THE DARK SIDE OF MARKETING ACCOUNTABILITY:

(1)

THE DARK SIDE OF MARKETING

ACCOUNTABILITY:

Expanding conceptual knowledge and determining

linkages to marketing activities and firm performance

Arjen Onrust

Marketing Department

SOM Research Master

First supervisor

prof. dr. P.C. (Peter) Verhoef

Second supervisor

prof. dr. K. (Koert) van Ittersum

(2)

Acknowledgements

First and foremost, I owe a debt of gratitude to my first supervisor, professor Peter Verhoef, for guiding me through the process of shaping and developing this research. By providing me with insights, a critical eye when needed, and also giving me the freedom to pursue my own ideas, it helped me to create a thesis that strongly feels as a product of my own doing but at the same time realizes a level of quality that I could not have achieved on my own. That is why I consider his input as a supervisor and researcher invaluable to this thesis. In the same breath I thank professor Koert van Ittersum for his enthusiasm in accepting the request to be my second supervisor and for investing the time to read and provide fresh insights on what has become a document of substantial volume. Also I would like to express my appreciation for the people that directly helped in realizing this research. I give thanks to the expert panel group for their keen professional eye on the quality and validity of the questionnaire that is used for this research, and a special thanks to Marjolijn Onrust and Vincent Kunst who took the time out of their own schedules to translate the questionnaire from Dutch to English and back again to Dutch. Also, I appreciate the help from professor Jaap Wieringa and professor Peter Verhoef for allowing me to use the participants of their marketing workshops for pretesting the survey.

I feel immense gratitude towards my family and friends that have supported me throughout the development and writing of this thesis. I am thankful for their patience with me when I seemed distracted with the assignment, or when my communications where absorbed with things about my research. I especially thank my parents for staying worried about me (as parents can do so well), my sister who supported me and helped me to put things into perspective, my girlfriend who was always able to lighten my mood and comfort me when I needed it, and lastly my best friend who always had my back even though we didn’t have as much time to keep contact as we wanted to.

(3)

Abstract

Marketing accountability has gained increased prominence in marketing literature as pressures increase from firms on marketing functions to express their contributions to the value of the firm. The overzealous focus on how increased marketing accountability would aid the financial justification of marketing expenditures, has resulted in severe lack of attention to 1) the potential dark side of marketing accountability, and 2) complementary forms of marketing accountability that affect marketing performance.

This study expands the conceptual understanding of accountability in marketing by adopting an interdisciplinary focus. The findings are accommodated in the conceptual framework. Subsequently, an empirical assessment is made of the marketing accountability dimensions (i.e., financial outcome accountability and procedural accountability) on the performance of explorative (e.g., long-term, innovative, and risk taking oriented) and exploitative (e.g., short-term, incremental, and refinement oriented) marketing activities, and of the performance of marketing activities on firm performance (i.e., market performance and innovation performance). We do so by collecting data from 59 of mostly marketing managers in large Dutch firms via an e-mail survey, and conducting manual stepwise multiple regression analyses.

Surprisingly, the data shows that the dark side of marketing accountability is less pronounced than anticipated. Overall, marketing accountability demonstrates positive effects on the performance of marketing activities, which means that a more capable and well-functioning marketing function (on accountability) performs better. Yet, it is found that firms’ increased emphasis on evaluating marketing activities with financial metrics instead of marketing metrics will have a detrimental effect on the performance of marketing activities. In addition, performance on explorative and exploitative marketing activities fully mediate the positive effect of procedural accountability on innovation performance, which indicates that a capable marketing function also increases the performance of the firm.

(4)

discussion on the potential extraneous effects of marketing accountability, as well as on what it means to be accountable.

(5)

Chapter 1 Introduction

Many functions of the firm are pressured by corporate executives to justify expenditure of budget (Luo, Slotegraaf, and Pan 2006b), and have to compete with other functions in order to appeal to future resources (Maltz and Kohli 1996; Gupta and Govindarajan 2000; Birkinshaw and Lingblad 2005). This is even more pertinent in situation of corporate budget constrains when business is stagnant, or in times of low economy as we currently experience (Lehmann 2004). It requires managers to provide evidence of their contribution and to demonstrate their value to the firm, hence proving they are a worthy part of the organization deserving of financial resources. A multitude of methods and metrics are created to monitor performance and evaluate these contributions (Simons, Dávila, and Kaplan 2000; Ittner and Larcker 1998; Eccles 1991). Likewise, the marketing function has come under considerable scrutiny for their inability to prove that its actions add value to the firm, which resulted in the loss of overall credibility and presence within the boardroom (Rust, Ambler, Carpenter, Kumar, and Srivastava 2004b; Moorman and Rust 1999; Verhoef and Leeflang 2009; O'Sullivan and Abela 2007). Aiming to find the solution to how marketing should improve their position has given rise to marketing accountability research. For this purpose, marketing accountability metrics and frameworks are created to show how marketing expenditures contribute to firm value.

(8)

8

can affect the firm in multiple ways that are currently left unaddressed by marketing research. How will increased accountable affect, for example, creativity and innovativeness, risk taking and permitting failures, efficiency of processes, and so forth? Does marketing accountability interact positively with these behavioral factors, or can it hurt them? We simply do not know, since they have yet to be empirically tested in the marketing setting (Gupta, Smith, and Shalley 2006). Only recently has a first study surfaced, by Mintz and Currim (2013), which addresses such effects of marketing accountability. The two scholars assess the effect of financial and marketing metrics on marketing-mix activities. One of their main findings is that the use of financial and marketing metrics is driven by the setting in which managers operate such as firm strategy, and firm and environmental characteristics. More importantly they find that the use of metrics is positively related to the perceived performance of marketing-mix elements, which provides new insights in how formal evaluation measures and metrics can affect the performance of marketing activities in itself. Still, this is only a fraction of what we need to know if we can convincingly implement accountability in marketing. A more comprehensive understanding of accountability effects on different marketing activities is needed.

(9)

9

and MacInnis 1989; Tetlock 1983; Lerner and Tetlock 1999). This high degree of stress hampers a function’s performance, and as a result could decrease its perceived value by the firm (Baumeister 1984; Ashton 1990). Such matters require additional attention to determine its impact on the marketing function within the firm.

This paper contributes to current marketing literature by opening the black box and increasing the knowledge on the diverse effects of accountability on the marketing function. It is a first attempt to expose the dark side of marketing accountability by addressing detrimental consequences of its presence within the firm and looking beyond solely the positives. An empirically analysis is conducted to assess the effects of marketing accountability on the marketing function’s performance of two distinct forms of activities (i.e. marketing exploration and marketing exploitation) leading to firm performance. Whereas extant literature is focused on improving marketing’s accountability by determining the link between marketing function’s activities and financial firm value, we aim to determine the effects of more accountable marketing functions on their ability to perform. Filling this knowledge gap may prevent unexpected complications and failures to implement accountable marketing practices.

(10)

10

(11)

11

Chapter 2 Literature background

2.1 MARKETING ACCOUNTABILITY (MA) LITERATURE

OVERVIEW AND DISCUSSION

Marketing Accountability (MA) receives substantial attention in current marketing literature due to the challenges that marketing practice is facing. Top executives call for marketing functions to demonstrate the contributions of their activities (Kumar and Shah 2009; Rust, Ambler, Carpenter, Kumar, and Srivastava 2004), while marketing’s inability to do so has led to a decline of their perceived importance within the firm. As a result, marketing is losing its standing in the firm and perceives a diminishing presence in the strategic planning process (Rust, Lemon, and Zeithaml 2004; Webster, Malter, and Ganesan 2004; Schultz 2003; O’Sullivan and Abela 2007). MA research is considered pivotal to remedying marketing’s decline, and is determined to help regain the strategic position of the marketing function, increase perceived influence, and improve firm performance (Schultz and Gronstedt 1997; O’Sullivan and Abela 2007; Verhoef and Leeflang 2009).

(12)

12

illustrate that studies can have different perspectives on how and what to research. Hence, in order to situate our study within the MA topic we present an overview of the main papers and define which different literature streams exist.

2.1.1 Literature overview and research streams

Since MA was first introduced in marketing research, a substantial and still growing body of literature has been created. The studies in it can be classified in two distinct streams of research: 1) linking marketing outcomes to financial firm performance and 2) relating marketing accountability to organizational behavior and perceptions. Table 1 provides an overview of noteworthy and impactful research that is conducted on marketing accountability.

The first stream pertains to research that provides evidence of linkages between marketing intermediate outcomes and activities to firm value, which tackles the primary concern from practice about marketing. This stream can be split by different dependent and independent foci used in the studies. The independent variable focus consists mostly of papers on customer metrics and measures, followed by brand metrics, and a few on specific marketing activities. The dependent variable of choice in this stream mainly revolves around the stock market and shareholder value. Several scholars address multiple financial firm metrics within the same paper and are therefore mentioned more than once in the table (e.g., Krasnikov, Mishra, and Orozco 2009; Kumar and Shah 2009). For a comprehensive review on methods and main findings of linking marketing to firm value, see Srinivasan and Hanssens (2009).

(13)

13

We situate this research towards the second literature stream, since we treat MA as an ability of the marketing function instead as marketing-finance measures and metrics. In addition, we are interested in the effects of accountability on marketing and the organization. Most notably, we focus on effects of MA on marketing intermediate performance (as well as firm/market performances) which has mostly been unaddressed in the literature. Many studies in research stream 2 are related to the problem at hand with marketing, and disregard all other context in which MA would influence marketing and the firm. Only now are studies emerging that look at how accountability of the marketing function affects multiple facets of the organization and marketing itself beyond the perceived influence on and perceptions on the firm that have plagued marketing practice (Currim and Mintz 2013).

Table 1. Overview of marketing accountability research

Main research

stream characteristics Stream variable focus Independent Studies variable focus Dependent Studies

Linking marketing outcomes to financial firm performance Focused on the development of frameworks and metrics to demonstrate marketing’s financial contributions. Studies explain how marketing activities or marketing intermediate outcomes affect firm value. Brand measures and metric

Mizik and Jacobson (2009); Krasnikov et al. (2009); Mizik and Jacobson (2008);

Srinivasan and Hanssens (2009; Luo et al. (2013); Madden et al. (2006) Cash flows / shareholder value Anderson et al. (2004); Krasnikov et al. (2009); Rust et al. (2004b); Rao and

Bharadwaj (2008); Srinivasan and Hanssens (2009) ; Gruca and Rego (2005); Kumar and Shah (2009) ; Srivastava et al. (1998); Gupta et al. (2004);

Morgan and Rego (2006); Madden et al. (2006) Customer

metrics and measures

Anderson et al. (2004); Aksoy et al. (2008); Gupta and Zeithaml (2006); Rust et al. (2004a); Srinivasan and Hanssens (2009); Mittal

et al. (2005); Luo and Bhattacharya (2006a); Fornell et al. (2006); Gruca and Rego (2005); Dotzel et al. (2013); Kumar and Shah (2009); Tuli and Bharadwaj (2009); Anderson et al. (1994); Gupta et al. (2004);

Morgan and Rego (2006)

Stock market valuation / returns and risk Aksoy et al. (2008); Krasnikov et al.(2009); Mizik and Jacobson (2008); O'Sullivan and Abela (2007); McAlister etl al. (2007); Luo and Bhattacharya (2006a); Fornell et al. (2006); Dotzel et al. (2013); Kumar and

Shah (2009); Tuli and Bharadwaj (2009) Specific marketing activities Vorhies et al. (2011); McAlister et al.(2007); Srivastava et al. (1998) Relating marketing accountability to organizational behavior and perceptions Research aims to find how marketing activities and MA affect the perceptions on marketing and behaviors within the firm. MA as an ability or capability

Moorman and Rust (1999), O'Sullivan and Abela (2007);

Lukas et al. (2005); Verhoef and Leeflang (2009); Engelen and Brettel (2011)

Perception’s and influence within the firm

Homburg et al. (1999); O'Sullivan and Abela (2007);

(14)

14

Even though we were able to find communalities between MA studies and categorize them in different research streams, a clear and comprehensive definition of MA is still lacking. To what it does and does not pertain is unclear, while a transparent or explicit debate on this has not yet manifested. Conceptual discussions and consensus are necessary for the progression of the research topic and continuation of knowledge creation (Yadav 2010; Webster and Watson 2002). One of the goals of this paper is to facilitate the start of such a discussion and to expand current conceptual knowledge of marketing accountability. We will do so below by 1) discussing the gaps in the current conceptual focus of MA, and 2) bridging the knowledge gap on accountability using closely related research fields to extend the MA concept (Whetten 1989).

2.1.2 Research gaps

We discuss two major gaps in the literature regarding marketing accountability that potentially impede the development of the research topic. We do so by asking ourselves two important questions. Is a focus on marketing’s financial justification enough when we address the marketing accountability concept, or is there more to be considered that may fall under the umbrella of MA? Is what we know now sufficient to recommend marketing accountability practices to marketing managers, or do we need to know more about its organizational implications? These are two questions that have been left unaddressed by extant marketing literature, and therefore generate gaps in the literature that need to be addressed if successful application of accountability in practice is to be achieved. We will now discuss both in more detail below, and why we believe they need to be subject of discussion.

(15)

15

dictionary (2005) describes accountable as being “responsible for your decisions or actions and [are] expected to explain them when you are asked.” Hence, accountability is more than justifying outcomes, and more than accommodating and incorporating financial information. In order to illustrate the limitations of only having a financial take on accountability and consequently an outcome focus, we consulted the affiliate research fields of management (control), psychology, and organizational behavior. Based on the work of Philip E. Tetlock in psychology on accountability, we describe three different accountability angles as depicted in figure 1, that have not yet been considered in marketing research (Lerner and Tetlock 1999; Brown 1999; O’Sullivan and Abela 2007). The ability to be accountable is a behavioral state of a person or group, to be held accountable is a behavioral state of an external person or group, and to feel accountable is the intrinsic affective state of a person or group. It is reasonable to assume that the three perspectives are related to each other to some extent. Marketing literature has primarily focused on the ability of marketing practice to be accountable, and has disregarded all other effects from what it means to feel the pressure of accountability, and what happens when managers and members in a firm are actually held accountable for their work.

(16)

16

Figure 1. Accountability perspective triangle

The second gap concerns whether we need to emphasize our academic attention more towards finding what implications marketing accountability will have on the behavior of marketing and the firm. And the answer is, yes we do. Concerns are expressed from both practice and academics on the potential detrimental effects that increased marketing accountability may have for marketing practitioners and the performance of the firm. Some scholars have cautioned the implementation of marketing accountability practice as it is suggested that it would hamper creativity and innovativeness of the marketing function potentially leading to detrimental performance. Concurrently, we hear practitioners voicing their concerns on the uncertainty surrounding the effectiveness of accountability and what additional problems its presence may present. During calls with marketing managers at the data collection stage, this discomfort and uncertainty regarding MA was also expressed. The concerns that surfaced illustrate the shortcomings from research to properly address these topics, and need to be replaced with knowledge and insights in order to improve MA’s application in practice. Therefore, an empirical assessment of the dark side of marketing accountability needs to be address if we want to progress the accountability topic.

(17)

17

managers not always emphasize objective, accurate information, but also rely on information that is socially acceptable to the evaluator of the accountable managers. Moreover, accountability will increase management’s confidence in results of their tasks, but does not per se lead to higher accuracy or effectiveness of those tasks. Therefore, we need to challenge the underlying assumption discussed by O’ Sullivan and Abela (2007) that an increasing focus and implementation of accountability processes in the firm is, by definition, beneficial to the firm.

2.2 EXTENDING THE MA CONCEPT

A broadening of the marketing accountability concept is achieved by accommodating knowledge from multiple closely related fields to marketing. Outside marketing research accountability is addressed as management’s responsibility for activities and their justification to the managers’ constituents (i.e., those persons or groups managers feel accountable to (Tetlock 1991; Brown 1999) by whom their activities are monitored and appraised (Siegel-Jacobs and Yates 1996). The accountability concept from an organizational perspective can broadly be defined as the ability of holding individuals or groups within the organization responsible for their activities and behaviors, and are evaluated leading to corresponding positive or negative consequences (Mulgan 2000). The felt accountability, by those that are held accountable, is considered the mental state that can also affect behavior (Tetlock 1983).

Marketing accountability specifically pertains to accountability in the marketing function of the firm. Considering the broader definition and the previous discussion in the first research gap, we distinguish between two subtypes of marketing accountability: financial outcome accountability and procedural accountability (Simonson and Staw 1992; Tetlock, Vieider, Patil, and Grant 2013; Langhe, van Osselaer, and Wierenga 2011). Each subtype addresses a distinctive orientation on accountability for the marketing function (Turner and Makhija 2006). This differentiation improves the MA concept to a multidimensional construct consisting of dimensions with unique characteristics and effects on the organization, as is supported in the accountability review paper by Lerner and Tetlock (1999).

2.2.1 Financial outcome accountability

(18)

18

metrics; however these studies have never before been characterized together in a single dimension. Therefore, no clear definition is readily available for what we term as financial

outcome accountability. Based on studies that concern the aforementioned we define

financial outcome accountability here as marketing’s ability to determine and demonstrate

the contributions of its activities to financial firm performance aided by formal financial evaluation tools (Brtek and Motowidlo 2002; Verhoef and Leeflang 2009; Moorman and

Rust 1999; O'Sullivan and Abela 2007). Improving the ability to be financial outcome accountable enables the marketing function to justify its decisions and expenditures, and allows for appraisal of marketing’s contribution to firm value by the firm (Anderson, Fornell, and Lehmann 1994; Gupta and Zeithaml 2006; Kumar and Shah 2009).

The formal financial evaluation tools consist of financial performance metrics and frameworks that are continually being developed to better assess marketing’s financial contribution to bottom-line firm performance in objectively quantified terms. Such metrics and frameworks are distinctively different from conventional marketing performance measures, since the latter looks at intermediate performance whereas accountability metrics aim to link intermediate marketing performance to financial firm performance (Mintz and Currim 2013; Anderson, Fornell, and Lehmann 1994; Anderson, Fornell, and Mazvancheryl 2004; Lehmann 2004).

(19)

19

2.2.2 Procedural accountability

We consider procedural accountability as a necessary addition to support and warrant activities of less standardized marketing activities (Brown 1999; Siegel-Jacobs and Yates 1996; Brtek and Motowidlo 2002; Doney and Armstrong 1996).

This dimension has largely been neglected by prior marketing accountability research. Its current novel application in marketing is adapted from its well-founded base in psychology, management (control) and organizational behavior literature. We define procedural accountability in marketing as the ability to justify and evaluate the process

quality of individuals or groups within the organization, regardless of their outcomes

(Langhe, van Osselaer, and Wierenga 2011; Siegel-Jacobs and Yates 1996; Brtek and Motowidlo 2002). This extension of the MA concept addresses when, how and why activities are executed in a certain way (Siegel-Jacobs and Yates 1996). Though intensifying the control on processes and specific activities seems more controlling and constraining relative to the financial outcome dimension, it should be clear that the nature of the control is different between the two (Siegel-Jacobs and Yates 1996; Brtek and Motowidlo 2002; Doney and Armstrong 1996; Simonson and Staw 1992).

(20)

20

Chapter 3 Conceptual model and

hypotheses

Figure 2. Conceptual model

3.1 MA AND PERFORMANCE OF MARKETING ACTIVITIES

(21)

21

that the main challenge of marketing is improving their ability to account for marketing activities. As such marketing has focused on how to improve this ability; however we look at the extent to which this ability may affect the marketing function’s ability to perform activities. In turn, we will later on determine the influence of this perceived ability to perform these activities on the performance of the firm (see figure 2). This view is consistent with Verhoef and Leeflang (2009) who similarly address the ability or inability of the marketing department to be accountable. Furthermore, we approach and measure all constructs in our conceptual framework as perceptions of behaviors and attitudes. Hence, we proceed from this point onwards to consider the constructs as perceptual without explicit mentioning it at every instance.

3.1.1 Performance of marketing activities

We determine the extraneous effects of both marketing accountability dimensions (i.e., financial outcome accountability and procedural accountability) by analyzing the concept in relation to the perceived ability of the marketing function to perform marketing activities. These activities pertain to the core responsibilities of the marketing function to the firm, namely developing and executing ideas to create customer value which in turn generates value for the firm (Woodruff 1997; Parasuraman 1997).

We categorize marketing activities on two distinct dimensions: exploration and exploitation (March 1991). These two terms are generally used in product innovation situations where exploiting existing products enable incremental innovations and exploring for new opportunities allows for radical innovations (Andriopoulos and Lewis 2009). Others have explained the distinction in terms of competences (Atuahene-Gima 2005), activities (Benner and Tushman 2003), or marketing strategic types to respectfully improve current expertise and develop new knowledge and skills (Kyriakopoulos and Moorman 2004). The seminal paper of March (1991) refers to the two terms in relation to organizational learning. He poses that exploitation is typically concerned with short-term development, whereas exploration has more long-term benefits. We summarize the main distinctive characteristics of both types in table 2 below.

(22)

22

activities presented in table 2. For example, strategic marketing planning and customer knowledge management are reasonably more explorative whereas marketing mix implementation can arguably be considered as exploitative.

Even though exploration and exploitation seem to be opposite ends of a continuum, extant literature has determined that they are two separate dimensions that can be considered as orthogonally separate. Hence, we pose that the marketing function can conduct both explorative and exploitative activities and can perform high or low on either dimension simultaneously (Gupta, Smith, and Shalley 2006).

We have strong reasons to believe that the marketing function is affected by the accountability pressures to justify expenditures and activities in financial terms (i.e., financial outcome accountability). Increased financial outcome accountability by the marketing function implies higher intensity on financial justification and more extensive use of accompanied financial measures and metrics. The better able managers are to express contribution and justification, the more elaborate and sophisticated the metrics that are likely to be used (Ambler and Roberts 2006). Consequently, this would increase the risks and/or benefits of accountability to execute other core marketing responsibilities. The need from practice to develop new financial metrics has resulted in avoidance by researchers to assess implications of increased emphasis on financially controlling and evaluating the marketing function in the firm (Turner and Makhija 2006). We argue that imposing metrics and systems on marketing management also means the inclusion of constraining effects on the ability of the marketing function to perform activities. Depending on the different types of marketing activities that are executed, these effects can vary (Brown 1999; Christensen, Kaufman, and Shih 2008).

Prior research in management provides insights on procedural and outcome accountability and their relationship to manager’s performance and decision making. A

(23)

23

comparison between the effects of both types of accountability, in general, produces positive associations of procedural accountability and detrimental associations of outcome accountability (Brtek and Motowidlo 2002; Doney and Armstrong 1996). We conclude that in most mentioned studies, that look at or control for accountability, participants were either unfamiliar to the task at hand, or the complexity of the task is perceived as high. In such cases, the best way to act for participants remains ambiguous which leads them to resort to decision making that is favorable to those they are held accountable to. Additionally, the unfamiliarity or complexity increases uncertainty of performing well and will subsequently increase stress levels of failing to perform (Brown 1999; Lerner and Tetlock 1999). Explorative marketing activities are creative, emphasize the search and development of new ideas, and promote experimentation (see table 2). These characteristics would require accountability practices to focus more towards assessments of the process in order to scrutinize and evaluate performance of these activities (Simonson and Staw 1992; Turner and Makhija 2006). Additionally, Christensen, Kaufman, and Shih (2008) provide a convincing argument for why several financial tools can actually “kill” innovation and the capacity to do new things. This argument is strengthened by the concept of ‘choking under pressure’ meaning that performance diminishes in situation where good or improved performance has increasing importance (Baumeister 1984). Also several studies show that “accounting forms of control are poorly suited to highly uncertain tasks (i.e. lower task analyzability and higher degree of variety in task encounters; Abernethy and Brownell 1997).

H1a: The ability of the marketing functions to be financial outcome accountable is negatively related to their ability to perform explorative activities.

(24)

24

and Currim (2013) that financial metrics improve marketing-mix activities. Marketing-mix activities can arguably be considered to be more tactical, short-term oriented, and as more routinized with incremental changes as opposed to creative risk-taking endeavors that have more task uncertainty (strategic planning, or developing novel products and services. In their paper, all marketing-mix activities can be considered more or less exploitative apart from the marketing-mix decisions of new product development. However, the nature of these last activities were not clear defined in terms of how radical innovation, standardization and task uncertainty.

H1b: The ability of the marketing functions to be financial outcome accountable is positively related to their ability to perform exploitative activities.

Procedural accountability provides management with a way to improve, by means of having some initial guidance to behave and improve. The necessity and pressure to always be successful is relatively lower for procedural accountability and is accompanied with diminished stress. Effects of procedural accountability on the performance of the marketing function will therefore be more favorable (Scholten, Van Knippenberg, Nijstad, and De Dreu 2007). This limits potential undesirable behavior with persons being held accountable (Siegel-Jacobs and Yates 1996; Simonson and Staw 1992).

H2a: The marketing function’s ability to be procedural accountable is positively related to their ability to perform explorative activities.

H2b: The marketing function’s ability to be procedural accountable is positively related to their ability to perform exploitative activities.

3.2 ORGANIZATIONAL CULTURE MODERATING MA AND

PERFORMANCE OF MARKETING ACTIVITIES

(25)

25

shared norms underlying MA, whereas accountability is part of the organizational climate: the routines of organizations and the behaviors that are rewarded and expected by the organization (Schneider and Rentsch 1987, quoted by Deshpandé and Webster 1989, p. 5). Both are conceptually different, yet the climate relies on the culture to a certain degree. This leads us to believe that the organizational culture affects the effectiveness of the organizational climate (i.e. the ability of the marketing function to be accountable).

There is no mistaking the importance of organizational culture in modeling the interactions between members of an organization. It affects how people behave outside the formal processes and how is reacted to the formal processes that are in place. Marketing is not unfamiliar to the importance of organizational culture. Market and customer orientation are two prime examples of how a culture can influence behavior and performance within the firm. The influential paper by Deshpandé and Webster (1989) provided relevant directions for research of organizational culture from a marketing perspective, and Deshpandé, Farley, and Webster (1993) where they assess organizational culture from a customer oriented standpoint. Ultimately, this lead to research emphasizing market orientation as the firm-wide shared values and norms in the firm (Homburg and Pflesser 2000; Slater and Narver 1995). Organizational culture can therefore not be dismissed in this research.

Market and customer orientation are well-established conceptualizations of organizational culture in marketing literature, however, we aim to present a different perspective of organizational culture in relation to marketing accountability and the performance of marketing activities. We aim to explain organizational culture in terms of individualism versus collectivism. Individualism “emphasize[s] separating from others and promoting one’s own internal attributes…”, and collectivism ”emphasize[s] maintaining relatedness and fitting in with relevant others” (Gelfand and Christakopoulou 1999). Though collectivism vs individualism has a widespread use as one of the Hofstede dimension for cultural differences between countries, it also has a considerable base of application in organizational culture.

(26)

26

own beliefs, values, and attitudes), and 4) emphasis on relationships (individualists are more oriented toward task achievement, sometimes at the expense of relationships, whereas collectivists put more emphasis on harmonious relationships, sometimes at the expense of task accomplishment).

Several of these attributes demonstrate a possible affiliation to how strong marketing accountability may affect their ability to perform different types of activities. Attribute 2 emphasizes that individualistic cultures prefer to follow own individual goals instead of corporate goals whereas collectivistic cultures would comply with goal setting of the firm. Considering accountability refers to demonstrating one’s contribution and performance to the person or group you are held responsible to, then individualistic cultures may have more difficulty to conform to such practices. This idea is strengthened by attribute 3 which states that individualistic cultures are driven by own values and norms, so not the values and norms of the firm which emphasizes scrutiny of financial accountability. However, attribute 4 says individualistic cultures focus more on task performance which would likely improve the effects of MA on the ability of marketing to perform activities. Thus, an individualistic culture would facilitate maximization of goal achievement and rewards good performances (Chatman and Barsade 1995). Therefore, we postulate a moderation effect individualistic and collectivistic organizational culture, which is in line with Gelfand and Realo (1999) who find support for this effect in the setting of accountability on negotiators’ psychological states, behaviors, and outcomes. We express no direction of moderation as the degree of which attributes manifest most may affect the direction, though we do perceive adequate support from literature that moderation is present.

H3a: The presence of an individualistic/collectivistic organizational culture positively or negatively moderates the effect of financial outcome accountability on (1) explorative marketing activities (2) exploitative marketing activities.

H3b: The presence of an individualistic/collectivistic organizational culture positively or negatively moderates the effect of procedural accountability on (1) explorative marketing activities (2) exploitative marketing activities.

(27)

27

relate positively to explorative marketing activities. Additionally, the study by Moorman and Blakely (1995) suggests collectivistic cultures to promote organizational citizenship behavior (behavior that is not directly or formally related to reward systems but promote effectiveness of the firm) providing indications of a positive relationship with exploratory marketing activities.

3.3 PERFORMANCE OF MARKETING ACTIVITIES AND

MARKET PERFORMANCE

Many different ways exist to measure different types of firm performances. The current development of firm value metrics and mechanisms for the marketing function are a testimony for the continuous need to evaluate performance of firms and organizations. The primary focus in this paper is the effect of the marketing accountability dimensions on the behavior of the marketing functions and their ability to perform explorative and exploitative activities. Hence, given the empirical setting we opt for the use of more marketing related performance measures of the firm to assess the effect of marketing initiatives on firm performance. We look at 2 types of firm performances, namely market performance, and innovation performance.

Kyriakopoulos and Moorman (2004) find support for marketing exploration strategies and marketing exploitation strategies, presenting high levels of both, having a positive impact on new product financial performance. This tells us that well-performing marketing actions will positively affect firm performance. More specifically, Drechsler, Natter, and Leeflang 2013 find that innovation performance positively affects business performance which is in line with our hypothesis that explorative marketing activities (innovation is an important characteristic) increases firm performance. Based on extant literature regarding the effects of marketing activities on firm performance we argue that marketing activities will positively relate to market performance (Raisch, Birkinshaw, Probst, and Tushman 2009).

H4a: Marketing function’s ability to perform explorative marketing activities is positively related to both (1) market performance and (2) innovation performance of the firm.

(28)

28

Additionally, we see increasing evidence from literature for the benefits of accommodation both exploration and exploitation leading to superior firm performance (March 1991; Benner Tushman 2003; He and Wong 2004). As such, we assess whether the ability of the marketing function to perform activities of both exploration and exploitation well will lead to increased market performance by balancing the activities in some manner e.g., ambidextrously (i.e., balancing exploration and exploitation by means of simultaneous consideration) (Raisch, Birkinshaw, Probst, and Tushman 2009).

H4c: Marketing functions that perform high on both explorative and exploitative dimensions have better market performance relative to marketing functions that are only able to perform high on a single dimension.

3.4 MEDIATION OF MARKETING ACTIVITIES’

PERFORMANCE ON MARKET PERFORMANCE

Extant literature has demonstrated a direct positive relationship of marketing accountability on firm performance (O'Sullivan and Abela 2007). Yet, with the addition of MA in conceptual frameworks no changes in the ability of the marketing function to perform tasks and activities are accounted for. Currently, the behavioral and perceptual effects that potentially facilitated the positive relationship direct relationship of MA on market performance (i.e., a type of firm performance) are considered as a black-box and are mostly obscured. In order to improve upon the research for intermediate effects of marketing accountability, we build further on Mintz and Currim (2013) who established the link between metric use and performance of marketing activities (i.e. marketing-mix activities). Additionally, we aim to improve the assessment of MA effects towards the broader classification of marketing activities as explorative and exploitative. We argue that, with the inclusion of the ability of the marketing function to perform explorative marketing activities and exploitative marketing activities, the direct relationship of MA on the two types of firm performance will fully or partially disappear.

(29)

29

strong reasons to investigate the extent marketing activity performance mediates the direct relationship of marketing accountability on market and innovation performance.

H5: The relationship between marketing accountability on market and innovation performance is mediated by the performance of marketing activities.

3.5 CONTROL VARIABLES

We account for a multitude of additional effects apart from the described hypotheses. This is to ensure the hypothesized relationships are due to the constructs under analysis that are of main importance to this study. Many different forces affect the firm and its members, who may cause changes in the focal constructs that, are not of primary interest. We, therefore, control for potential effects that might bias the estimations of the relationships in our model. Secondly, the unit of analysis is very heterogeneous. Marketing functions across firms have different tasks, goals, and compositions. Similarly, different firms have different characteristics that may result in deviating behavior by the presence of marketing accountability, while the markets in which firms operate are also heterogeneous which potentially affect the degree to which marketing function activities affect firm performance. As such, we use the perspectives of Mintz and Currim (2013) to determine the control variables of 1) the marketing function, 2) the firm, and 3) the market environment.

In order to control for the heterogeneity within the marketing function we assess the marketing function’s size, their budget, and the experience of marketing management. In compliance with the resource-based view, we select the firm characteristics. We add firm size, the type of ownership, the offering of goods or services, and the degree of business done with consumers or other businesses. Lastly, market environment characteristics are chosen on the contingency perspective, where the most important and comprehensive factors are market growth and market dynamics.

(30)

30

Chapter 4 Methodology

To test our conceptual model a self-administered e-mail survey is developed which consists of closed-formulated questions. Data is collected for the variables of interest via multiple-item questions asking participants to indicate their perceptions on accountability abilities of the marketing function, performance of marketing-related activities, firm performances, organizational and managerial settings, and environmental conditions. 7-point Likert scales are used for the main variables in the model, and mostly have proven to be effective by previous research, though for some concepts adapted or new scales needed to be developed. Adapted and new scales, received similar wording to related proven constructs including the 7-point Likert scale to maintain consistency throughout the questionnaire, and to improve the development of reliable and valid measures. Furthermore, our questionnaire consists of both formative and reflective scales which are to be combined if allowed. For an overview of measurement scales used in the questionnaire of both main variables and control variables, see appendix A.

4.1 MEASUREMENTS

(31)

31

accountability, and to lesser extent considered as the ability to be accountable as is most prevalent within marketing literature (see figure 1). As such, a new scale is developed by first adapting the wording of questions of the proven financial outcome accountability scale into a procedural orientation. Since, to our knowledge, this study is a first to empirically assess procedural accountability, additional items have been added that characterize the marketing function’s ability to be accountable for the process. This aids the development of a good measurement scale with a set of items that represent the concept best. This resulted in a 7-item reflective Likert scale that determines the perceived ability of the marketing function to be procedural accountable.

The performance of marketing activities has been measured in many different ways by extant research. Leading in our research is the split between exploration and exploitation. This well-known distinction in innovation research comprehensively covers all behaviors in two dimensions, regardless whether it pertains to competences, strategies, or activities. In this paper, the focus is on the latter by looking at the performance of marketing activities. It is important to note that we do not look at the degree to which the two types of activities are executed, but the degree the marketing function is able to perform on these activities. This is an important distinction as one of our objectives is to assess how an increasing ability to be accountable can affect the performance of the marketing function, and the extent that this performance in turn relates to changes in firm performance. To address the marketing function’s performance on exploration and exploitation activities respectfully, two sets of questions have been adapted from March (1991), and Mom, Van Den Bosch, and Volberda (2007) to arrive at two reflective 6-item Likert scales. The perceived performance of the marketing function is administered for marketing activities that are characteristic for exploration (e.g., searching for new opportunities, risk taking, and long-term focus), and exploitation (e.g., existing knowledge, efficiency, and short-term goals).

(32)

32

managers need to cope with potential conflicting abilities to be accountable and to perform on marketing tasks. We measure organizational culture with Individualism versus Collectivism as ends of a continuum. In search of an appropriate scale for the concept the number of items was an important criterion to minimize overall questionnaire length. A tested and supported measurement scale was found in the paper of Moorman and Blakely (1995) in which the concept was divided into three different indicators (i.e., beliefs, values and norms) with the first two indicators consisting of 3-item questions, and the third of 5-items. All items are developed as a Likert scale. The items to each indicator are reflective whereas the three indicators are formative to the organizational culture concept.

To determine firm performance we look at market performance and innovation performance of the firm. Both constructs are formative with a 7-point Likert scale, though the former consists of five items and the latter of three items. Participants are to provide their perception on the firm’s performance relative to stated objectives by the firm from 1=”much worse” to 7=”much better”. Scales from Moorman and Rust (1999), and Verhoef and Leeflang (2009) are used as the basis for both constructs. Additionally, for market performance we assess the firm’s performance relative to its main competitors.

(33)

33

marketing’s ability to be accountable (also see figure 1). These are the intensity of being held accountable by the firm, number of average appraisal measures used for marketing activities, relative focus on marketing versus financial metrics, focus on outcome versus process appraisal, a reflective Likert scale for outcome control focus (Jaworski, Stathakopoulos, and Krishnan 1993), and a reflective Likert scale for process control focus. Though these accountability controls are of a precautionary nature in this study, they also hold considerable promise as main variables considering the lack of attention to the perception triangle of the accountability concept that is previously discussed, as well as the connection between a function’s abilities and the use (financial) metrics.

4.2 QUESTIONNAIRE DEVELOPMENT

We used the Qualtrics software package to create our questionnaire. Qualtrics has multiple advantageous, build-in applications that improve survey quality, user friendliness, and time-efficiency of development. It enables researchers to organize questions in blocks, build in skips (or loops), and has a wide selection of question formatting possibilities. In this section, we will discuss in more detail how we developed the survey with use of this software.

(34)

34

wished to do so. This cover letter was uploaded to the build-in mail application of Qualtrics, which allows developers to upload multiple letters for multiple e-mail address panels, as well as for reminders.

Secondly, a questionnaire instruction text was designed for potential participants that clicked on the link to open the survey (see appendix C). This was used to introduce the questionnaire further and provide supporting information on answering the questions. The subject was shortly mentioned, information was provided that questions mainly pertain to participants’ perception, attention was asked to reverse coding and to carefully read questions, time to participate was displayed, and browser usage was recommended.

Thirdly, the items of each construct and measurement scales were entered. The survey was designed in such a way that participants were not required to scroll while participating1_. This was realized by creating question blocks (mostly 1 block per construct) with each block receiving a title on the subject that is questioned. Participants could use back and forth arrows within the window of the website to navigate through the pages of questions. A progress bar was displayed to inform the participant how long completion would take. If participants would perceive the survey to be too long and they would close the survey, then progress would be saved and participants could continue where they left when reopening the survey.

Lastly, several techniques were applied to facilitate the correct questions were answered correctly. Two skip patterns were created. For the first skip it was questioned whether the firm of participant maintained a marketing function. If not, they were led to the first of two consistency checks asking participants whether the first question was correctly entered. If it was correct, the participant was exited from the survey, if not correct, the participant was kindly requested to use the back-button to correct their answer. The participant would continue to the rest of the questionnaire, when the firm did have a marketing function. The second skip pattern was for the organizational culture question block. It pertained to whether the participant worked in work groups. Depending on the answer a list of questions was show with slightly different wording when the person did or did not work in work groups. The second consistency check was placed before the last page where the participant would see a message that they exited the questionnaire and that results would be submitted. This was done to avoid participants skimming through the pages and accidentally exiting the survey before being able to fill in their responses. Lastly, we

(35)

35

randomized the ordering of question blocks for each respondent. This is done to reduce bias from learning effects from subsequent questions, fatigue at the end of questionnaire, and other potential effects due to ordering of questions.

4.2.1 Pretesting

In order to test the quality and validity of our questionnaire several pretests were conducted before data collection took place.

An expert panel was used to assess the concepts with their respective measurement items and scales. The assessment pertained validity of used items, ambiguity of wording, clarity of what is questioned, completeness, and so forth. This group consisted of 6 people of which 3 PhD candidates (two with an affinity to marketing, one with little affinity to the field), two Research Master graduates in marketing, and one marketing professor2_{. Feedback} was provided and incorporated in the questionnaire, when it was perceived as quality enhancing.

The next step was a third person back-translation of the measurement scales and items. First, we translated the measurements from English to Dutch. Then one of two PhD candidates from the expert panel, translated it to English, and the other subsequently translated that back to Dutch. The PhD candidates are both fluent in English and Dutch. Translations are compared for differences in wording and interpretation, so English-to-English and Dutch-to-Dutch. Little problems arose, and some different wordings were resolved after discussions with both translators. The result was a well-translated questionnaire in Dutch that was used for pretests with persons from practice.

Two of such pretests were performed with members of firms. The participants from each pretest were taken from two separate workshops of two professors at the University of Groningen. In total 6 from the first group responded and four from the second group. Both groups were asked to fill in the questionnaire and to comment on it, based on clarity, ambiguity, errors, etc. Feedback from the first pretest was incorporated for the second to maximize the value from each subsequent pretest group. Overall, little changes were required apart from some textual alteration for questions unrelated to the main variables in the conceptual model.

2_{The supervisor of this thesis was also considered as part of the expert panel as the questionnaire}

(36)

36

4.3 PARTICIPANTS

The participants of our questionnaire consist of profit-seeking firms located in the Netherlands that are large enough to support and sustain a separate marketing function within the organization. For this study, firms that have 200 or more employees are considered3_{. The members of the firm that are targeted are management with a close} affiliation to the marketing function, such as senior marketing managers, marketing executives, and general managers. Members of the firm that participate in the questionnaire should be knowledgeable of marketing function’s ability and behaviors, firm level performances, as well as environmental conditions. We believe members of the organization in management positions are most appropriate and are targeted for this study.

4.3.1 Sampling frame

To acquire participants for our survey we adopt several selection frames. Firstly, we acquired contacts via the Customer Insights Center (CIC) which is an organization maintained by the University of Groningen that produces practically relevant research from which companies can derive relevant insights for their businesses. This source contains mainly large firms of which personal contacts are available. Secondly, we acquired firm contact information database from Graydon. A list of 622 contacts was obtained that would meet the criteria for our population (profit-seeking firms, head of marketing contact information, large firms with employees > 200, situated in the Netherlands, contact information of headquarters only).

Firms were contacted personally via telephone to acquire e-mail addresses for questionnaire distribution as no such information was available in the contact database. However, before we could do so, it became clear from a preliminary check that still not all contacts met our criteria and had to be eliminated from the list. Foundations, municipalities, branch firms, unions, associations had to be taken off the list. They are outside our population of this research as we believe they either have no distinct marketing function in their organization, or if they have, its function is likely to deviate substantially from those within profit-seeking firms. This resulted in 497 firms that were eligible.

3_{This cut off value of 200 employees is mainly due to the catergorization of the acquired contact list}

(37)

37

4.3.2 Response rate

Information from the Graydon contact database, that was used to phone the firms, were 1) name of the firm, 2) phone number of headquarters, 3) last name of head of marketing, and 4) gender of head of marketing. Additional firms were deleted from the list of those that no longer existed, had an international headquarter from where marketing was conducted, or did not have a marketing department. Call backs were conducted if so requested by the firm. This resulted in a total of 299 e-mail addresses consisting of 96 general e-mail addresses and 203 personal email addresses. For 48 general e-mail addresses we could provide names to which the e-mail was addressed4_.

The CIC provided a database of personal contact information of members from 26 firms, including e-mail addresses. The most eligible person per firm was determined with use of LinkedIn. They were directly contacted via the build-in e-mail distribute program of Qualtrics. After disappointing initial response from this group a new set of secondary e-mail addresses was selected from the CIC contact database of the firms that had not participated. Four firms initially participated, meaning 22 new e-mail addressed were used. Hence, in total 48 people were contacted from the CIC.

A number of methods were used to enhance response rate. Two reminders were used for the CIC5_{and Graydon panel (see appendix B). They were planned approximately with} intervals of two weeks depending on the rate of response. Each subsequent reminder letter was adapted to persuade participation. As such, reminder mails were shortened providing more condensed information methods to enhance responses. The last reminder was again shortened providing only core information. Additionally, important information was accentuated by underscoring and bolding of text. A different method to enhance response was the promise of giving something in return for participation. Respondents that fully completed the survey would receive insights from the research in terms of benchmark statistics of their own scores relative to averages scores of participating firms. Additional advantages are that the likelihood of obtaining fully completed surveys, once started, is higher, and if managers are truly interested in how they compare to other firms their answer will be more truthful. This resulted in a response size of 58 responses of which 51 from

4_{Although the Graydon list provided names of head of marketing, it was frequently the case that the}

person was either not from marketing, never worked at the firm (or not to their knowledge), or had already left the firm for quite some time. Correct names of marketing were not at all instances provided by the firm.

5_{Both the first and secondary CIC panel received two reminders. The initial invitation of the second}

(38)

38

Graydon and 7 from the CIC. After eliminating the questionnaires that were not usable in terms of completeness6_{the respective response rates, relative to the number of sent e-mail} addresses, are 15.7% (i.e., 47 responses) and 14.6% (i.e., 6 responses) giving an overall response rate of 15.3%. The Graydon response rate relative to the total number of eligible firms is 9.5%.

In light of the minimal response in absolute terms from data collection efforts, it was decided to include response from the pretest groups. As no significant changes resulted from the pretests to the questionnaire we can reasonably argue that little differences would result from the textual changes to the questionnaire that were unrelated to variables of the main model. In total, six responses were usable for analyses from the pretest group. To be certain, the initial responses and the pretest responses were compared for differences. Independent samples t-tests were performed for the main constructs of interest in our model. No differences were found (p > .05). Therefore, the data points were added resulting in a total sample of 59.

4.3.3 Non-response bias

In order to test for non-response bias we use the technique from Armstrong and Overton (1977) for which the database is separated in 3 equal parts in chronological order of response time. The first group (i.e., early respondents) and last group (i.e., late respondents) are compared for differences. The underlying thought of this technique is extrapolation, which argues that late respondents would more likely resemble non-respondents than early respondents would. If those two groups are compared, and differences were to be found, then that would be an indication that non-response bias may be an issues. For this test the database was split into 3 chronological groups, for which the first and last chronological groups were compared. The t-test demonstrates that there are no differences, so non-response bias is not likely to be an issue.

Also, the collected e-mail addresses from the Graydon list were contacted on three different occasions of approximately 100 e-mails at each time. This was done out of time considerations so data could be collected while the remainder of firms could still be called and mailed at a later time. Since we have 3 different groups of collected responses from the contact database, and two additional group from the CIC and pretests, we also compared differences between these groups to ensure they behave similarly and it is justified to

6_{The deleted firms, according to the respondents, did not have a marketing function, and, hence, were}

(39)

39

combine them for analysis. Separate t-tests for the main variables in our model show that no significant differences were found for an alpha of .05, nor for 0.10, hence we are allowed to combine the groups into one database.7

4.4 RELIABILITY AND VALIDITY

4.4.1 Reliability and convergent validity of measures

We assess whether the measures of our constructs are in fact related. We do so by estimating Cronbach’s Alpha’s for all multi-item scales in our model. Specifically, we test for the internal consistency reliability of measurement items to the degree they produce similar results. The Chronbach’s Alpha estimates the average inter-item correlation, which should be .7 or higher for the measures of a construct to be reliable in terms of internal consistency (Hair Black Babin and Anderson, 2010). Based on the Chronbach’s Alpha’s provided for the constructs in table 3 and appendix A, we can conclude that overall all construct of the main model have a high internal consistency. Of control variables, customer orientation (α=.659) and management experience (α=.424) are less internally consistent, which is not unexpected for management experience as it is not reasonable to assume that years as manager, years at the firm, and years at current position are highly interrelated. Also, deleting one item does not improve internal consistency enough. So we consider the three items as separate variables. All other multi-item constructs have a high internal consistency, so we can conclude that the measures are overall reliably consistent for their respective constructs.

Table 3. Internal consistency values of constructs

Construct # of

items Chronbach’s α Highest α if one item is deleted Conclusion

Financial outcome

accountability 3 .834 .868 Create composite variable Procedural accountability 6 .853 .850 Create composite

variable Performance of explorative marketing activities 6 .887 .889 Create composite variable Performance of exploitative marketing activities 6 .802 .825 Create composite variable Individualism vs. 11 .792 .793 Create composite

7_{In correct order, we first addressed differences within Graydon list groups, then within differences}

(40)

40

collectivism variable

Market performance 5 .844 .897 Create composite variable Innovation performance 3 .850 .804 Create composite

variable Intensity of held

accountability 4 .785 .814 Create composite variable Outcome control focus 5 .877 .924 Create composite

variable Process control focus 4 .873 .903 Create composite

variable Customer orientation 9 .659 .709 Do not create

composite variable Management experience 3 .424 .476 Do not create

composite variable Market turbulence 6 .777 .773 Create composite

variable

THE DARK SIDE OF MARKETING ACCOUNTABILITY: