Monczka (1973) while one item was added. Next to these measures, control variables are included in the research.  Age  Gender  Function: Technical Consultant / Sales / Other  Years of experience within sector  Experience within selling function  Tenure time with company  Country 31

The wording of some measures was slightly adjusted such that they are applicable for the context in which the research is held. To ensure that the instructions, questions, and scale items were understandable for participants, five people were asked to check these measures before the final version of the survey was distributed.

3.4 Data Analysis Partial Least Squares Structural Equation Modeling (PLS-SEM) is used to test the hypotheses. Since some of the constructs included in the model are expected to both directly and indirectly influence other constructs (e.g. Product Newness and Subjective Norms), SEM is very useful. It distinguishes itself from other regression methods because a construct that acts as an independent variable in one relationship can be the dependent variable in another relationship (Hair et al., 2006). SEM therefore will provide a path analysis between the different constructs easily.

32

4

Data Analysis and Results

An invitation to the survey was sent to 166 people, to which 112 (67,5%) people responded and answered the very first question. In this chapter, these responses are analyzed. For the model analysis, Partial Least Squares Structural Equation Modelling (PLSSEM), specifically SmartPLS software, is used to run the analysis and test the hypotheses. For some additional analysis, SPSS Statistics is used. First, the dataset is examined and assumptions are tested. Thereafter, a factor analysis is conducted to confirm underlying constructs, including reliability and validity tests. Finally, paragraph 4.4 and 4.5 elaborate on the actual model analysis.

4.1 Examination of data Hair, Black, Babin, Anderson & Tatham (2010) state that, before the actual analysis can start, the data have to be examined rigorously for missing values, outliers, and whether assumptions are violated, which prevents for potential biases. Therefore, a first goal is to create a dataset where all these influences have been checked for. Because the sample is not very big (N=112), it was tried to prevent case deletion. 4.1.1 Missing values In the software used to collect the data, settings were adjusted such that missing data could only occur when a participant stopped somewhere during the survey. Unfortunately, 15 participants did not complete the full survey, and answered less than 90% of the questions. According to Hair et al. (2010), missing value analysis is only useful when a respondent answered at least 90% of the questions. Therefore, it was decided to delete these 15 cases. The remaining 97 observations, a sufficient number to run PLS-SEM, are further investigated. Although Hair et al. (2010) propose a four-step analysis regarding missing data (with step three and four being randomness determination and imputation method), and because the non-response bias was already elaborated on in paragraph 3.1.1, no further analysis regarding the missing values is needed for the 97 cases left. 4.1.2 Outliers Second, the dataset is checked for outliers. Hair et al. (2010) defined outliers as ‘a unique combination of characteristics identifiable as distinctly different from the other observations’ (p. 63). Univariate (a), and multivariate outliers (b) are examined here. (a) Univariate: The software used to conduct the survey only allowed respondents to answer according to a Likert-scale with pre-set values. For this reason, univariate outliers do not occur. (b) Multivariate: Although also bivariate outliers exist, Hair et al. (2010) provide two arguments why not to check for bivariate outliers: (i) a large amount of graphs is needed to identify the outliers, and (ii) only two variables can be checked at a time. Therefore, only multivariate outliers are checked here. To test for multivariate outliers, the Mahalanobis D2 measure is used. This method measures each observation's distance in multidimensional space from the mean center of all observations (Hair et al., 2010). The D2 is calculated in SPSS, then transformed to a χ2 distribution, and is consequently subtracted from 1. Hair et al. (2010) propose a cut-off point of 0,001. Two separate χ2 distributions were calculated: one for commitment and effort, and one for all factors expected to influence SFNP Adoption. Results show one multivariate 33

outlier related to the antecedents. Inspection suggests that this participant had the perception of being involved to a high extent. Since this is possible, it is decided to refrain from deleting this case. 4.1.3 Assumption testing Normality of data is an important assumption in multivariate analysis. PLS-SEM is able to deliver reliable results even with a small sample size and is robust against a violation of this assumption (Hair et al., 2010). However, with a relatively small sample (n=97) and because Esposito Vinzi, Chin, Henseler, & Wang (2010) state that this method is still affected by skewness of variables, normality is investigated. Furthermore, socially desirable answers might occur because the questionnaire covers questions about a participant’s sales performance and effort devoted to selling a product within the context of their current employer. Normality of data can be tested both graphically as well as statistically. The results of the statistical analysis (see Appendix B) show most items to be non-normally distributed. With the exception of the involvement items (PerInv), most items are negatively skewed. To overcome non-normality of items, they are transformed. Tests with squared or cubed items however show the occurrence of a similar pattern: if a transformation diminishes skewness, kurtosis increases. Therefore, it is chosen to not transform the data. Although PLS-SEM is robust against violations of normality, these violations should be kept in mind. 4.1.4 General information about participants After the examination of the data, the analysis can be conducted with the remaining 97 cases. Table 2 displays some information about these cases. Table 2: General information about participants

Characteristic Sex Age Function

Category Male Female In years Technical Consultant Account Manager Else

Tenure Sector Tenure Function Tenure AkzoNobel Note: N=96 due to one missing value.

Total (N=96) 94 (97,9%) 2 (2,1%) 49,1 (sd = 7,5) 26 (27,1%) 68 (70,8%) 2 (2,1%) 24,4 (sd = 9,9) 19,4 (sd = 8,7) 11,8 (sd = 7,7)

Characteristic Category Country NL UK Ireland USA Canada Australia

Total (N=96) 10 (10,4%) 10 (10,4%) 4 (4,2%) 56 (58,3%) 9 (9,4%) 7 (7,3%)

4.2 Factor analyses In this section, the factor analyses and tests related to reliability and validity are elaborated on. First, because the measures for product advantage and – newness are partially derived from literature but also include new items, an exploratory factor analysis (EFA) is conducted to test the distinguishability of these dimensions. However, although all other measures are adopted from validated scales, and are therefore theoretically justified, we will still analyze them. Thereafter, a confirmatory factor analysis (CFA) is conducted and reliability and validity tests are discussed.

34

4.2.1 Exploratory factor analysis on product perception As explained in paragraph 3.3, the three items for product advantage and the four items for product newness are represented together in the survey. On a scale from 1 (strongly disagree) – 7 (strongly agree), participants had to indicate to what extent they agreed with seven statements regarding the new product. Most items were positively worded, while two items were negatively worded. Before this analysis is conducted, the two negatively worded items (PN1 and PN4) are reversed. Field (2009) states that the KMO measure of sampling adequacy should be above 0.7, Bartlett’s test should be significant, and all eigenvalues should exceed 1 (Field, 2009). Using oblique rotation, all these criteria are met. The outcome of the analysis is not in line with expectations. Contrary to expectations, three factors can be distinguished (see Table 3): (1) related to product advantage, (2) related to newness items that were negatively worded, (3) related to newness items that were positively worded. This suggests that this measure might be subject to a response bias, referred to as the reversed item bias (Weijters, Baumgartner, & Schillewaert, 2013; Weijters, Geuens, & Schillewaert, 2009). This bias would decrease the reliability of the measure. A closer examination of the data reveals that, indeed, the answers of multiple participants seem to be influenced by the fact that two items were negatively worded. In these cases, where the answers for PN1 and PN4 differ much from PN2 and PN3, participants had the tendency to agree with an item regardless of its statement. Additionally, because the data are heavily skewed (see paragraph 4.1.3), potentially the data are subject to another bias, namely acquiescence. Table 3: Factor matrix product perception

To overcome this problem, two approaches can be adopted. The first option is to remove all participants with a high difference between the positively and negatively worded items. To do this, first the average of the positively worded (PN2 and PN3) and negatively worded (PN1 and PN4) is calculated. Subsequently, the absolute difference between these two averages is calculated and cases with a difference of more than 2,5 are deleted. We are aware that this cut-off value of 2,5 is rather subjective. However, if cases with a difference of more than 2,5 are deleted, this would lead to the deletion of 31 cases (N=66). The advantage of this approach is the data used are reliable, and therefore the findings will be more reliable. However, an important drawback of this approach would be the fact that the sample size decreases to an unacceptably low level, which would decrease the reliability. A second option is the deletion of the two negatively worded items. An important disadvantage of this approach is that the influence of the bias is not really accounted for. However, the advantage of this approach is that the sample size remains large enough to 35

draw statistically valid conclusions. Because evidence suggests that not all measures are affected by response biases equally (Podsakoff et al., 2012), and the sample size remains sufficiently large, the second approach is adopted in the remainder of this research. When rerunning the analysis without the two reversed items, two factors are clearly distinguishable (see Table 4). These results plead for the deletion of the two reversed items. Although a cross-loading occurs, Field (2009) states that the distance between the factor loadings should at least be 0,2, while Hair et al. (2010) state that cross-loadings should not exceed 0,6. Both these criteria are met. Appendix C shows the results of the exploratory factor analysis with all items included. Table 4: Factor matrix product perception after case deletions (N=97)

4.2.2

Confirmatory factor analysis

Although all other measures are adopted from previous research, they are analyzed as well. The results of this analysis, including reliability and validity tests are displayed in Table 5. Prior to the analysis, items were standardized. The remainder of paragraph 4.2 will elaborate on the reliability and validity tests that will ultimately result in the deletion of some items. Item reliability: Hair et al. (2010) state that with a sample size of 100 and above, a factor loading of 0,55 is appropriate. Because the sample size for this research is 97, a cut-off value of 0,55 is used. Several items have a loading below 0,55 and might ultimately be subject to deletion. Internal consistency: Cronbach’s alpha has the tendency to underestimate internal reliability in PLS path models and therefore, the more appropriate composite reliability (CR) measure is applied here (Henseler, Ringle, & Sinkovics , 2009). Values above 0.7 are regarded as satisfactory (Nunnally & Bernstein, 1991). As can be seen in Table 5, either Cronbach’s alpha or the CR is insufficient in two cases: sales performance and product newness. In both cases, the value of both measures increases when items with low item reliability are deleted (see Table 5, ‘after item deletion’). Regarding sales performance however, it can be argued that distinct types of performance are measured. While item 1-3 cover target obtainment, item 4 measures a salesperson’s performance relative to others.

36

Table 5: Confirmatory factor analysis, reliability- and validity tests

37

Continuation of Table 5.

Convergent validity: The convergent validity can be assessed using the Average Variance Extracted, or AVE-value (Fornell & Larcker, 1981). If a latent variable has an AVE-value above 0,5 (cut-off according to Hair et al. (2010)), this indicates that this variable explains more than half of the variance in the items. Only the AVE-value of product newness does not surpass the cut-off value while the value after deletion of the two reversed items is well above the cut-off point. Discriminant validity: To assess the discriminant validity, the (a) Fornell‐Larcker criterion and the (b) cross‐loadings of the items are examined. (a) Fornell-Larcker criterion: the scholars after which the criterion is named state that the correlations of any construct in the model and the construct of interest should never exceed the square root of the AVE-value of the construct of interest. As can be seen in Appendix D.1, is this criterion met for all constructs, although the control-systems show to be problematic. Three important conclusions can be drawn from this table. First, these findings show that commitment and effort are distinct from each other, which justifies their interaction to form SFNP adoption. Second, these findings show that the two dimensions of subjective norms (‘what others think’ and ‘motivated to comply with wishes’) can be discriminated from each other. Third, it is doubtful whether the control system measures truly measure something different, something we will elaborate on below. (b) Cross-loadings: Field (2009) states that if an item loads on more factors, the distance between them should at least be 0,2 while Hair et al. (2010) state that cross-loadings should not exceed 0,6. Appendix D.2 shows that some items have cross-loadings higher than 0,6: Com3, PA1, PA3, PO1, PO4 and many control-system items. The cross-loadings between commitment and product advantage indicate that salespeople are more committed to a new product when they believe it has clear advantages. It can be argued that, because commitment includes the acceptance of the new product, better products will be accepted more easily and therefore a cross-loading occurs between product advantage and commitment. 38

Regarding the performance orientation items, PO1 states: ‘It is very important to me that my supervisor sees me as a good salesperson’. It can be argued that, although this is linked to performance orientation, a salesperson with a higher degree of learning orientation wants to be considered a good salesperson too, and therefore a cross-loading occurs. Regarding PO4 however, a deletion can be justified because its factor loading is below the cut-off point, a cross-loading within the 0,2 range does occur, and because a deletion of PO4 significantly increases the AVE-value. Last, many cross-loadings exist within the control-system construct. A possible explanation for this is, as Oliver and Anderson (1994) stated, that ‘output and behavior control are polar opposites and that management may elect to position their control strategy at various levels of the continuum (p.53)’. Because a stronger influence of behavioral control is expected during the period before adoption (see Hypothesis 6), it is therefore decided to delete the outcome control items. In sum, some items prove to be problematic, and the following reflective measures are therefore deleted: SP4, PN1, PN4, PO4, and the whole outcome control construct (CSOC1-4). While some suggest a minimum of two indicator variables per construct (Bullinger, Anderson, Cella, & Arronson, 1993), others propose a minimum of three (Raubenheimer, 2004). In order to keep the product newness construct in the analysis, at least 2 items have to be retained to remain a reliable instrument, a condition that is met.

4.3 Correlations, mean and standard deviation Now reliability and validity have been tested, we will investigate the correlations between constructs, displayed in Table 6. In line with what can be expected, there are a few correlation-values that are medium-high (>0,6), namely both dimensions of adoption with the adoption construct. Furthermore, apparent differences between commitment and effort occur, something that will be elaborated on in paragraph 4.5.3. Last, perceived involvement significantly correlates to only three other constructs in the model: sales performance, product newness, and learning orientation. Table 6: Correlations, Mean and Standard Deviation

39

4.4 Model results Having investigated the correlations between constructs, the model can be investigated. Structural Equation Modelling (SEM), particularly SmartPLS, is used to analyze the data. In this software tool, default settings were applied. In order to test the hypothesis, Model 1 is created (see Table 7). Figure 2: Model results for testing hypothesis

Table 7: Overview models for testing hypotheses Model 1

2

R= 0.153 Paths Modelled Coefficient t-value Hypothesis *** SFNP Adoption H1 Sales Performance 0,392 3,755 Supported *** Product Advantage H2 SFNP Adoption 0,360 2,928 Supported Product Newness H3a SFNP Adoption -0,141 1,446 Not supported *** Product Newness H3b Product Advantage -0,318 3,355 Supported *** Subjective Norms H4a Product Advantage 0,320 3,627 Supported *** Subjective Norms  H4b Product Newness -0,373 3,840 Supported * Subjective Norms  H4c SFNP Adoption 0,175 1,697 Marginally Supported Learning Orientation  H5 SFNP Adoption 0,038 0,401 Not supported Performance Orientation  H5 SFNP Adoption 0,104 1,024 Not supported *** Behavioral Control  H6 SFNP Adoption 0,210 2,625 Supported Perceived Involvement  H7 SFNP Adoption 0,083 0,718 Not supported Note: significance levels ***p1.65)

As can be seen in Table 7 and Figure 2, are some hypothesis supported while others are rejected. Note that these findings help us answer sub-question 3. Interestingly, both product advantage and – newness significantly influence product adoption individually. However, if these product perceptions are combined, the significant effect of product newness disappears due to its significant link with product advantage. 40

4.5 Additional analysis Now the hypothesized relationships are tested, additional analyses are conducted to check for other effects. First, this section elaborates on moderation effects. Next, a model is created with additional direct effects. Last, a model is created in which adoption is replaced by either commitment or effort to investigate the difference between these constructs. 4.5.1

Moderating effects

To test whether relationships are influenced by other variables, moderation effects are tested for. In order to overcome multi-collinearity, mean-centered variables are used (Fielding, McDonald, & Louis, 2008), an option that can be chosen in the software tool used. Many moderations were tested (see Appendix E) but only two effects were significant, displayed in Table 8. Table 8: Moderation effects

Moderating variable Learning Orientation Age

Relationship moderated Product Advantage → SFNP Adoption Product Advantage → SFNP Adoption

Coefficient t-value 0,127* 1,694 -0,203** 2,144

Increase of R2 0,022 0,029

Note: significance levels ***p1.65)

When combining these moderation effects in one model, the R2 in adoption increases from to 0,534 to 0,576. The moderating effect of learning orientation however becomes insignificant in this combined model. A visualization of these effects (see Figure 3 and Figure 4 Figure 4) enables the interpretation of them. Figure 3: Moderation Learning Orientation

Figure 4: Moderation Age

Figure 3 shows that the relationship between product advantage and adoption is stronger when a salesperson is more learning oriented while Figure 4 shows that the higher the age of a salesperson, the weaker the relationship between product advantage and adoption.

41

4.5.2

Additional direct paths

From the correlation-table, we can derive that sales performance correlates with several antecedents. Additionally, perceived involvement correlates (marginally) significant with only sales performance, product newness, and learning orientation. Therefore, additional direct paths are tested. Figure 5 and Appendix F display the results of these analyses, and show that four additional paths are significant: product newness and perceived involvement both directly influence sales performance, perceived involvement decreases the degree to which a salesperson thinks a new product is incompatible with experience and consumption patterns, and learning orientation negatively influences the perception of being involved. Interestingly, when all these effects are combined to one model, the effects of adoption on sales performance, and involvement on sales performance become less significant. Figure 5: Additional direct paths

Note: significance levels ***p1.65)

4.5.3

Difference between Commitment and Effort

From Table 6 we derive that some constructs relate very differently to commitment and effort. For example, learning orientation strongly relates to commitment, while its correlation with effort is not significant. Therefore, differences between these constructs are tested for. First, the items of both constructs are mean centered. When, in the basic model (see Figure 2), adoption is then replaced by either commitment or effort, these two newly created models show very different results. Most surprising about these results (see Table 42

9), is that while adoption correlates most strongly to effort, the models tested here show that the ‘commitment-model’ shows stronger similarities based on the explanatory power (R2). Furthermore, learning orientation does not influence effort, while it significantly influences commitment. Table 9: Difference between Commitment and Effort

Last, when these two constructs, and their interaction are combined in one model, we are able to predict 15,7% of the variance in ‘sales performance’, roughly the same percentage as with the original model.

43

5

Discussion and conclusion

The introduction of this thesis states that the main purpose of this study is to uncover what factors influence new product adoption by the sales force such that the degree to which salespeople adopt a new product will be increased. In particular, the effect of factors related to the product, social influences, and the organization are investigated. Next to that, the effect of adoption on sales performance is investigated. This chapter serves as a wrap up of this research. First, the main findings will be discussed. Thereafter, the theoretical and managerial implications are elaborated on. Finally, limitations and directions for future research are provided.

5.1 Discussion of results Innovation adoption literature investigates the choices and decisions individuals make to adopt (or reject) an innovation. Factors that might influence the decision are formed over time (Straub, 2009) and can relate to the process, and to beliefs and attitudes regarding the product. Furthermore, cognitive, emotional, social, and contextual factors influence adoption as well (Burkhardt, 1994; Kraut, Rice, Cool, & Fish, 1998; Straub, 2009; Zablah, Chonko, Bettencourt, Allen, & Haas, 2012). This led us to expect that different categories of antecedents might influence the degree to which salespeople adopt a new product. To determine these antecedents, antecedents derived from research streams focusing on (i) product adoption, (ii) the link between attitudes, intentions, and behavior, and (iii) new product selling were reviewed on their effect size. Based on this, we selected five antecedents: ‘product advantage’ and ‘product newness’ (both product related), ‘subjective norms’ as a social factor, ‘goal orientation’ as a personal trait of the salesperson, and ‘control system’ as influential factor during the process. Based on a literature review on change behavior, a sixth antecedent is added: ‘perceived involvement’. In order to find out whether these concepts influence the degree to which a new product is adopted, structural equation modelling (SEM) was used to test the hypothesized relationships. The basic model (See Table 7 and Figure 2 for results) reveals that 6 out of 10 hypotheses are supported, while one hypothesis is marginally supported. In line with previous research (e.g. Hultink & Atuahene-Gima, 2000), this research shows that the degree to which a salesperson adopts a new product positively influences his or her sales performance. Next, factors potentially influencing product adoption were investigated. Results show that a salesperson’s perception of the product’s advantage strongly influences the degree to which this product is adopted: the more advantageous a product is, the more it is adopted. This finding conflict with Ahearne et al. (2010), who found a negative relationship between product perception and the effort devoted to selling it. While these researchers used one objective performance measure to measure effort (total number of phone calls made), we additionally asked participants to rate their effort on ‘collecting market information’, ‘using market information’, and ‘building customer relationships’. Possibly, because these measures were self-reported, and because we asked for more than one performance item, we find conflicting results.

44

In line with the TAM (Davis, 1985; Venkatesh & Davis, 2000), a relationship is found between a salesperson’s perception of a product’s newness and the perception of the product’s advantage. Next to that, as hypothesized, factors in a salesperson’s social environment strongly influence his or her perception of a product’s advantage and the newness of the product while we find marginal support for the link between subjective norms and adoption. This is in line with findings of for example Fu et al. (2010), who found a positive relationship between subjective norms and selling intention, and Venkatesh et al. (2003), who found a significant positive relationship between social influences and behavioral intention. These results however suggest that subjective norms have a more profound influence on product adoption indirectly (via product perception), rather than directly. While the results of the CFA lead to the deletion of the outcome control items, behavioral control positively influences adoption. This positive relationship is in line with, for example, the agency theory (Anderson & Oliver, 1987), and shows that the type of control system not only has a profound impact once a product is adopted (e.g. Hultink & Atuahene-Gima, 2000; Hultink et al., 2000), but it also directly influences adoption. However, some hypotheses were not supported. Although the beta has the expected sign (β=-0,141), the effect of product newness on adoption is not significant. Similar results have been found related to the TAM; while the link between a product’s usefulness and intentions is consistently strong, this is not the case for perceived ease of use (Davis, 1985; Ma & Liu, 2004; Venkatesh & Davis, 2000). Furthermore, both learning - and performance orientation do not relate to adoption, while literature shows strong positive relationships between both types of orientations and effort (see for example Sujan et al., 1994; VandeWalle et al., 1999). This will be elaborated on when discussing the difference between adoption, commitment, and effort. Last, contrary to expectations, involvement in the development process is not directly related to product adoption. A possible explanation for this is that being involved only in the development process is not enough to foster adoption. Because there might be a long time between the development of a product, and the actual launch of that product, it can be hypothesized that involvement in an early stage (e.g. the development of a product) has no influence on adoption, while involvement in stages closer to the launch will. Additionally, as the additional analysis show, involvement could influence adoption indirectly (via product newness). From the additional analyses (see Figure 5 for results), several conclusions can be drawn. Concerning moderation effects, the relationship between product advantage and adoption is moderated by the age of a salesperson such that a higher age weakens the relationship. Although a fundamental difference exists between UTAUT and the current research, these findings are comparable with findings of Venkatesh et al. (2003). Furthermore, we investigated whether these data support additional direct effects. Most importantly, sales performance is strongly influenced by product newness. One should realize that sales performance is an indication of the degree to which customers adopt a new product. Acquiring a new product might require customers to change their behavior, which often involves costs, both actual transaction costs as well as psychological costs (Gourville, 2006), which particularly hold for products with higher levels of newness. Although a new product might offer substantial benefits over its substitutes (gains), a potential change in behavior (e.g. in the case of high newness) might feel like a loss. Since 45

prospect theory shows that “losses loom larger than gains” (Kahneman & Tversky, 1979, p.279), this might explain why ‘compatibility with experience and consumption patterns’ directly influences sales performance, while product advantage does not directly influence the sales performance in this case. Additionally, sales performance is directly influenced by the degree to which people perceive to be involved in the development process. I cannot explain this direct link. It should be noted though that the addition of these direct effects weakens the relationship between adoption and sales performance, while individually, their relationships with sales performance are strong. An analysis of relationships between antecedents reveals that the perception of being involved negatively influences the perception of product newness. A possible explanation for this is the fact that being involved in the development of a new product could create a ‘hands-on’ experience with the new product. For this reason, salespeople believe a product is more compatible with experiences and more easy to use, and potentially see better ways to sell it. Alternatively, and in line with findings of Judson et al. (2006), it could be argued that involvement of salespeople leads products where the changes are less radical. Based on the effect of involvement on both product newness and sales performance, we can conclude that involvement does influence the sales performance, where the indirect effect is strongest. Next, the results show that learning orientation negatively influences perceived involvement. Both constructs cover an investment of time: investing time in learning how to be a better salesperson, and investing time in communicating new ideas and developing new products. Because a salesperson has to make a trade-off between improving his own skills (learning orientation), and involvement, the negative relationship the negative relationship could be explained. Based on apparent differences between commitment and effort in the correlation matrix, it was investigated how exactly these constructs differ from each other. The results of this comparison show that commitment is most similar to adoption regarding the amount of variance explained. Additionally, with the antecedents mentioned, a much larger percentage of the variance in commitment (59,7%) is explained, compared to the variance explained in effort (32,6%). Furthermore, while literature suggests a strong link between both types of goal orientation and effort (Sujan et al., 1994), our findings do not confirm this. In fact, only learning orientation relates to commitment, while performance orientation has no significant effect on either one of the two dimensions of adoption. A possible explanation for the link between learning orientation and commitment is that learning orientation stems from an intrinsic interest in one's work (Meece, Blumenfeld, & Hoyle, 1988). People high on learning orientation have the desire to increase their task competence and learn from new selling situations. It might be argued that, in order to learn from these new situations, they have to be more eager to accept and internalize the goals of a new product. Additionally, effort is not influenced by either learning- or performance orientation, while previous research suggest that both dimensions (Sujan et al., 1994), or learning orientation (VandeWalle et al., 1999) influence the effort devoted to selling. A possible explanation for this difference is the fact that effort is measured in different ways. Sujan et al. (1994) investigate the influence of both orientations on working smart, which covers planning and adaptive selling, and working hard, which covers the persistency with which activities are executed. VandeWalle et al. (1999) on the other hand investigated the influence of both 46

orientations on ‘intended effort’ compared to other salespeople. We however asked participants to rate, in retrospect, the effort they did devote to several selling related activities. Based on these results, can we answer the question whether adoption is an attitude, a behavior, or indeed an interaction between these two? In trying to explain human behavior, two theories are well-established: the Theory of Reasoned Action (TRA; Fishbein & Ajzen, 1975), and an extension of the TRA: the Theory of Planned Behavior (TPB; Ajzen, 1991). These theories have one thing in common: they both conclude that the performance of a behavior is determined by the intentions to perform that behavior. They assume intentions “to capture the motivational factors that influence a behavior” (Ajzen, 1991, p.181), and state that intentions indicate the extent to which people plan to exert effort to perform the behavior. Investigating what influences behavioral intentions, the TRA postulates two determinants of intentions: attitude toward behavior, and subjective norms. The TPB later added a third determinant to this: perceived behavioral control (e.g. Fishbein & Ajzen, 1975; Ajzen, 1991). These two theories however explained general behavior. More specifically investigating technology adoption and usage behavior, the Technology Acceptance Model (TAM; Davis, 1985) was developed based on the TRA. This model suggests that two specific behavioral beliefs, perceived ease of use (PEoU) and perceived usefulness (PU), determine an individual's attitude toward use, and behavioral intention to use that technology. We conceptualize adoption as the interaction between commitment (which is an attitude towards the new product), and effort (which covers the activities to achieve results). Importantly, it thus covers an attitude, and a behavior, while intentions are not included. This is interesting, specifically because Hultink & Atuahene-Gima (2000) state that SFNP adoption is a motivational force that energizes the sales force to work towards the short and long-term success of the new product. Additionally, Fu et al. (2010) state that intentions are the antecedent to effort. Thus, when comparing our conceptualization of adoption with the earlier mentioned theories, the essential motivational concept mediating attitudes and behavior, intentions, is missing. From a theoretical perspective, and in line with Fu et al. (2010), it might be hypothesized that including intentions would lead to a better conceptualization of what ‘SFNP adoption’ intends to represent: a motivational force. If we investigate the findings of Figure 2, the comparison between adoption, commitment and effort, and the findings of the TRA and an extension of the TAM, TAM2 (Venkatesh & Davis, 2000), strong similarities exist. In line with the TRA, believes about the product (advantage and newness) influence the attitude towards it, and in line with findings of TAM2, we find subjective norms to influence adoption both directly, and indirectly (via beliefs of the product). Interestingly, based on the variance explained by the antecedents, and the variance it can explain in the sales performance, commitment towards the new product is most similar to adoption. This shows that these antecedents much better predict the attitudinal –rather than the behavioral- dimension of adoption. This too leads us to hypothesize that including intentions would lead to a better conceptualization of what ‘SFNP adoption’ intends to represent: a motivational force. Overall, we conclude that sales performance is most strongly influenced by the newness of a product, while product adoption, and perceived involvement directly influence it too. 47

In turn, product adoption is most strongly influenced by a salesperson’s perception of the product’s advantages. Next to that, salespeople should be monitored, directed, and evaluated based on their behavior, because this increases their rate of adoption and creates a sense of security and steadiness. Additionally, the influence of the social environment should not be underestimated. Together, this answers the main research question.

5.2 Implications 5.2.1

Theoretical Implications

In the endeavor to answer all research questions, valuable insights were created. This research therefore deepens our understanding of the new product selling process, and provides valuable theoretical insights concerning this process. Most importantly, this research is the first to empirically investigate how several antecedents can influence product adoption by the sales force. Although Atuahene-Gima (1997) conceptually explored this research area, empirical support was lacking. Below, the theoretical implications of the findings are elaborated on. First, while others focus on factors influencing selling intentions (Fu et al., 2010) or selling effort (Ahearne et al., 2010), we investigate the influence of subjective norms and productrelated antecedents on product adoption. The link between these antecedents and adoption has been investigated for situations in which the adopter and the user were the same individual (e.g. TAM, Davis (1985)), something that had not been done for situations in which the adopter is only an influencer in the user’s decision making process. In a sellingrather than usage context, findings from the current research are similar to the findings of the TAM. It therefore adds to our understanding of product adoption in a selling- rather than usage context. However, because our results conflict with the findings of Ahearne et al. (2010), it remains unclear what the actual direction of the relationship is. Next to that, this research adds to existing literature the influence of learning- and performance orientation on product adoption. While their effect on effort has been investigated (e.g. Sujan et al., 1994; VandeWalle et al., 1999), their relationship with product adoption and commitment had not been investigated before. Contrary to expectations though, no significant effect between either one of the two orientations and adoption occurs. Third, we are the first to investigate the influence of the control system on product adoption, a perspective that did not receive any attention yet. In line with the agency theory (e.g. Anderson & Oliver, 1987), behavioral control positively influences product adoption. These findings indicate that the type of control system not only has a profound impact once a product is adopted (e.g. Hultink & Atuahene-Gima, 2000; Hultink et al., 2000), but control systems also directly influence adoption. Fourth, the current research adds to existing literature the introduction of involvement. While adoption is conceptualized as pro-change behavior and involvement has been shown to foster pro-change behavior, no direct effect is found between involvement and adoption in the current research. However, involvement influences sales performance both directly and indirectly via product newness, which is an important theoretical implication. Finally, the difference between commitment and effort, their relation to sales performance, and how antecedents relate to these constructs has been investigated. To our knowledge, this comparison has not been made yet. We investigated how our conceptualization of adoption relates to behavioral theories such as the TRA, and adoption 48

theories such as the TAM. From a theoretical perspective, and in line with Fu et al. (2010), it might be hypothesized that including intentions would lead to a better conceptualization of what ‘SFNP adoption’ intends to represent. Based on the empirical results however, we can conclude that the current conceptualization of adoption strongly relates to the TRA and the TAM. However, because stronger similarities exist between adoption and commitment, the proper representation of product adoption by the sales force still remains unclear. 5.2.2

Managerial Implications

Next to these theoretical implications, this research provides some valuable insights for sales managers, as well as those who are responsible for product launches. Most importantly, this research emphasizes the importance of a positive product perception by salespeople. As can be seen in Figure 5, do both product-related constructs strongly influence the sales performance, either directly (product newness) or indirectly (via adoption). It thus is essential to establish a positive perception of the product amongst those who are in touch with an organization’s customers. Assuming an organization strives to develop the best products, the findings of this research imply two ways to increase the product perception. First of all, the social environment strongly influences the product perception. The stronger a salesperson thinks others consider selling the new product to be important, or are motivated to comply with wishes of others, the more positive their perception of the product. Second, product newness is negatively related to involvement in the development process. This means that if a salesperson is involved in the development process to a greater extent, they believe customers will find a new product more compatible with existing experiences. Additionally, involvement also directly influences the sales performance. So how can the perception of being involved be increased? Combining the influence of being involved and the impact of the social environment, managers should do several things. Because most organizations are too big to involve everyone, managers should find out which individuals are decision makers and/or opinion leaders. As Rogers (1962) showed, “opinion leaders are individuals who lead in influencing others' opinions about innovations” (p271). These opinion leaders (within the different reference groups) should be involved such that they influence the product perception of others. These opinion leaders themselves thus are the ‘social influences’ described under subjective norms. How can the product perception of these opinion leaders be shaped? Importantly, as Wanberg & Banas (2000) mention, the reason for change should be communicated such that the consequences of the change are less uncertain. Additionally, managers and product developers should emphasize the importance of the opinion leaders’ input in the development process, because they really know the customers, and their needs. Therefore, these influential individuals should be encouraged to ask their customers about their needs, and they should motivate their colleagues to do the same. Once information about the customers’ needs has been gathered, feedback should be provided on what has been achieved using their input. Last, once a product has been developed and is being tested, these opinion leaders should again be involved in the testing and launch phase, for example by receiving training on how to sell it, or let them attend at meetings concerning launch strategies. During this ‘involvement process’, all employees should receive updates about their involvement regularly, they should be trained in how to gather customer information, organizations should have guidelines for the collection process, and reporting methods for 49

the feedback provided by customers should be established (Judson et al., 2006). Unfortunately, customer relationship management (CRM) systems are rarely used to communicate information about customer needs, while for less influential individuals, these systems could foster the perception of being involved in the development process. Therefore, companies should seek means to motivate employees to provide feedback in such CRM systems by providing incentives (Judson et al., 2006). Another important implication of this research relates to the set of procedures an organization uses to monitor, direct, evaluate, and compensate its employees. This research shows that focusing on procedures positively influences the degree to which salespeople adopt a new product. Related to the selling of a new product, this means that managers should actively guide a salesperson, evaluate their behavior, and provide feedback on how to accomplish performance goals. However, because appropriate flexibility should be provided to the salespeople, caution is needed with this approach, as indicated by Ahearne et al. (2010). Last, the findings of this research indicate that perceived involvement is negatively influenced by learning orientation. As mentioned before can this be explained by a timingaspect. This finding shows that managers should make a conscious trade-off between demanding a salesperson’s time on learning to improve their skills or being involved in the development process.

5.3 Limitations and future research This research attempts to shed light on factors influencing product adoption by customer-contact employees including factors related to the product, and the environment in which these employees operate. Despite the promising findings, highlighting a number of limitations and recommendations for future research appears to be important to judge these findings on their value. Several limitations relate to the data used. First, a relatively small sample size is used (N=97). Increasing the sample size would increase the strength of the results and conclusions. Second, the normality-assumption is violated. With few exceptions, data were either skewed, or kurtosis occurred. Furthermore, the reversed item bias occurs. Although PLSSEM is robust against a violation of this assumption, the results and conclusions should be interpreted with caution. Third, the data seem to be subject to the common method bias and the reversed item bias. Although Podsakoff et al. (2010) suggest that not all measures are affected by response biases equally the findings of this research should be interpreted with caution. Fourth, self-reported measures are used. Particularly the self-report measure of sales performance deserves special attention. Because Churchill, Ford, Hartley, & Walker (1985) find evidence for the validity and usefulness of such measures, and because the measures have been used in prior research successfully, these measures are used. Additionally, because our findings conflict with the findings of Ahearne and his colleagues (who used objective measures for effort), objective could increase the reliability of the research. Fifth, these data were collected at one point in time. Because predictor and outcome variables were collected from one person, together with the third limitation, this means this research could be subject to the common method bias. Unfortunately though, balancing 50

positive and negative items (a procedure to control for the common method bias) lead to a bias referred to as the reversed item bias. This means participants had the tendency to agree with items regardless of its statement. Although Podsakoff et al. (2012) indicate that not all measures are affected by response biases equally, this bias decreases the reliability of the findings. Future research should use objective performance measures, obtain data from different sources, separate predictor and criterion, eliminate common scale properties, improve scale item to eliminate ambiguity, reducing social desirability, and balancing positive and negative items (Podsakoff et al., 2012). This should control for different sources of the method bias, and therefore should make findings more reliable. Sixth, these data were gathered within a single company. Although we are grateful for AkzoNobel to make this research possible, these findings apply to a company that operates in a business-to-business market that is very mature. Therefore, precaution is needed when generalizing these results. Future research could shed light on whether these findings differ for different kinds of products. One can for example investigate whether findings in a business-to-business setting differ from findings in a business-to-consumer environment, whether there is a difference between technological and physical products, and in line with Fu et al. (2010) whether products that are new-to-market or are a line extension differ. Another opportunity is to investigate whether there are differences between ‘pushed’ products and ‘pulled’ products. The above describes limitations related to the data used. Below, several recommendations for future research, not related to the data, are described. First, as mentioned, data are gathered at one point is time. Because a cross-sectional rather than a longitudinal research design is used, conclusions about causality, and how relationships between variables change over time cannot be made. Future research should address this shortcoming. It could for example investigate the influence of different stages of the process (development, testing, launch etc.) or the maturity of a market, but also how the influence of different categories of antecedents changes over time. It might for example be reasoned that the success of a product influences the degree to which salespeople adopt the new product, and a feedback-loop therefore could occur. Furthermore, it might be interesting to investigate when involvement yields the highest effect on performance. Second, we conclude that our conceptualization of adoption is not in line with behavioral theories such as the TRA and the TAM, and the inclusion of intentions might lead to a better representation of what adoption intends to represent. Future research should investigate how, in the context of selling new products, adoption can best be conceptualized. Third, the current research only investigates the effect of adoption on sales performance. Although this is important, future research could elaborate on other outcome variables of product adoption such as job satisfaction or customer satisfaction. Fourth, few possible antecedents of adoption were investigated in this research. While the TPB postulates three determinants of intentions (attitudes, subjective norms, and perceived behavioral control), we only included attitudes and subjective norms. Future research should include perceived behavioral control, specifically because this concept directly relates to behavior (see Ajzen, 1991). Fifth, while this research takes the perspective of the salesperson, it does not shed light on a sales manager’s (or any other stakeholder’s) perception of SFNP adoption and how they might influence this process. Future research could investigate a network effect: do

51

influences differ per organizational department, and which people have the most profound influence of a salesperson’s product perception. Sixth, because both product newness and sales performance are influenced by involvement, future research should investigate factors that influence involvement, whether involvement influences any other constructs than the ones covered in this research, if involvement during different stages (testing, launch) yield different results, and how different types of involvement (see for example Alam (2002)) influence different constructs. Seventh, salespeople have to assess the value of a new product for someone else, namely their customers. While Ahearne et al. (2010) investigated both a salesperson’s-, and a customer’s perception of the new product, we only elaborated on the salesperson’s perception. Future research could investigate whether there is a difference between these two perceptions and how this might influence for example sales performance. A research setup used by Mullins, Ahearne, Lam, Hall, & Boichuk (2014) could be used in this case. Finally, while participants originated from six countries, cultural differences were not taken into account. It can be argued that the degree of power distance influences the effect of subjective norms, the influence of a control system or how people approach their goal. Cultural differences are therefore worth to investigate. If these shortcomings are addressed by future research, it will add to the understanding of the new product selling process.

52

References

Abramovici, M. & Bancel-Charensol, L. (2004). How to take customers into consideration in service innovation projects. The Service Industries Journal, 24(1), 56–78. Agarwal, R. & Prasad, J. (1998). The antecedents and consequents of user perceptions in information technology adoption. Decision Support Systems, 22(1), 15–29. Ahearne, M., Rapp, A., Hughes, D. E. & Jindal, R. (2010). Managing sales force product perceptions and control systems in the success of new product introductions. Journal of Marketing Research, 47(4), 764–776. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Processes, 50(2), 179–211.

Decision

Alam, I. (2002). An exploratory investigation of user involvement in new service development. Journal of the Academy of Marketing Science, 30(3), 250-261. Anderson, E. & Robertson, T. S. (1995). Inducing multiline salespeople to adopt house brands. The Journal of Marketing, 16–31. Armstrong, J. S., & Overton, T. S. (1977). Estimating nonresponse bias in mail surveys. Journal of marketing research, 396-402. Atuahene‐Gima, K. (1995). An exploratory analysis of the impact of market orientation on new product performance. Journal of product innovation management, 12(4), 275-293. Atuahene-Gima, K. (1996). Market orientation and innovation. Journal of Business Research, 35(2), 93–103. Atuahene-Gima, K. (1997). Adoption of new products by the sales force: the construct, research propositions, and managerial implications. Journal of Product Innovation Management, 14(6), 498–514. Atuahene-Gima, K., & Li, H. (2002). When does trust matter? Antecedents and contingent effects of supervisee trust on performance in selling new products in China and the United States. Journal of Marketing, 66(3), 61-81. Atuahene-Gima, K. & Micheal, K. (1998). A contingency analysis of the impact of salesperson’s effort on satisfaction and performance in selling new products. European Journal of Marketing, 32(9), 904–921. Avlonitis, G. J. & Gounaris, S. P. (1997). Marketing orientation and company performance: industrial vs. consumer goods companies. Industrial Marketing Management, 26(5), 385–402. Borgh, W. van der (2012). Selling new products (Doctoral dissertation). Technische Universiteit Eindhoven, Eindhoven.

53

Brown, S. P. & Peterson, R. A. (1994). The effect of effort on sales performance and job satisfaction. The Journal of Marketing, 70–80. Bullinger, M., Anderson, R., Cella, D., & Aaronson, N. (1993). Developing and evaluating cross‐ cultural instruments from minimum requirements to optimal models. Quality of Life Research, 2(6), 451‐459. Burkhardt, M. E. (1994). Social interaction effects following a technological change: A longitudinal investigation. Academy of Management Journal, 37(4), 869–898. Button, S. B., Mathieu, J. E. & Zajac, D. M. (1996). Goal orientation in organizational research: A conceptual and empirical foundation. Organizational Behavior and Human Decision Processes, 67(1), 26–48. Churchill, G. A., Ford, N. M., Hartley, S. W., and Walker, O. C. The determinants of salesperson performance: A meta-analysis. Journal of Marketing Research 22, 103–118 (1985). Davis, F. D. (1985). A technology acceptance model for empirically testing new end-user information systems: Theory and results. Davis, F. D., Bagozzi, R. P. & Warshaw, P. R. (1989). User acceptance of computer technology: a comparison of two theoretical models. Management Science, 35(8), 982–1003. Di Benedetto, C. A. (1999). Identifying the key success factors in new product launch. Journal of Product Innovation Management, 16(6), 530–544. Dillman, D. A., Phelps, G., Tortora, R., Swift, K., Kohrell, J., Berck, J., & Messer, B. L. (2009). Response rate and measurement differences in mixed-mode surveys using mail, telephone, interactive voice response (IVR) and the Internet. Social Science Research, 38(1), 1-18. Dobosz-Bourne, D. and Jankowicz, A.D. (2006), “Reframing resistance to change: experience from general motors Poland”, International Journal of Human Resource Management, Vol. 17 No. 12, pp. 2021-34. Erez, M. & Arad, R. (1986). Participative goal-setting: Social, motivational, and cognitive factors. Journal of Applied Psychology, 71(4), 591 – 597. Engel, J. F., Blackwell, R. D., & Miniard, P. W. (1995). Consumer behavior, 8th. New York: Dryder. Esposito Vinzi, V., Chin, W. W., Henseler, J., & Wang, H. (2010). Handbook of partial least squares: Concepts, methods and applications. Farr, J. L., Hofmann, D. A., & Ringenbach, K. L. (1993). Goal orientation and action control theory: Implications for industrial and organizational psychology. International review of industrial and organizational psychology, 8(2), 193-232. Field, A. (2009). Discovering statistics using SPSS. Sage publications. Fielding, K. S., McDonald, R., & Louis, W. R. (2008). Theory of planned behavior, identity and intentions to engage in environmental activism. Journal of Environmental Psychology, 28(4), 318‐ 326.

54

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An introduction to theory and research. Fornell, C. & Larcker, D. F. (1981). "Evaluating structural equation models with unobservable variables and measurement error". Journal of marketing research, p. 39–50. Fu, F. Q., Jones, E. & Bolander, W. (2008). Product innovativeness, customer newness, and new product performance: a time-lagged examination of the impact of salesperson selling intentions on new product performance. Journal of Personal Selling and Sales Management, 28(4), 351–364. Fu, F. Q., Richards, K. A., Hughes, D. E. & Jones, E. (2010). Motivating salespeople to sell new products: the relative influence of attitudes, subjective norms, and self-efficacy. Journal of Marketing, 74(6), 61–76. Fu, F. Q., Richards, K. A. & Jones, E. (2009). The motivation hub: Effects of goal setting and selfefficacy on effort and new product sales. Journal of Personal Selling \& Sales Management, 29(3), 277–292. Gatignon, H., & Robertson, T. S. (1989). Technology diffusion: an empirical test of competitive effects. The Journal of Marketing, 35-49. Giangreco, A., & Peccei, R. (2005). The nature and antecedents of middle manager resistance to change: evidence from an Italian context. The international journal of human resource management, 16(10), 1812-1829. Glew, D.J, O’Leary-Kelly, A.M., Griffin, R.W. and Van Fleet, D.D. (1995) Participation in organizations: a preview of the issues and proposed framework for analysis, Journal of Management, 21(3), pp. 395–421. Gourville, J. T. (2006). Eager Sellers, Stony Buyers, Harvand Business Review, 99-106. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2010). Multivariate data analysis. Engelwood Cliffs: Prentice-Hall. Hair, J. F., Busch, R. P. & Ortinau, D. J. (2006). Marketing Research. Within a Changing Information Environment. New York: McGraw‐Hill Irwin. Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. The Journal of Marketing Theory and Practice, 19(2), 139-152. Hair, J. F., Sarstedt, M., Ringle, C. M., & Mena, J. A. (2012). An assessment of the use of partial least squares structural equation modeling in marketing research. Journal of the Academy of Marketing Science, 40(3), 414-433. Haslam, S. A. (2004). Psychology in organizations. Sage. Henseler, J., Ringle, C. & Sinkovics, R. (2009). "The use of partial least squares path modeling in international marketing". Advances in International Marketing (AIM), 20, p. 277–320

55

Herold, D. M., Fedor, D. B., & Caldwell, S. D. (2007). Beyond change management: a multilevel investigation of contextual and personal influences on employees' commitment to change. Journal of Applied Psychology, 92(4), 942. Homburg, C., Wieseke, J. & Kuehnl, C. (2010). Social influence on salespeople’s adoption of sales technology: a multilevel analysis. Journal of the Academy of Marketing Science, 38(2), 159–168. Hulland, J. (1999). "Use of partial least squares (PLS) in strategic management research: a review of four recent studies". Strategic management journal, 20(2), p. 195–204. Hultink, E. J. & Atuahene-Gima, K. (2000). The effect of sales force adoption on new product selling performance. Journal of Product Innovation Management, 17(6), 435–450. Hultink, E. J., Atuahene-Gima, K. & Lebbink, I. (2000). Determinants of new product selling performance: an empirical examination in The Netherlands. European Journal of Innovation Management, 3(1), 27–36. Jaramillo, F., Mulki, J. P. & Marshall, G. W. (2005). A meta-analysis of the relationship between organizational commitment and salesperson job performance: 25 years of research. Journal of Business Research, 58(6), 705–714. Jaworski, B. J., & MacInnis, D. J. (1989). Marketing jobs and management controls: toward a framework. Journal of Marketing Research, 406-419. Joshi, A. W. (2010). Salesperson influence on product development: insights from a study of small manufacturing organizations. Journal of Marketing, 74(1), 94-107. Judson, Kimberly, Denise D. Schoenbachler, Geoffrey L. Gordon, Rick E. Ridnour, and Dan C. Weilbaker (2006), “The New Product Development Process: Let the Voice of the Salesperson be Heard,” Journal of Product & Brand Management, 15 (2–3), 194–202. Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica: Journal of the Econometric Society, 263-291. Kelman, H. C. (1958). Compliance, identification, and internalization: Three processes of attitude change. Journal of conflict resolution, 51-60. Kraut, R. E., Rice, R. E., Cool, C. & Fish, R. S. (1998). Varieties of social influence: The role of utility and norms in the success of a new communication medium. Organization Science, 9(4), 437–453. Krishnan, B. C., Netemeyer, R. G. & Boles, J. S. (2002). Self-efficacy, competitiveness, and effort as antecedents of salesperson performance. Journal of Personal Selling & Sales Management, 22(4), 285–295. Latham, G. P., & Pinder, C. C. (2005). Work motivation theory and research at the dawn of the twenty-first century. Annu. Rev. Psychol., 56, 485-516. Leong, S. M., Randall, D. M. & Cote, J. A. (1994). Exploring the organizational commitment— Performance linkage in marketing: A study of life insurance salespeople. Journal of Business Research, 29(1), 57–63.

56

Lines, R. (2004). Influence of participation in strategic change: resistance, organizational commitment and change goal achievement. Journal of Change Management, 4(3), 193–215. Lu, J., Yao, J. E. & Yu, C.-S. (2005). Personal innovativeness, social influences and adoption of wireless Internet services via mobile technology. The Journal of Strategic Information Systems, 14(3), 245– 268. Ma, Q. & Liu, L. (2004). The technology acceptance model: a meta-analysis of empirical findings. Journal of Organizational and End User Computing (JOEUC), 16(1), 59–72. Mathieu, J. E. & Zajac, D. M. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108(2), 171. Mayer, R. C. & Schoorman, F. D. (1998). Differentiating antecedents of organizational commitment: a test of March and Simon’s model. Journal of Organizational Behavior, 19(1), 15–28. Meece, Judith L., Phyllis C. Blumenfeld, and Rick H. Hoyle (1988), "Students' Goal Orientations and Cognitive Engagement in Classroom Activities", Journal of Educational Psychology, 80 (4), 514-23. Moore, G. C. & Benbasat, I. (1991). Development of an instrument to measure the perceptions of adopting an information technology innovation. Information Systems Research, 2(3), 192–222. Moriarty, R. T. & Kosnik, T. J. (1989). High-tech marketing: concepts, continuity, and change. Sloan Management Review, 30(4), 7–17. Mullins, R. R., Ahearne, M., Lam, S. K., Hall, Z. R., & Boichuk, J. P. (2014). Know Your Customer: How Salesperson Perceptions of Customer Relationship Quality Form and Influence Account Profitability. Journal of Marketing, 78(6), 38-58. Nunnally, J. C. & Bernstein, I. H. (1991). "Psychometric theory. 1994". McGraw, New York. Oliver, R. L., & Anderson, E. (1994). An empirical test of the consequences of behavior-and outcomebased sales control systems. The Journal of Marketing, 53-67. Oreg, S. (2003). Resistance to change: developing an individual differences measure. Journal of Applied Psychology, 88(4), 680–693. Pauwels, K., Silva-Risso, J., Srinivasan, S. & Hanssens, D. M. (2004). New products, sales promotions, and firm value: The case of the automobile industry. Journal of Marketing, 68(4), 142–156. Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of applied psychology, 88 (5), 879. Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources of method bias in social science research and recommendations on how to control it. Annual review of psychology, 63, 539-569. Pugh, D. (1993) ‘Understanding and Managing Organizational Change’. In Mabey, C. and MayonWhite, B. (eds) Managing Change. London: Paul Chapman Publishing Ltd in association with The Open University. 57

Raubenheimer, J. (2004). An item selection procedure to maximize scale reliability and validity. SA Journal of Industrial Psychology, 30(4), 59-64. Renn, R. W. (1998). Participation’s effect on task performance: Mediating roles of goal acceptance and procedural justice. Journal of Business Research, 41(2), 115–125. Robinson Jr, L., Marshall, G. W. & Stamps, M. B. (2005). Sales force use of technology: antecedents to technology acceptance. Journal of Business Research, 58(12), 1623–1631. Rogers, E. M. (1962). Diffusion of innovations. Simon and Schuster. Sagie, A., & Koslowsky, M. (2000). Participation and empowerment in organizations: Modeling, effectiveness, and applications. Thousand Oaks, CA: Sage Samli, A. C., Wirth, G. P. & Wills, J. R. (1994). High-tech firms must get more out of their international sales efforts: An exploration in developing a competitive edge. Industrial Marketing Management, 23(4), 333–342. Savery, L. K. (1994). The Influence of the Perceived Styles of Leadership on a Group of Workers on their Attitudes to Work. Leadership & Organization Development Journal, 15(4), 12–18. Schepers, J. & Wetzels, M. (2007). A meta-analysis of the technology acceptance model: Investigating subjective norm and moderation effects. Information & Management, 44(1), 90–103. Schillewaert, N., Ahearne, M. J., Frambach, R. T. & Moenaert, R. K. (2005). The adoption of information technology in the sales force. Industrial Marketing Management, 34(4), 323–336. Searfoss, D. G. & Monczka, R. M. (1973). Perceived participation in the budget process and motivation to achieve the budget. Academy of Management Journal, 16(4), 541–554. Shum, P., Bove, L. and Auh, S. (2008), “Employee’s affective commitment to change: the key to successful CRM implementation”, European Journal of Marketing, Vol. 42 Nos 11/12, pp. 1346-71. Singh, J. (1998). Striking a balance in boundary-spanning positions: An investigation of some unconventional influences of role stressors and job characteristics on job outcomes of salespeople. The Journal of Marketing, 69–86. Singleton, Jr., R. A. & Straits, B. C. (2005). Approaches to social research. New York: Oxford University Press Straub, E. T. (2009). Understanding technology adoption: Theory and future directions for informal learning. Review of Educational Research, 79(2), 625–649. Sujan, H., Weitz, B. A. & Kumar, N. (1994). Learning orientation, working smart, and effective selling. The Journal of Marketing, 39–52. Van Dam, K., Oreg, S. & Schyns, B. (2008). Daily work contexts and resistance to organizational change: The role of leader-member exchange, development climate, and change process characteristics. Applied Psychology, 57(2), 313–334.

58

VandeWalle, D., Brown, S. P., Cron, W. L. & Slocum Jr, J. W. (1999). The influence of goal orientation and self-regulation tactics on sales performance: A longitudinal field test. Journal of Applied Psychology, 84(2), 249 – 259. Venkatesh, V. & Davis, F. D. (2000). A theoretical extension of the technology acceptance model: four longitudinal field studies. Management Science, 46(2), 186–204. Venkatesh, V., Morris, M. G., Davis, G. B. & Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 425–478. Wanberg, C. R., & Banas, J. T. (2000). Predictors and outcomes of openness to changes in a reorganizing workplace. Journal of Applied Psychology, 85(1), 132-142. Weijters, B., Baumgartner, H., & Schillewaert, N. (2013). Reversed item bias: An integrative model. Psychological methods, 18(3), 320-334. Weijters, B., Geuens, M., & Schillewaert, N. (2009). The proximity effect: The role of inter-item distance on reverse-item bias. International Journal of Research in Marketing, 26(1), 2-12. Wieseke, J., Homburg, C. & Lee, N. (2008). Understanding the adoption of new brands through salespeople: a multilevel framework. Journal of the Academy of Marketing Science, 36(2), 278– 291. Zablah, A. R., Chonko, L. B., Bettencourt, L. A., Allen, G. & Haas, A. (2012). A job Demands-Resources (JD-R) perspective on new product selling: a framework for future research. Journal of Personal Selling & Sales Management, 32(1), 73–87.

59

Appendices

A.

Measures .......................................................................................................... 61

B.

Test for normality – Skewness & Kurtosis.......................................................... 64

C.

Exploratory factor analysis ................................................................................ 65

D.

Discriminant validity ........................................................................................... 66 1.

Fornell – Larcker criterion .............................................................................. 66

2.

Cross-loadings ............................................................................................... 66

E.

Analysis for moderation ..................................................................................... 68

F.

Additional paths investigated ............................................................................. 70

G.

Early versus late response ................................................................................ 71

H.

Notifications....................................................................................................... 72

I.

1.

Pre-notification to customer-contact employee............................................... 72

2.

Pre-notification to sales manager ................................................................... 73

3.

Invitation to customer-contact employee ........................................................ 74

4.

Notification invitation to sales manager .......................................................... 75

5.

First reminder to customer-contact employee ................................................ 75

6.

First reminder to sales manager..................................................................... 76

7.

Second reminder to customer-contact employee ........................................... 76

8.

Second reminder to sales manager ............................................................... 77

List of antecedents mentioned in literature ............................................................ 78

60

A. Measures Construct Sales performance SP1 SP2 SP3 SP4 Commitment Com1 Com2 Com3 Com4 Effort Eff1 Eff2 Eff3 Eff4 Product Advantage PA1 PA2 PA3 Product Newness PN1 PN2 PN3 PN4

Statement

Adapted from

Achieving targets related to the turnover of this new product. Achieving targets related to the market penetration of this new product. Achieving targets related the number of customers using this new product. When you compare your performance with the performance of others in your agency, you are in the..(1) Upper 10%, (2) Upper 1/3, (3) Middle 1/3, (4) Lower 1/3, (5) Lower 10%.

Sujan et al. (1994) Sujan et al. (1994) Sujan et al. (1994) Oliver & Anderson (1994) Hultink & Atuahene-Gima (2000)

I feel emotionally attached to the success of this new product. Achieving objectives for this new product has a great deal of personal meaning to me. I enjoy discussing this new product with other salespeople. I feel a strong sense of duty to ensure the success of this new product. Hultink & Atuahene-Gima (2000) Planning sales calls Collecting market information Using market information Building customer relationships The new product offers unique benefits to customers. The new product offers benefits superior to the competition. Overall, customers will find the new product useful.

Atuahene-Gima (1996) Atuahene-Gima (1996) Jones et al. (2002)

It takes time before customers really understand the advantages of the product. (Rev.) Learning to use the new product is easy for the customer. The use of the new product is self-explanatory. The use of this product requires significant prior knowledge or training. (Rev.)

Atuahene-Gima (1995) Robinson Jr. et al. (2005) Added Added

61

Subjective Norms SNoth1 SNoth2 SNoth3 SNoth4 SNmot1 SNmot2 SNmot3 SNmot4 Goal Orientation LO1 LO2 LO3 LO4 PO1 PO2 PO3 PO4 Control System CSOC1 CSOC2 CSOC3 CSOC4 CSBC1 CSBC2 CSBC3 CSBC4

Fu et al. (2010) I think ‘my sales manager’ considers selling the new product to be 'very important'. My sales manager My product manager Fellow sales reps. A technical consultant I am motivated 'to a great extent' to comply with the wishes of 'the sales manager'. My sales manager My product manager Fellow sales reps. A technical consultant VandeWalle et al. (1999) An important part of being a good salesperson is continually improving your sales skills. It is worth spending a great deal of time learning new approaches for dealing with customers. Learning how to be a better salesperson is of fundamental importance to me. I put in a great deal of effort sometimes in order to learn something new. It is very important to me that my supervisor sees me as a good salesperson. I very much want my co-workers to consider me to be good at selling. I always try to communicate my accomplishments to my manager. I spend a lot of time thinking about how my performance compares with other salespeople's. Jaworski & MacInnis (1989) My immediate boss monitors the extent to which I attain my performance goals. If my performance goals were not met, I would be required to explain why. I receive feedback from my immediate superior concerning the extent to which I achieve my goals. My pay increases are based upon how my performance compares with my goals. My immediate boss monitors the extent to which I follow established procedures. My immediate boss evaluates the procedures I use to accomplish a given task. My immediate boss modifies my procedures when desired results are not obtained. I receive feedback on how I accomplish my performance goals. 62

Perceived Involvement PerInv1 PerInv2 PerInv3 PerInv4 Additional information Age Func TenSec TenFunc TenAkz Country

Product developers have asked me about any product specifications I think should be considered in the new product. The new product includes changes I have suggested. I was involved in the development of the new product. I provided a manager or someone from another department with information, such that the new product could be better developed.

Age Function Tenure Sector Tenure Function Tenure Akzo Country

63

Searfoss & Monczka, 1973 Searfoss & Monczka, 1973 Searfoss & Monczka, 1973 Added

B. Test for normality – Skewness & Kurtosis KolmogorovSmirnov Statistic

Sig.

Shapiro-Wilk Statistic

Skewness

St. error of Skewness

Kurtosis

Sig.

df = 97

St. error of Kurtosis

Zvalue Skewness

Zvalue Kurtosis

SP1

0,179

0,000

0,901

0,000

-0,656

0,245

-0,678

0,485

-2,677

-1,398

SP2

0,174

0,000

0,906

0,000

-0,616

0,245

-0,727

0,485

-2,516

-1,498

SP3

0,156

0,000

0,909

0,000

-0,529

0,245

-0,846

0,485

-2,159

-1,742

SP4

0,215

0,000

0,861

0,000

0,568

0,245

-0,241

0,485

2,318

-0,496

Com1

0,251

0,000

0,832

0,000

-0,876

0,245

0,569

0,485

-3,576

1,172

Com2

0,273

0,000

0,804

0,000

-0,991

0,245

1,801

0,485

-4,046

3,712

Com3

0,284

0,000

0,756

0,000

-1,371

0,245

2,477

0,485

-5,595

5,104

Com4

0,325

0,000

0,733

0,000

-1,432

0,245

1,758

0,485

-5,847

3,623

Eff1

0,269

0,000

0,859

0,000

-0,780

0,245

0,608

0,485

-3,184

1,252

Eff2

0,224

0,000

0,895

0,000

-0,446

0,245

-0,154

0,485

-1,821

-0,317

Eff3

0,208

0,000

0,893

0,000

-0,379

0,245

0,050

0,485

-1,549

0,103

Eff4

0,254

0,000

0,873

0,000

-0,692

0,245

0,389

0,485

-2,824

0,801

PA1

0,242

0,000

0,799

0,000

-1,350

0,245

1,849

0,485

-5,510

3,809

PA2

0,188

0,000

0,878

0,000

-0,851

0,245

0,023

0,485

-3,474

0,048

PA3

0,284

0,000

0,778

0,000

-1,619

0,245

2,905

0,485

-6,608

5,985

PN1

0,179

0,000

0,904

0,000

-0,660

0,245

-0,422

0,485

-2,695

-0,869

PN2

0,160

0,000

0,918

0,000

-0,579

0,245

-0,385

0,485

-2,362

-0,793

PN3

0,128

0,000

0,949

0,001

0,003

0,245

-0,728

0,485

0,011

-1,501

PN4

0,148

0,000

0,937

0,000

-0,285

0,245

-0,750

0,485

-1,162

-1,544

SNoth1

0,324

0,000

0,739

0,000

-1,385

0,245

1,445

0,485

-5,655

2,976

SNoth2

0,320

0,000

0,750

0,000

-1,280

0,245

0,984

0,485

-5,226

2,028

SNoth3

0,243

0,000

0,812

0,000

-1,043

0,245

0,701

0,485

-4,259

1,445

SNoth4

0,312

0,000

0,718

0,000

-1,617

0,245

2,665

0,485

-6,601

5,490

SNmot1

0,349

0,000

0,717

0,000

-1,262

0,245

0,573

0,485

-5,151

1,180

SNmot2

0,248

0,000

0,816

0,000

-0,916

0,245

-0,108

0,485

-3,738

-0,223

SNmot3

0,260

0,000

0,801

0,000

-1,076

0,245

0,517

0,485

-4,394

1,064

SNmot4

0,323

0,000

0,732

0,000

-1,478

0,245

1,972

0,485

-6,034

4,063

LO1

0,410

0,000

0,569

0,000

-2,618

0,245

8,482

0,485

-10,686

17,477

LO2

0,257

0,000

0,728

0,000

-2,098

0,245

7,248

0,485

-8,563

14,934

LO3

0,314

0,000

0,755

0,000

-1,124

0,245

0,564

0,485

-4,588

1,163

LO4

0,357

0,000

0,700

0,000

-1,539

0,245

2,901

0,485

-6,284

5,977

PO1

0,339

0,000

0,722

0,000

-1,470

0,245

1,946

0,485

-6,000

4,009

PO2

0,276

0,000

0,797

0,000

-0,978

0,245

0,007

0,485

-3,994

0,014

PO3

0,169

0,000

0,887

0,000

-0,546

0,245

0,025

0,485

-2,227

0,052

PO4

0,161

0,000

0,908

0,000

-0,680

0,245

-0,147

0,485

-2,776

-0,303

CSOC1

0,253

0,000

0,798

0,000

-0,886

0,245

0,308

0,485

-3,618

0,634

CSOC2

0,245

0,000

0,805

0,000

-0,771

0,245

0,325

0,485

-3,145

0,670

CSOC3

0,250

0,000

0,816

0,000

-0,710

0,245

-0,296

0,485

-2,897

-0,610

CSOC4

0,214

0,000

0,860

0,000

-0,671

0,245

-0,517

0,485

-2,741

-1,065

CSBC1

0,234

0,000

0,837

0,000

-0,550

0,245

-0,370

0,485

-2,245

-0,762

CSBC2 CSBC3

0,247 0,238

0,000 0,000

0,849 0,890

0,000 0,000

-0,465 -0,457

0,245 0,245

-0,399 -0,619

0,485 0,485

-1,899 -1,865

-0,822 -1,275

64

CSBC4

0,219

0,000

0,852

0,000

-0,442

0,245

-0,664

0,485

-1,806

-1,368

PerInv1

0,262

0,000

0,746

0,000

1,456

0,245

1,249

0,485

5,944

2,574

PerInv2

0,316

0,000

0,712

0,000

1,573

0,245

1,743

0,485

6,419

3,591

PerInv3

0,371

0,000

0,650

0,000

1,710

0,245

1,937

0,485

6,980

3,991

PerInv4

0,206

0,000

0,846

0,000

0,474

0,245

-1,245

0,485

1,936

-2,565

C. Exploratory factor analysis Settings: Eigenvalue>1, rotation = oblique.

65

D. Discriminant validity 1. Fornell – Larcker criterion

SP COM EFF Pr.Adv. Pr.New. SNoth SNmot GOLearn GOPerf CSOut CSBeh

SP

COM

0,857 0.402 0.315 0.340 0.504 0.255 0.264 0.210 0.171 0.101 0.104

0,882 0.496 0.621 0.459 0.485 0.490 0.481 0.381 0.430 0.438

EFF

0,874 0.457 0.380 0.334 0.352 0.199 0.282 0.209 0.323

Pr. Adv.

0,901 0.455 0.437 0.289 0.313 0.215 0.213 0.174

Pr. New.

SNoth

SNmot

GO Learn

GO Perf

CS Outc.

CS Beh.

PI

AVE-values SP COM EFF Pr.Adv. Pr.New. SNoth SNmot GOLearn GOPerf CSOut CSBeh

0,653 0.347 0.321 0.153 0.167 0.155 0.160

0,735 0,778 0,764 0,811 0,426 0,737 0,763 0,631 0,583 0,576 0,667

0,858 0.528 0,873 0.374 0.352 0,794 0.256 0.456 0.614 0,764 0.462 0.422 0.456 0.492 0,759 0.385 0.434 0.466 0.527 0.759 0,817 PI 0.260 0.090 0.164 0.156 0.296 0.129 0.021 0.031 0.079 0,819 PI 0,671 0.057 0.180 Note(i): square root of AVE-value of latent variable on diagonal. Note(ii): the values of the correlations are without items being deleted. Table 6 displays correlations between constructs used in the model.

2. Cross-loadings SP1 SP2 SP3 SP4 Com1 Com2 Com3 Com4 Eff1 Eff2 Eff3 Eff4 PA1 PA2 PA3 PN1 PN2 PN3 PN4 SNoth1 SNoth2 SNoth3 SNoth4

Sales Perf. 0.967 0.960 0.968 -0.381 0.351 0.378 0.302 0.396 0.381 0.205 0.227 0.290 0.298 0.219 0.383 -0.137 0.532 0.383 0.051 0.166 0.189 0.274 0.243

Commit 0.394 0.346 0.401 -0.233 0.851 0.881 0.886 0.910 0.477 0.366 0.381 0.513 0.605 0.402 0.657 -0.186 0.503 0.243 -0.283 0.393 0.393 0.419 0.463

Effort 0.317 0.295 0.330 -0.007 0.333 0.414 0.549 0.431 0.842 0.916 0.897 0.847 0.460 0.319 0.420 -0.233 0.368 0.275 -0.193 0.370 0.315 0.273 0.202

Prod. Adv. 0.319 0.371 0.308 -0.089 0.457 0.545 0.622 0.551 0.471 0.328 0.395 0.405 0.923 0.859 0.904 -0.330 0.488 0.266 -0.235 0.416 0.380 0.336 0.377

Prod. Newn. 0.506 0.472 0.460 -0.273 0.440 0.381 0.366 0.441 0.367 0.297 0.285 0.381 0.384 0.383 0.440 -0.201 0.932 0.874 -0.094 0.358 0.210 0.295 0.318

66

SN others 0.245 0.238 0.223 -0.214 0.380 0.502 0.369 0.466 0.303 0.252 0.271 0.345 0.423 0.291 0.479 -0.062 0.334 0.275 -0.095 0.895 0.758 0.888 0.886

SN motiv. 0.227 0.247 0.233 -0.313 0.413 0.466 0.370 0.486 0.308 0.275 0.286 0.364 0.344 0.087 0.318 -0.160 0.361 0.194 -0.052 0.456 0.346 0.504 0.493

Learning Orient. 0.183 0.201 0.182 -0.233 0.371 0.491 0.399 0.436 0.206 0.128 0.182 0.181 0.291 0.257 0.300 -0.078 0.162 0.074 -0.155 0.350 0.250 0.369 0.307

Perf. Orient. 0.144 0.154 0.153 -0.228 0.282 0.346 0.312 0.404 0.306 0.261 0.213 0.207 0.238 0.123 0.229 -0.015 0.234 0.054 0.032 0.201 0.246 0.261 0.178

Output Control 0.127 0.091 0.071 -0.054 0.343 0.417 0.304 0.459 0.221 0.132 0.102 0.276 0.214 0.069 0.267 -0.115 0.216 0.014 -0.123 0.345 0.378 0.430 0.433

Behav. control 0.073 0.113 0.116 -0.042 0.398 0.395 0.321 0.442 0.286 0.312 0.254 0.281 0.213 0.014 0.211 -0.134 0.190 0.034 -0.221 0.344 0.305 0.339 0.335

Perc. Involv. 0.252 0.236 0.268 -0.080 0.155 0.055 0.083 0.033 0.159 0.202 0.084 0.129 0.189 0.058 0.087 -0.333 0.307 0.186 -0.158 -0.073 -0.075 -0.011 -0.043

SNmot1 SNmot2 SNmot3 SNmot4 LO1 LO2 LO3 LO4 PO1 PO2 PO3 PO4 CSOC1 CSOC2 CSOC3 CSOC4 CSBC1 CSBC2 CSBC3 CSBC4 PerInv1 PerInv2 PerInv3 PerInv4

Sales Perf. 0.187 0.202 0.226 0.309 0.293 0.023 0.189 0.131 0.144 0.149 0.113 0.127 0.070 0.069 0.091 0.078 0.047 0.115 0.145 0.051 0.295 0.293 0.134 0.049

Commit 0.420 0.375 0.393 0.520 0.338 0.187 0.439 0.450 0.412 0.337 0.176 0.088 0.404 0.345 0.391 0.107 0.426 0.357 0.261 0.372 0.069 0.025 0.015 0.165

Effort 0.353 0.332 0.237 0.309 0.066 0.060 0.232 0.183 0.306 0.230 0.110 0.121 0.173 0.151 0.218 0.058 0.298 0.288 0.196 0.263 0.108 0.145 0.054 0.188

Prod. Adv. 0.339 0.152 0.196 0.317 0.282 0.102 0.256 0.296 0.269 0.179 0.020 0.044 0.235 0.091 0.166 0.142 0.212 0.181 0.059 0.099 0.162 0.128 0.018 0.127

Prod. Newn. 0.290 0.194 0.301 0.333 0.188 -0.011 0.176 0.079 0.175 0.155 0.078 0.037 0.171 0.039 0.154 0.078 0.159 0.108 0.175 0.093 0.305 0.331 0.153 0.097

SN others 0.511 0.443 0.455 0.436 0.192 0.370 0.262 0.388 0.285 0.177 0.176 0.090 0.465 0.234 0.411 0.246 0.366 0.345 0.152 0.361 -0.025 -0.040 -0.129 -0.055

67

SN motiv. 0.861 0.824 0.899 0.910 0.261 0.211 0.323 0.284 0.510 0.422 0.121 0.095 0.398 0.324 0.370 0.147 0.367 0.347 0.301 0.400 0.132 0.136 0.021 0.069

Learning Orient. 0.413 0.263 0.263 0.285 0.776 0.708 0.851 0.816 0.653 0.481 0.342 0.247 0.420 0.340 0.378 0.223 0.481 0.397 0.166 0.433 -0.133 -0.166 -0.245 -0.113

Perf. Orient. 0.457 0.357 0.363 0.413 0.485 0.378 0.593 0.439 0.919 0.889 0.617 0.519 0.412 0.392 0.343 0.387 0.514 0.493 0.185 0.475 0.059 0.004 -0.048 0.005

Output Control 0.419 0.399 0.349 0.309 0.342 0.193 0.412 0.411 0.504 0.345 0.389 0.235 0.835 0.717 0.889 0.550 0.690 0.640 0.451 0.672 0.041 0.005 -0.004 0.041

Behav. control 0.481 0.378 0.319 0.338 0.219 0.254 0.424 0.476 0.512 0.351 0.424 0.360 0.624 0.427 0.769 0.408 0.841 0.896 0.679 0.835 0.067 0.075 0.096 0.042

Perc. Involv. 0.073 0.069 0.128 0.179 -0.062 -0.312 -0.187 -0.074 -0.025 0.030 0.072 0.054 0.003 0.048 0.054 -0.025 -0.080 0.054 0.215 0.111 0.906 0.887 0.777 0.676

E. Analysis for moderation Moderating variable Product Advantage Product Newness Subjective Norms Learning Orientation Performance Orientation Outcome Control Behavioral Control Perceived Involvement Product Newness Subjective Norms Learning Orientation Performance Orientation Behavioral control Perceived Involvement Product Advantage Subjective Norms Learning Orientation Performance orientation Behavioral control Perceived Involvement Product Advantage Product Newness Learning Orientation Performance Orientation Behavioral Control Perceived Involvement Product Advantage Product Newness Subjective Norms Performance Orientation Behavioral control Perceived Involvement Product Advantage Product Newness Subjective Norms Learning Orientation Behavioral control Perceived Involvement Product Advantage Product Newness Subjective Norms Learning Orientation Performance Orientation Perceived Involvement

Relationship moderated SFNP Adoption → Sales Performance SFNP Adoption → Sales Performance SFNP Adoption → Sales Performance SFNP Adoption → Sales Performance SFNP Adoption → Sales Performance SFNP Adoption → Sales Performance SFNP Adoption → Sales Performance SFNP Adoption → Sales Performance Product Advantage → SFNP Adoption Product Advantage → SFNP Adoption Product Advantage → SFNP Adoption Product Advantage → SFNP Adoption Product Advantage → SFNP Adoption Product Advantage → SFNP Adoption Product Newness → SFNP Adoption Product Newness → SFNP Adoption Product Newness → SFNP Adoption Product Newness → SFNP Adoption Product Newness → SFNP Adoption Product Newness → SFNP Adoption Subjective Norms → SFNP Adoption Subjective Norms → SFNP Adoption Subjective Norms → SFNP Adoption Subjective Norms → SFNP Adoption Subjective Norms → SFNP Adoption Subjective Norms → SFNP Adoption Learning Orientation → SFNP Adoption Learning Orientation → SFNP Adoption Learning Orientation → SFNP Adoption Learning Orientation → SFNP Adoption Learning Orientation → SFNP Adoption Learning Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Behavioral Control → SFNP Adoption Behavioral Control → SFNP Adoption Behavioral Control → SFNP Adoption Behavioral Control → SFNP Adoption Behavioral Control → SFNP Adoption Behavioral Control → SFNP Adoption

68

Coefficient -0,096 -0,062 -0,006 0,027 0,016 0,027 -0,006 0,004 0,022 0,033 0,127* 0,151 0,068 0,040 0,022 0,051 0,074 0,005 0,027 -0,112 0,033 0,051 0,002 0,052 0,059 -0,031 0,127* 0,074 0,002 -0,043 -0,087 0,098 0,151 0,005 0,052 -0,043 -0,081 0,029 0,068 0,027 0,059 -0,087 -0,081 0,086

t-value 1,522 0,814 0,057 0,285 0,163 0,288 0,076 0,047 0,223 0,335 1,694 1,615 0,687 0,387 0,222 0,551 0,819 0,051 0,326 1,036 0,326 0,571 0,025 0,845 0,654 0,372 1,694 0,811 0,024 0,766 1,294 1,112 1,618 0,055 0,803 0,782 1,287 0,338 0,680 0,324 0,678 1,310 1,250 1,087

Moderating variable Product Advantage Product Newness Subjective Norms Learning Orientation Performance Orientation Behavioral Control Age Age Age Age Age Age Age Age Age Age Age TenSec TenSec TenSec TenSec TenSec TenSec TenSec TenSec TenSec TenSec TenSec TenFunc TenFunc TenFunc TenFunc TenFunc TenFunc TenFunc TenFunc TenFunc TenFunc TenFunc TenAkz TenAkz TenAkz TenAkz TenAkz TenAkz TenAkz

Relationship moderated Perceived Involvement→ SFNP Adoption Perceived Involvement→ SFNP Adoption Perceived Involvement→ SFNP Adoption Perceived Involvement→ SFNP Adoption Perceived Involvement→ SFNP Adoption Perceived Involvement→ SFNP Adoption SFNP Adoption → Sales Performance Product Advantage → SFNP Adoption Product Newness → SFNP Adoption Subjective Norms → SFNP Adoption Learning Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Behavioral Control → SFNP Adoption Perceived Involvement → SFNP Adoption Product Newness → Product Advantage Subjective Norms → Product Advantage Subjective Norms → Product Newness SFNP Adoption → Sales Performance Product Advantage → SFNP Adoption Product Newness → SFNP Adoption Subjective Norms → SFNP Adoption Learning Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Behavioral Control → SFNP Adoption Perceived Involvement → SFNP Adoption Product Newness → Product Advantage Subjective Norms → Product Advantage Subjective Norms → Product Newness SFNP Adoption → Sales Performance Product Advantage → SFNP Adoption Product Newness → SFNP Adoption Subjective Norms → SFNP Adoption Learning Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Behavioral Control → SFNP Adoption Perceived Involvement → SFNP Adoption Product Newness → Product Advantage Subjective Norms → Product Advantage Subjective Norms → Product Newness SFNP Adoption → Sales Performance Product Advantage → SFNP Adoption Product Newness → SFNP Adoption Subjective Norms → SFNP Adoption Learning Orientation → SFNP Adoption Performance Orientation → SFNP Adoption Behavioral Control → SFNP Adoption 69

Coefficient 0,040 -0,112 -0,031 0,098 0,029 0,086 -0,154 -0,203** 0,008 -0,106 -0,051 0,052 0,045 0,100 -0,122 -0,014 -0,034 -0,079 -0,023 -0,022 0,006 -0,112 -0,042 0,007 0,064 0,064 -0,032 0,090 -0,075 -0,063 -0,061 -0,073 -0,002 -0,007 -0,011 0,058 -0,126 -0,003 0,139 -0,128 -0,011 -0,055 -0,074 -0,011 0,033 0,050

t-value 0,407 1,063 0,353 1,206 0,336 1,055 1,330 2,144 0,075 1,024 0,514 0,652 0,541 1,160 1,123 0,121 0,241 0,959 0,248 0,216 0,073 1,364 0,561 0,108 0,822 0,534 0,300 0,782 0,774 0,571 0,545 0,802 0,024 0,085 0,134 0,776 1,061 0,025 1,241 1,270 0,112 0,579 0,770 0,165 0,456 0,669

TenAkz TenAkz TenAkz TenAkz

0,026 Perceived Involvement → SFNP Adoption 0,352 -0,017 Product Newness → Product Advantage 0,105 0,008 Subjective Norms → Product Advantage 0,065 0,049 Subjective Norms → Product Newness 0,494 Note: significance levels ***p1.65)

F. Additional paths investigated Additional paths investigated Path modelled

Coefficient

tvalue

R

2

Increase

0,154

Original model Product Advantage

Sales Performance

0,179

1,481

0,174

0,020

Product Newness

 

Sales Performance

-0,419***

4,006

0,295

0,141

Subjective Norms



Sales Performance

0,098

0,802

0,160

0,006

Learning Orientation



Sales Performance

0,067

0,508

0,157

0,003

Performance Orientation



Sales Performance

-0,024

0,212

0,154

0,000

Behavioral Control



Sales Performance

-0,092

0,872

0,160

0,006

Perceived Involvement



Sales Performance

0,234***

2,583

0,208

0,054

Perceived Involvement



Product Newness

-0,285***

3,590

-

-

Learning Orientation



Perceived Involvement

-0,230*

1,789

-

-

Model adjusted based on these paths Model 1

Paths Modelled

Coefficient *** ***

SFNP Adoption



Sales Performance

0,392

Product Advantage



SFNP Adoption

0,360

Product Newness



SFNP Adoption

-0,141

Product Newness



Product Advantage

-0,318

Subjective Norms



***

***

Product Advantage

0,320

***

2

R= 0,154 tvalue

Model 2

2

R= 0,312

Coefficient

t-value

3,755

0,211**

1,987

2,928

0,381***

3,140

1,446

0,161*

1,762

3,355

-0,317***

3,463

3,627

0,310***

3,068

Subjective Norms



Product Newness

-0,373

3,840

-0,371***

4,287

Subjective Norms



SFNP Adoption

0,175

*

1,697

0,147*

1,722

Learing Orientation



SFNP Adoption

0,038

0,401

Performance Orientation



SFNP Adoption

0,104

1,024

0,120

1,543

0,229***

2,957

***

Deleted

Behavioral Control



SFNP Adoption

0,210

2,625

Perceived Involvement



SFNP Adoption

0,083

0,718

Product Newness



Sales Performance

-0,377***

3,517

Perceived Involvement



Sales Performance

0,138*

1,871

Perceived Involvement



Product Newness

-0,282***

3,642

Learning Orientation



Perceived Involvement

-0,286***

3,093

Deleted

Note: significance levels ***p1.65)

70

G. Early versus late response

71

H. Notifications 1. Pre-notification to customer-contact employee

Sassenheim, February 2015. Request to participate in a survey

Dear Sir, Madam, Would you like to do us a favor? Scientific research has shown that if a salesperson does not support a new product, has not ‘adopted’ the new product, customers will neither. Furthermore, those who are actually in close contact with the customer, the sales force, has a crucial role in the success of product introductions. Because the influence of customer-contact employees, we would therefore like to investigate what factors influence new product adoption. This research takes place as part of a graduation research conducted by a student of the Eindhoven University of Technology (the Netherlands) in collaboration with AkzoNobel’s SMU Vehicle Refinishes. The aim of the research is to better understanding the behavior of customer-contact employees before and during product introduction and thereby provide AkzoNobel with tools to be even more successful when introducing new products and services. Soon you will receive an invitation to participate in this research. When you participate, you have the option to receive a personalized benchmark in which your situation and those of colleagues is compared. Since confidentiality and anonymity are of paramount importance to us, your response will be strictly anonymous. When you receive the invitation, I hope you will spend just 10 minutes of your precious time to help us with this research. Furthermore, we hope that you take this opportunity to share with us your thoughts and opinions on the strategically important topic of product introductions. If you have any questions, you can contact Thomas Janssens or Jacqueline Revet. Thank you very much for your cooperation, it is highly appreciated. Kind regards,

72

2. Pre-notification to sales manager

Sassenheim, February 2015. Request to participate in a survey Dear Sir, Madam, I am currently working in-house at Sassenheim in The Netherlands, and my direct Supervisor is Jacqueline Revert, in the Color Marketing group of Ben Zweers. I am a student from the Eindhoven University of Technology and in the context of my master Innovation Management I am investigating the adoption of new products within organizations. Because of my technical background I choose to conduct this research within a high-tech company where technology is key: AkzoNobel. Scientific research has shown that if a salesperson adopts a new product, they are better able to sell it and their sales performance will increase. In line with these findings I want to investigate what factors influence product adoption by those persons actually visiting AkzoNobel’s customers: technical consultants and account managers (sales). The aim of the research is to create a quantitative understanding of the factors that lead to product adoption. I hope to provide AkzoNobel with tools to improve product adoption and therefore be even more successful when introducing new and improved products or services. In order to draw conclusions that are statistically relevant, I need the help of you and your team. As agreed with Remco Maassen van den Brink (VR Marketing Director), soon I would like to distribute a digital survey amongst a large number of technical consultants and account managers. As the findings of this research are valuable for you and your team as well, I would like to ask you to raise awareness amongst your employees about this survey and request them to participate in the survey. When I have distributed the survey, you will receive a notification again via email. The survey contains questions about the current level of adoption, perceived sales performance and factors affecting product adoption. Completing it takes only 10 minutes of their time. Furthermore, the answers will be kept confidential and only aggregated results will be published, results which I am happy to share with you later. Attached, you will find a first version of the research: a theoretical foundation of factors I would like to investigate by means of this survey. If you have any questions, you can contact Thomas Janssens or Jacqueline Revet. Thank you very much for your cooperation, it is highly appreciated. Kind regards,

73

3. Invitation to customer-contact employee

Sassenheim, February 2015. Request to participate in a survey Dear Sir, Madam, Would you like to do us a favor? Scientific research has recently shown that if a salesperson is not enthusiastic about a new product, has not ‘adopted’ the new product, customers will not adopt it either. We would therefore like to investigate what factors influence new product adoption by customer-contact employees. The best way to conduct such a research is by making use of a survey. This research takes place as part of a graduation research conducted by a student of the Eindhoven University of Technology (the Netherlands) in collaboration with AkzoNobel’s SMU Vehicle Refinishes. The aim of the research is to better understanding the behavior of customer-contact employees before and during product introduction and thereby provide AkzoNobel with tools to be even more successful when introducing new products and services. We would like to invite you to answer a few questions. These questions will be about your opinion and experiences concerning the selling of Akzonobel’s new products. Answering these questions will take only 10 minutes of your time and will provide us and yourself with valuable information. After completion namely, you are given the opportunity to retrieve a personal benchmark in which your selling experience is compared with those of your colleagues (anonymity is guaranteed). We hope you will spend 10 minutes of your precious time to help us with this research. Furthermore, we hope that you take this opportunity to share with us your thoughts and opinions on the strategically important topic of product introductions. We would kindly ask you to complete the survey, that is available via https://nl.surveymonkey.com/s/Akzo-English , before February 22nd. If you receive an error message when following the link, simply click on the link again, this should ‘solve’ the problem and enable you to start the survey. If you have any questions, you can contact me or Jacqueline Revet. Thank you very much for your cooperation, it is highly appreciated. Kind regards,

74

4. Notification invitation to sales manager

Dear Sir, Madam, On the 12th of February, I sent you an email in which I tried to explain the aim and relevance of the research I’m conducting as well as why a high response rate is desirable. Today, I invited all North-American account managers and technical consultants to participate in this research by completing a survey I designed. Since completion takes only 10 minutes of their time, I would like to ask you to raise awareness amongst your employees about this survey and request them to participate in the survey. Thank you very much for your cooperation, it is highly appreciated. If you have any questions, you can contact me or Jacqueline Revet. Kind regards,

5. First reminder to customer-contact employee

Reminder: survey product adoption Dear, Recently, we invited you to participate in a research about new product adoption by customer contact employees. Your participation is important since it helps to improve product introductions of AkzoNobel. The survey is written in English and completing it takes only 10 minutes of your time. It is possible to stop at any time and finish the survey later. We will take care of your personal data and your anonymity is guaranteed. You can start directly by clicking on the following link: https://nl.surveymonkey.com/s/Akzo-English Unfortunately, problems may arise when trying to participate. If this occurs, please try one of the following 3 things: 1. Click on the link again 2. Many of you will be connected via Junos Pulse. Apparently, when you disconnect from Junos Pulse and then follow the link to the survey, you should be able to participate. 3. A last solution is related to settings concerning ‘javascript or cookies’. I’d like to refer you to the following site if none of the above seems to solve the problem: http://help.surveymonkey.com/articles/en_US/kb/We-are-having-accessibility-ortechnical-issues-What-should-we-do I’d like to thank you for participating! Kind regards, 75

6. First reminder to sales manager

Reminder: survey product adoption Dear, Recently, we informed you about a research concerning new product adoption. To date, the response rate to the questionnaire is pretty good. Unfortunately however, it is still too low to be able to make some well-grounded analysis. Given that this research will help us improve product introductions within AkzoNobel, I would like to ask you to raise awareness amongst your employees about this survey and request them to participate in it. Completing it only takes 10 minutes of their time and provides us with valuable information. Thank you very much for your cooperation, it is highly appreciated. Kind regards, 7. Second reminder to customer-contact employee

Reminder: Please complete the survey on product adoption Dear, I would like to remind you about a graduation research about new product adoption by customer contact employees currently being conducted within AkzoNobel. To date, the response rate to the questionnaire is pretty good. Unfortunately however, it is still too low to be able to make some well-grounded analysis. Please help me in reaching the right number of responses. In the beginning of March, I’d like to finish the data collection and we want to be sure that everybody who has not yet responded has had the opportunity to participate. After this, I can start with the creation of the benchmark. Completing the survey takes only 10 minutes of your time. I will take care of your personal data and your anonymity is guaranteed. You can start directly by clicking on the following link: https://nl.surveymonkey.com/s/Akzo-English Unfortunately, problems may arise when trying to participate. If this occurs, please try one of the following 3 things: 1. Click on the link again 2. Many of you will be connected via Junos Pulse. Apparently, when you disconnect from Junos Pulse and then follow the link to the survey, you should be able to participate.

76

3. A last solution is related to settings concerning ‘javascript or cookies’. I’d like to refer you to the following site if none of the above seems to solve the problem: http://help.surveymonkey.com/articles/en_US/kb/We-are-having-accessibility-ortechnical-issues-What-should-we-do I’d like to thank you for participating! Kind regards, 8. Second reminder to sales manager

Reminder: last request to emphasize questionnaire product adoption Dear, Recently, we informed you about a research concerning new product adoption. In the beginning of March, we’d like to finish the data collection and we want to be sure that everybody who has not yet responded has had the opportunity to participate. I would like to ask you one last time to raise awareness amongst your employees about this survey and request them to participate in it (for example by sending an email). Completing it only takes 10 minutes of their time and provides us with valuable information. Thank you very much for your cooperation, it is highly appreciated. Kind regards,

77

I. List of antecedents mentioned in literature Category

Antecedent

Perceived Usefulness (PU)

Product

Performance expectancy (PE)

Perceived Ease of use (PEoU)

Effort Expectancy (EE)

Discussed by (+ model / theory) Davis (1985), (TAM) Venkatesh & Davis (2000), (TAM2) Lu, Yao, & Yu (2005) Schillewaert, Ahearne, Frambach, & Moenaert (2005) Venkatesh et al. (2003), (UTAUT) Venkatesh et al. (2003), (UTAUT) Davis (1985), (TAM) Venkatesh & Davis (2000), (TAM2) Lu et al. (2005) Schillewaert et al. (2005) Lu et al. (2005) Davis (1985), (TAM) Schepers & Wetzels (2007) Venkatesh et al. (2003), (UTAUT) Venkatesh et al. (2003), (UTAUT)

Relationship

Effect size

PU  Intention to Use

0.79

PU  Intention to Use

0.55

PU  Intention to Adopt

0.81

PU  Adoption

0.37

PE  Behavioral Intention

0.40 – 0.43

PE  Attitude towards use

0.28 – 0.32

PEoU  Intention to Use

.18

PEoU  Intention to Use

0.17

PEoU  Intention to Adopt PEoU  Adoption PEoU  PU PEoU  PU PEoU  PU

0.24 0.14 0.22 0.30 0.48

EE  PE

0.30 – 0.34

EE  Behavioral Intention

0.22

78

Comment

Increases over time

Decreases over time

Category

Antecedent

Social

Subjective norms (SN)

Social Influences (SI)

Peer Usage

Discussed by (+ model / theory) Venkatesh & Davis (2000), (TAM2) Venkatesh & Davis (2000), (TAM2) Fu et al. (2010) Schepers & Wetzels (2007) Schepers & Wetzels (2007) Lu et al. (2005)

Relationship

Effect size

Comment

SN  PU

0.20 – 0.50

Depends on voluntariness

SN  Intention to use SN  Selling intention SN  PU SN  Behavioral Intention SI  PU

0.24 – 0.31 0.31 0.31 0.16 0.22

Depends on voluntariness

Lu et al. (2005)

SI  PEoU

0.11

Venkatesh et al. (2003), (UTAUT) Venkatesh et al. (2003), (UTAUT) Venkatesh et al. (2003), (UTAUT) Schillewaert et al. (2005) Schillewaert et al. (2005) Schillewaert et al. (2005)

SI  PE

0.29 – 0.33

SI  EE

-0.21

SI  Behavioral intentions

0.19 – 0.31

Peer Usage  PU Peer Usage  PEoU Peer Usage  adoption

0.15 0.28 0.19

79

Category

Antecedent

(Characteristics of) Salesperson

Role conflict (RC)

Role ambiguity (RA)

Goal orientation

Personal innovativeness (PI)

Perceived personal competence (PPC) Self-efficacy Job Satisfaction

Discussed by (+ model / theory) Siguaw, Brown, & Widing (1994) MacKenzie, Podsakoff, & Ahearne (1998) Zablah, Franke, Brown, & Bartholomew (2012) Brown & Peterson (1994) Johnston, Parasuraman, Futrell, & Black (1990)

Relationship RC  Organizational Commitment RC Organizational Commitment Role conflict  Organizational Commitment RA  Effort

Effect size -0.17 -0.24

RA  Commitment

-0.22

Siguaw et al. (1994)

RA  organizational commitment

-0.14

Zablah et al. (2012) MacKenzie et al. (1998) Sujan et al. (1994) Sujan et al. (1994) Sujan et al. (1994) VandeWalle, Brown, Cron, & Slocum Jr (1999) Lu et al. (2005) Lu et al. (2005) Schillewaert et al. (2005) Schillewaert et al. (2005) Jones et al. (2002) Robinson Jr, Marshall, & Stamps (2005) Mathieu & Zajac (1990)

RA  organizational commitment RA  Commitment Learning orientation Effort Learning orientation Effort Performance orientation  Effort Learning orientation  Effort PI  PU PI PEoU PI Adoption PI  PEoU PI  Use to the fullest extent PI  PEoU PPC  Organizational Commitment

-0.10 -0.51 0.38 0.58 0.33 0.21 0.23 0.33 0.24 0.22 0.42 0.81 Correlation of .63

Fu et al. (2010) Fu et al. (2009) Krishnan et al. (2002) Bateman & Strasser (1984)

Self-efficacy  Selling intention Self-efficacy  Effort Self-efficacy  Effort JS  Organizational commitment

0.3-0.36 0.2-0.25 0.28-0.43

80

Comment

-0.11 -0.16

Bateman &

(JS)

Process / Context

Category

MacKenzie et al. (1998) Testa (2001) Zablah et al. (2012) Johnston et al. (1990)

Antecedent User Training

Discussed by (+ model / theory) Schillewaert et al. (2005) Schillewaert et al. (2005) Ahearne et al. (2010) Hultink et al. (2000)

JS  Organizational commitment JS  Organizational commitment JS  Organizational commitment JS  Organizational commitment

Relationship User training  Adoption User training  PEoU Behavior CS moderates negative effect of product perception on effort Outcome CS  selling performance

0.37 0.84 0.49 0.36

Effect size 0.13 0.28

Hultink & Atuahene-Gima (2000)

Behavior CS  selling performance Outcome CS moderates SFNP adoption on selling performance

81

Comment

0.21 0.26

Control system (CS) Hultink et al. (2000)

Strasser (1984) state that satisfaction can also be viewed as an outcome of, for example, performance and effort. This is in line with e.g. Brown & Peterson (1994); Christen, Iyer, & Soberman (2006).

-0.34 0.15

Contrary to expectations