An Empirical Analysis on Digital Orientation: Exploring the Relationship between Digital Orientation and Innovation Performance and the Moderating Role of the Presence of a Chief Digital Officer

(1)

An Empirical Analysis on Digital Orientation: Exploring the

Relationship between Digital Orientation and Innovation Performance

and the Moderating Role of the Presence of a Chief Digital Officer

Yanlan W. van Wonderen* | Supervised by N.E. Fabian** | Co-assessed by Prof. Dr. Q.J. Dong** * Student MSc BA Strategic Innovation Management | Faculty of Economics and Business | University of Groningen

** Faculty of Economics and Business | University of Groningen

Keywords

Digital Orientation · Dynamic Capabilities · Chief Digital Officer · Strategy · Innovation Performance

Abstract: In an era where technologies are entwined with our every-day lives, changing, transforming

and disrupting industries, we have to extend our understanding on how these technologies can contribute to innovation. Introducing a new strategic route; digital orientation may offer the solution to firms competing in today’s market. Based on the dynamic capabilities theory, sensing, seizing and transforming, three elements were defined to reflect digital orientation, namely, digital marketing capability, technological capability and cross-functional integration. Additionally, the need for a link between strategy and technology is tested through the moderating effect of the presence of a Chief Digital Officer and its impact on the relationship between digital orientation and innovation. The hypotheses were tested with a longitudinal dataset over a period from 2010 to 2019 using a sample of 609 U.S. based firms listed on the S&P 500 in the manufacturing, service, retail and telecommunications sectors. The empirical findings support the positive relationship between digital marketing capability, technological capability and innovation performance. No empirical results were found for cross-functional integration and moderation effect. My research contributes to the limited literature available on digital orientation. Finally, it provides practical implications for managers.

“90% of CEOs believe the digital economy will impact their industry, but less than 15% are executing on a digital strategy.”

- MIT Sloan and Capgemini

(2)

List of Figures and Tables

Table 1. Skill Clusters Example ... 20

Table 2. Measurements of Variables ... 24

Table 3. Descriptive Statistics ... 27

Table 4. Correlation Matrix ... 27

Table 5. Regression Analysis ... 29

(5)

1. Introduction

Our era is characterized by digital technologies. Everywhere you go, everywhere you look, digital technologies are infused in our lives. We turn to our mobile phones to connect with people, to the web to book a flight and to our laptops to stream music. What was once primarily used by technical people, is now used by people all around the world. The ease and the comfort digital technologies offers to people has not only changed the way individuals embrace technology, but also caused firms to re-think their involvement in digital technologies. Technology is no longer seen as a supportive tool, it is embedded at the very core of products and services of many firms. The significant importance of digital technologies is illustrated as 78.85% of firms rate digital technologies as very important for their business (Statista, 2019). As technology developed, new winners and losers were formed. The pace of innovation that coincided with technological advances disrupted and transformed industries. The rising evolution of technology continued and their impact began to grow, incumbent firms’ realized that digital technologies were no longer an operational concern, but a strategic one offering a new strategic route; digital orientation.

(6)

opportunities in digitally oriented firms is reflected in technological capability (Wang, Zhang, Lo & Xue, 2006). This satisfies the exploitation of opportunities through their technological knowledge base (Halaç, 2015) and leveraging this by engaging in emerging technologies that can be used to pursue innovation activities (Day & Schoemaker, 2000). Lastly, transforming capabilities in digitally oriented firms are reflected in cross-functional integration (Troy, Hirunyaipada & Paswan, 2008). This satisfies the integration of knowledge through technical mechanisms that facilitate knowledge transfer and sharing that allow firms to learn from past performance to ensure future performance (Scherman, Berkowiz & Souder, 2005).

Digital orientation is often measured against innovation performance as they share a disruptive and ever-changing nature (Yang, Wang, Zhu & Wu, 2012). Mentioned previously, digital orientation has received limited attention in existing literature (Yang et al., 2012; Hortinha, Lages & Lages, 2011). However, when researched, this was often in the context of SMEs (Yang & Wu, 2012; Hortinha, Lages & Lages, 2011), which are significantly different from large firms. Moreover, existing literature focuses on surveys and interviews (Yang et al., 2012; Hortinha, Lages & Lages, 2011), failing to use other data. There is a need for more research in regards to digital orientation, what it entails and how it affects innovation performance as they have the ability to transform and destroy industries (Day & Schoemaker, 2000). In efforts to fill this gap, I will focus my research on large firms listed on the Standard & Poor’s 500 by analyzing big data that involve skill-clusters that represent the capabilities needed for digital orientation. This leads to the following research question:

RQ: What is the relationship between digital orientation and innovation performance? And how does the presence of a Chief Digital Officer moderate this relationship?

(7)

The empirical results support the positive relationship between digital marketing capability and innovation. Similarly, a significant positive result was found for technological capability and innovation performance. Cross-functional integration displays a negative non-significant result. Lastly, the moderation effect was partially supportive for the interaction effect between digital marketing and Chief Digital Officers, not supporting the other two. My findings contribute to existing literature as it adds on the scarce research done on digital orientation and innovation by (1) broadening the knowledge on digital orientation (2) providing a working definition on digital orientation, (3) identifying what elements it takes to become digitally oriented (4) testing this concept in large U.S based firms (5) assessing digital orientation using big data. Thid resulted into managerial recommendations and suggestions for future research.

(8)

2. Theoretical Background and Hypothesis Development

2.1 Digital Orientation: Towards a Definition

Digital orientation hereafter referred to as DO, falls under the overarching term known as a firm’s strategic orientation. A strategic orientation can be seen as a package of managerial choices made to achieve desired business objectives (Venkatraman,1985; Leskovar-Spacapan & Bastic, 2007).Two principal typologies examined strategic orientations, namely, Porter’s Generic strategies (1980) and Miles & Snow’s strategic orientation (1978). Both use different business-level criteria that assess business strategies, yet share a similar goal to increase the number of dimensions a firm’s strategic orientation is evaluated upon (Govindarajan, 1986). They differ as Porters’ typology (1980) is more generic and Miles & Snow’s typology (1978) is more specific in nature (Segev, 1989). Porters’ generic strategies are classified based on economic factors, centred around a firm’s competitive advantage and the scope of their target market, handling an inside-out approach (González –Benito & Suárez-González, 2010). In contrast, Miles & Snow (1978) take an outside-in approach, evaluation is decided upon the external environment of the firm (Segev, 1989). Despite differences, one overlapping conclusion is found as both models do not vouch for firms “being stuck in the middle” meaning those not adopting a specific strategy (Segev, 1989). Both streams of research conclude that this negatively affects firms’ success as they are ill-conceived strategies that are incompatible with their external environments failing to cater to demands in the market (Porter, 1980; Miles & Snow, 1978). In conclusion a specific and well-formulated strategic orientation is advised (Segev, 1989).

Working towards a definition of DO both typologies are considered. Given that DO is characterized by rapid innovation and uncertain and complex external environments (Byrne, Oliner & Sichel, 2017), the alignment with those external factors is crucial to sustainable performance, hence DO would lend itself better to be classified against external factors following Miles & Snow’s typology (1978). Traditionally, Miles & Snow (1978) identified four strategic orientations; entrepreneurial, defender, analyser and reactor. Noted here is that the reactor orientation is often disregarded because its value is questionable and often leads to failure (Matsuno & Mentzer, 2000; Miles & Snow, 1978). Building on those original four, the development of three new strategic orientations firms adopt; market orientation, entrepreneurial orientation and learning orientation (Kropp, Linsdsay & Soham, 2008; Hakala, 2011; Long, 213; Huang & Wang, 2011; Baker & Sinkula, 2009).

(9)

through the improved alignment between customer demands and offerings of firms (Slater & Narver, 2000). Regardless, this is conditioned by the external environment of firms (Grinstein, 2008). This positive effect was mostly found in firms operating in markets that are not subject to high technological turbulence (Grinstein, 2008). Especially in SMEs, market orientation offers significant advantages, the flexibility and minimized layers of firms allows them to diffuse information quickly across the organization and respond rapidly to customer demands (Verhees & Meulenberg, 2004). However, firms that are too dependent on information provided by customers may cause firms to miss other opportunities to innovate (Zhou, Yim & Tse, 2005). As argued by Grinstein (2008), this is especially problematic in markets that experience turbulence due to fast technological changes. In such markets, learning orientation may offer more benefits.

Learning orientation refers to the ability of firms to create and use knowledge to enhance their competitive advantage (Calantone, Cavusgil & Zhao, 2002). This orientations depends on all types of information sources, internal and external, competitors and customers to steer new developments (Calantone, Cavusgil & Zhao, 2002). Learning orientation influences how knowledge is obtained and interpreted making it an important complement to market orientation (Mavondo, Chimhanzi & Stewart, 2005). Especially in competitive and turbulent environments learning orientation offers significant valuable input for innovation whereas market orientation may miss out (Baker & Sinkula, 1999). In market orientation, the emphasis is on customer information, but they might not be the most resourceful sources when it comes to complex and uncertain environments. Other sources such as competitor analysis may be more resourceful as they are better able to understand the complexities (Zhou, Yim & Tse, 2005). Particularly for innovation purposes, firms deal with uncertain environments in which they need to handle complex information that learning orientation can assist in (Baker & Sinkula, 1999). However, in the context of innovation one orientation has been prominently studied; entrepreneurial orientation (Lumpkin & Dess, 1996; Miller, 1983; Wiklund & Schepherd, 2005).

(10)

Being stuck in the middle has gotten numerous negative comments (Porter, 1980; Miles & Snow, 1978). However, in recent studies this perspective has changed. Instead of seeing these orientations has mutually exclusive, they should be seen as complements (Slater & Narver, 200; Quionton et al., 2017). Hence all three orientations have elements that contribute to an adequate definition of DO. In coherence with Quinton et al. (2017) and Gatignon and Xuereb (1997), DO is defines as:

The deliberate strategic position of a firm to take an advantage of opportunities presented by digital technologies and the will to acquire substantial technological knowledge and use this in the

development of new products.

This definition covers three key elements: 1) combining elements of each orientation where market and entrepreneurial orientation help sense and seize new opportunities. And learning orientation to foster continuous renewal based on technological knowledge; 2) linking digital orientation to innovations, finally, 3) taking a firm-level approach.

2.3 Digital Orientation and Innovation

(11)

2.4 Dynamic Capabilities and Digital Orientation

The environment in which DO is adopted advocates the need for dynamic capabilities (Teece, Pisano & Shuen, 1997) and thus used to identify what elements are needed for digital orientation. Dynamic capabilities are defined as a “firm’s ability to integrate, build and reconfigure internal and external competences to address rapidly changing environments” (Teece, Pisano & Shuen, 1997). These uncertainties make it difficult to probe for the future hence dynamic capabilities offer firms the flexibility to reconfigure and adapt competences to achieve sustainable financial performance (Teece, Pisano & Shuen, 1997). Dynamic capabilities can be disaggregated into the capacity of (1) sensing and shaping opportunities, (2) seizing opportunities and (3) maintain competitiveness through transforming those capabilities to meet changing demands (Teece, 2007). Dynamic capabilities are closely related to innovation as they share similar characteristics and purposes (Breznik & Hisrich, 2014). Both are focused on the role of learning to sense and seize new opportunities and the continuous process of adjustment (Brexnik & Hisrich, 2014). These overlap with the definition provided of digital orientation hence the dynamic capability perspective will be used to illustrate elements that constitute digital orientation.

2.5 Sensing and Shaping Opportunities

(12)

Digital Marketing Capability

Digital marketing capability is a component of marketing that utilizes internet and online-based technologies and platforms to promote products and services. Besides promoting products and services, marketeers are also retrieving information from the market that allow them to spot new opportunities. In this capability the effects of technologies are crystal clear and how it has shifted its perception towards technology. What once was considered a non-technological department, now has technology at the very core of their operations (Moncrief & Cravens, 1999).

Technology offers significant benefits to current marketing activities. It facilitates a fast, transparent and easy way of communication with customers (Moncrief & Cravens, 1999). This communication takes place via web-based platforms and has become the primary way of interacting with customers (Moncrief & Cravens, 1999). Moreover, this transparency helps firms better detect what went wrong and what went well (Chatterjee, 2001). Customers can voice their opinion on websites, known as online reviews which can be used for firms to increase customer knowledge (Chatterjee, 2001). Additionally, marketeers are able to form personalized relationships with customers through digital marketing (Wind & Rangaswamy, 2001). As a result of online communication channels deep customer data is generated while interacting with customers (Stone & Woodcock, 2013). Leveraging this data is done through analytical activities that help deepen understanding on customers, detecting patterns and provide new insights (Leeflang et al., 2014). This data can be used to improve existing relationships whilst probing for potential relationships (Leeflang et al., 2014). The ability to retrieve and assimilate this data has become an important tool to outperform competition (Davenport, 2006). Systematically tracking customer data can help sensing opportunities that lead to new segments, new business models and revenue streams that are likely to achieve market acceptance (Leeflang et al., 2014; Teece, 2007).

(13)

Despite of previous concerns, digital marketing capability is promising for fostering innovation. Contrary to seizing and transforming capabilities which are focused on internally improving business, sensing capabilities are focused on external sources for improvements. This outside-in focus has been studied widely in the context of innovation performance (Weerawardena, 2003; Merrilees, Rundle-Thiele & Lye, 2010; Mariadoss, Tansuhaj & Mouri, 2011). Digital marketing capabilities have a dual nature in influencing innovation. Namely, in the (1) development stage and facilitates innovation (2) in the commercialization stage (Weerwardena, 2003). Since digital marketing capability is used to sense opportunities, the emphasis is on the former. Digital marketing capabilities generate information that is a crucial determinant of the development of new products (Song, Nason & Di Benedetto, 2008). In coherence with Mariadoss, Tansuhaj & Mouri (2011) who argue that adequate marketing capabilities are needed to recognize opportunities that will be developed into successful innovations. In conclusion, digitally oriented firms use customer data resulting into rich insights on demands, patterns, expectations and provide information that can lead to the identification of new opportunities (Leeflang et al., 2014). The role fulfilled by marketeers has changed significantly in respect to DO and their input has become much more influential in guiding firms where to go next (Tiago & Verrísimo, 2014).

Therefore, the following hypothesis is proposed:

H1. Firms that have high levels of digital marketing capability positively influence a firm’s innovation performance.

2.6 Seizing and Exploiting Opportunities

Once an opportunity is sensed, it must be seized through new products, processes or services (Teece, 2007). This includes; improving technological competences, investing in new technologies and adapting a business model that will result into market acceptance (Teece, 2007). Argued by Chandler (1990), successful firms are able to pursue a three-facet strategy; (1) early and large-scale investments in new technologies; (2) investment in specific marketing, distribution and purchasing networks and (2) recruiting and organization adequate managers to coordinate functional activities. Therefore, seizing opportunities in a digitally oriented firm is reflected in technological capability. This capability fulfils seizing opportunities by building a knowledgebase in which technological superiority acts as the foundation to exploit opportunities and the engagement in emerging technologies.

Technological Capability

(14)

capability is defined as “a set of pieces of knowledge that includes both practical and theoretical know-how, methods, procedures, experience and physical devices and equipment (Wang et al., 2006). Increasing technological capability requires investments in R&D and new technologies (Afuah, 2002). These investments increase a firms’ ability to accumulate and assimilate new information (Berkhout, Hartmann & Trott, 2010). This can lead to a superior level of technological capability that results into successful exploitation of opportunities (Benner & Tushman, 2003). Additionally, high technological capability increase receptiveness to external knowledge flows that are crucial to innovation (Berkhout, Hartmann & Trott, 2010). Superior technological capability involves high-quality technical know-how that increases exploitative learning that leads to recombination of existing technologies to fulfil new uses (RosenKopf & Nerkar, 2001). The strength of technological capability heavily depends on how technical knowledge is bundled together and distributed throughout the firm. This is tricky as knowledge involved in technologies are often tacit making it difficult to codify and transfer (Wang et al., 2006).

Once a strong technical base is formed, building new solutions can be achieved through harnessing emerging technologies, which hold the potential to create and transform industries fostering innovation (Day & Schoemaker, 2000). Emerging technologies are relatively fast growing and radical novel technologies that hold potential for a considerable impact on industries, but these impacts are most accountable for the future, but as they are emergent, their long-term impacts are uncertain and ambiguous (Rotolo, 2016). These emerging technologies can foster both radical and incremental innovations (Day & Schoemaker, 2000). Their contribution is valuable because the life-cycles of technological products are relatively short (Day & Schoemaker, 2000) evidently speeding up the rate of innovation. Consequently, the value of technology vanishes at a much faster rate (Srinivasan, 2008). To tackle this, a level of flexibility is needed foundational to the dynamic capabilities view (Teece, 2007). Emerging technologies offer such flexibility as relatively less is known their durability is longer and can contribute to new innovations (Srinivasan, 2008). Given the novelty of these technologies, the life-cycle rate is likely to be longer than existing technologies (Rotolo, 2016). Superiority in emerging technologies may give firms a significant advantage by being first-movers and lead to sustainable business performance (Suarez & Lanzolla, 2007; Rotolo, 2016). Though, the successful deployment of such technologies strongly depend on existing technological capability given the novelty, a higher technological capability is preferable as it can better accumulate and assimilate the related technical knowledge (Halaç, 2015). Moreover, DO is built around a technological knowledge base, a failure to recognize and harness emerging technologies could lead to an obsolete business model (Srinivasan, 2008).

(15)

2.7 Transforming and Renewal

To transform and foster continuous renewal, the knowledge accumulated in the first two capabilities is centric to a firm’s success and its integration across the firm is crucial (Teece, 2007). This can be done through proper knowledge management, providing incentives to create and create mechanisms that facilitate knowledge transfer (Teece, 2007). Actively integrating and diffusing knowledge across the firm allows to keep taps on past performance while ensuring future performance (Teece, 2007). Transforming in digitally oriented firms is reflected in cross-functional integration. This is because the solid presence of technology influences all functions regardless and thus integration is required (Joshi, 1998). Additionally, digitally oriented firms facilitate knowledge transfer mechanisms through databases that help store information that is beneficial for transforming.

Cross-Functional Integration

While former capabilities are focused on shorter term emphasis, the latter is focused on transforming capabilities that foster long-term sustainable performance (Teece, 2007). To achieve the latter in digitally oriented firms, cross-functional integration is used. This concept refers to combining functional activities within an organization, bridging boundaries and allowing information to flow throughout a firm. Effective cross-functional integration can enhance learning, knowledge sharing, solve complex problems and encourage change (Meyer, 1993). It allows firms to be more flexible and agile to any changes needed to respond accordingly that can lead to continuously renewal (Meyer, 1993). Argued by several researchers cross-functional integration plays a key role in the success of innovations as innovation is a multi-facet concept (Sherman, Souder & Jenssen, 2000; Sherman, Berkowitz & Souder, 2005; Troy, Hirunyawipada & Paswan, 2008). Some prime examples of cross-functional integration are: R&D/manufacturing, R&D/supplier and R&D/marketing integrations (Sherman, Souder & Jenssen, 2000). Foremost, cross-functional integration requires management of knowledge between departments and how that knowledge is distributed and utilized (Troy, Hirunyawipada & Paswan, 2008).

(16)

balances between efficiency and effectiveness with very different actors (Song, Thieme & Xie, 1998). Achieving a mutual understanding, a shared standard and aligned objectives is necessary but tricky (Song, Thieme & Xie, 1998). Fundamental differences are challenging to overcome. For example, R&D employees tend to be focused on long-term performance orientation and radical innovations whereas marketing employees have short-term orientation that includes more incremental innovations (Sherman, Souder & Jenssen, 2000). They also base their decisions on different sources of information, R&D is mainly involved in scientific information whilst marketing focuses on market information (Sherman, Souder & Jenssen, 2000). Therefore, successful integration requires careful coordination and managerial attention that imposes coordination costs on the firm (Troy et al,2008; Song, Thieme & Xie, 1998). This can damper the positive effects of cross-functional integration on innovation, conflicts may disturb the process of innovation, time-consuming and managers need specific training to manage such teams (Song, Thieme & Xie, 1998; Scherman, Souder & Jenssen, 2000).

On the other hand, in a digital environment the effect of cross-functional integration lies somewhat different. In digitally oriented firms, technology is the baseline and their presence is complex, interconnected and influences all organizational functions (Joshi, 1998). In the context of DO, cross-functional integration is demanded as it influences all department. Cross-cross-functional integration in digitally oriented firms may be favourable, yet achieving this may be even more complicated in comparison to non-digitally oriented firms. Effective integration requires transfers of knowledge flows, which in the case of DO is often tacit knowledge and difficult to codify (Wang et al., 2006). The technologies used in DO generate big data sets that are complex and difficult to understand (Leeflang et al., 2014). Achieving cross-functional integration in such conditions, requires even better coordination between departments to reach a mutual understanding and shared standard (Brettel, Heineman, Engelen & Neubauer, 2011). This can be done through the incorporation of knowledge management mechanisms. Effective knowledge management allows firms to collect all knowledge in a way that it is able to recode, retrieve and review information of past performances to ensure future performance (Sherman, Berkowitz & Souder, 2005). In digitally oriented firms, these mechanisms can be introduced by means of technological databases, these offer transparency, easy access, real-time adaption and make sure the knowledge is available to the whole firm enhancing flexibility and response rate (Shaw, Subramanian, Tan & Welge, 2001). These mechanism facilitate cross-functional integration that is demanded for digitally oriented firms (Mangold & Faulds, 2009). Therefore, cross-functional integration can be used to effectively diffuse knowledge throughout the firm by means of knowledge management mechanism that is needed to transform (Teece, 2007).

(17)

2.5 The Moderating Role of Chief Digital Officer

Further building on the dynamic capabilities of Teece (2007), it is argued that for firms to successfully transform, dynamic capabilities need to be used in conjunction with strategy as that is a core concept in business models (Teece, 2018). This leads to the assumption that the above mentioned elements of digital orientation need to be paired with a fitted strategy (Rumelt, 2011). This too fits the strategic nature in which digital orientation is categorized. To achieve this alignment, this study will focus on the presence of the presence of a Chief Digital Officer that will ultimately strengthen the relationship between DO and innovation performance.

A Chief Digital Officer (CDO) is an executive role that is responsible for the transformation of operational processes to digital processes and generating new business opportunities presented by the adoption of digital technologies (Singh, Klarner & Hess, 2020). The strategic imperative is undeniably linked to digital orientation and thus a new executive position; CDO is introduced (Singh & Hess, 2017). This role has grown tremendously as more firms realise that the fit between business and technology becomes vital (Sigh, Klarner & Hess, 2020). Given the novelty on this position, the true contribution of a CDO is difficult to determine (Saldanha, 2019). This is seen in recent studies, where a lot of confusion exists, due to limited understanding CDOs are often interchanged with other executive functions such as Chief Information Officer (CIO) and Chief Data Officer for example (Singh & Hess, 2017). Yet in practice these are differ in their responsibilities, CDOs are responsible for digital initiatives, CIOs are focused on strategic deployment of IT resources and Chief Data Officers are focused on data management and analysis (Grossman & Rich, 2016). Where CIOs tend to focus on established norms and routines, CDOs are often appointed to pursue digital orientation activities and thus driving change across the firm (Singh & Hess, 2017). Especially in digitally oriented firms, the presence of a CDO is a strong indicator of DO and an important and formal steps firms need to take when striving for DO making them the front leaders (Peltier, Schibrowsky & Zhao, 2009; Quinton et al., 2017; Singh, Klarner & Hess, 2020). Ultimately CDOs’ influence is much more prominent when adopting a DO in comparison to other executive roles or strategic orientations (Singh & Hess, 2017). CDOs possess expertise and experience required to guide strategic efforts in achieving DO (Singh & Hess, 2017; Quinton et al., 2017). Regardless, as business and technology becomes more tangled, it is even more difficult to determine the added value a CDO can offer to firms (Saldanha, 2019). Their roles and responsibilities become less clear in an era where technology is embedded in the very core of firms (Saldanha, 2019). Traditionally, CDOs were appointed to link business and technology, yet with more firms turning digital, there is need for a better understanding on the role CDOs and their contribution in a context where business and technology are no longer separate pillars (Saldhana, 2019).

(18)

presence or absence of such managers have on firm performance (Nath & Mahajan, 2008; Taylor & Vithayathil, 2018; Tumbas, Berente & vom Brocke, 2018; Singh, Klarne & Hess, 2020). To illustrate, Nath & Mahajan (2008) investigated how Chief Marketing Officers’ (CMO) positively influences innovation. This was because the presence of a CMO helped translate complex market insights whilst overseeing marketing resources ensuring a better alignment with customer needs (Nath & Mahajan, 2008). Contrary, Taylor & Vithayathil (2018) argued that having a Chief Information Officer surpasses the presence of a CMO. Their rationale being, that technology changes faster than ever, posing challenges on firms to adopt to new technologies that are tackled by a Chief Information Officer (Taylor & Vithayathil, 2018). A powerful technological leader leads to a sustainable firm performance in comparison to a powerful marketing leader, as technology becomes exponentially important (Taylor & Vithayathil, 2018).

When exploring the specific condition of a CDO, it can be argued that they positively affect innovation activities. Given that CDOs are not looking to conform to established routines, rather using digital expertise to explore new paths (Tumbas, Berent, vom Brocke, 2018). Additionally, CDOs emphasize the importance of strategy in all their activities opposed to functional IT managers, who are more involved in operational integration (Tumbas, Berent, vom Brocke, 2018). The core contribution of any CDO was the creation of new revenue streams, interconnected with digital expertise and innovations (Tumbas, Berent, vom Brocke, 2018). CDOs that were appointed usually had experience in start-ups and firms operating in highly dynamic environments, they also portrait risk-taking characteristics opposite to risk-averse behaviours that CIOs portray (Tumbas, Berent, vom Brocke, 2018), this is believed to have a positive effect on innovation. Moreover, CDOs are focused on utilizing their digital expertise to set emphasis on customers’ to provide a high quality end product as well as use customers to drive further insights (Tumbas, Berent, vom Brocke, 2018). Noted by Wiesböck & Hess (2018), firms that are moving towards DO require adaption of the entire business model. Thus there are two things that need to happen at executive level; (1) executive managers create a digital roadmap upon which digital innovation activities are derived and delivered to the customers and (2) executive managers need to foster the search for new digital products and services (Wiesböck & Hess, 2018). This transformation can only take place if the required expertise and experience is included in strategic decision-making as argued before (Wiesböck & Hess, 2018).

Hence the following hypotheses are proposed:

(19)

H4b: The positive relationship between technological capability and innovation performance is moderated by the presence of a Chief Digital Officer such that the presence of a Chief Digital Officer positively influences the relationship between technological capability and innovation performance.

H4c: The positive relationship between cross-functional integration and innovation performance is moderated by the presence of a Chief Digital Officer such that the presence of a Chief Digital Officer positively influences the relationship between cross-functional integration and innovation performance.

The conceptual model illustrated in Figure 1 present the proposed hypotheses in this study.

Figure 1. Conceptual Model

(20)

3. Methodology

3.1 Data Collection

The data that was collected for the purpose of this study was mostly done by means of secondary data. The primary dataset was retrieved from Labor Insights which is provided to us by Burning Glass Technologies. Labor Insights is a database that extracts information on the labor market and provide an analytical insights on employment vacancies. The database consist of job postings collected through a technique called web-scraping, meaning that jobs are scraped from major job boards in the US. This results into a rather complete overview of all job postings listed on labor websites throughout he US. Once accessed the database, you can collect data on firms based on various identifiers. After this, you will be able to export the datafile into an excel sheet that will visualize the job postings and the skills demanded for those postings. This results into a division of job postings in main clusters and sub clusters. For this study, the Labor Inisght database was primarily used to collect data. While the overall databases generates main clusters and sub-clusters, in this study I chosen focus on sub-clusters that reflect all three elements of DO. To do so, three main clusters were used to identify the selected sub-clusters. The main clusters are Marketing And Public Relations, Information Technology And Analysis. A representation of these can be found in Table 1. I have chosen to focus on the companies and job postings that are listed on the Standard and Poor’s 500 firms over a time period of 10 years (from 2010 to 2019). This dataset was then extended by collecting financial data from Compustat retrieving innovation performance and the chosen control variables. For the moderator, I used the Boardex database which contains information on boards of firms. Boardex allows you access to data that discloses information on board compositions, board roles, divisions of gender and so forth. To merge the datasets together, company ticker and year were used as identifiers. This lead to a final sample of 605 firms and 5,432 observations.

Table 1. Skill Clusters Example

Main Clusters Sub Clusters Example of Skills

Marketing and Public Relations

Social Media Online Marketing

Instagram, Social Media Platforms, Social Media Trends Search Engine Optimization (SEO), Digital Marketing,

Keywords Research Information

Technology

Artificial Intelligence Augmented Reality

Virtual Agents, Artificial Intelligence, AI ChatBot Oculus, Virtual Reality

(21)

3.2 Measurements Dependent Variable

(22)

Independent Variables

The following paragraphs will explain the measurements of the elements of digital orientation. Due to the skewness of those variables, they all have been log transformed to counteract this. The literature used to identify the measurements can be found in Table 2. It is important to note that the measures for the independent variables are proxy measures. Proxy measures are indirect measures of the desired outcome and are highly correlated to a specific outcome. This is mainly used when direct measures of a specific outcome are unobservable or unavailable. In this study the direct measurements of these capabilities match those unavailability, hence these are metrics are proxy measures.

Digital Marketing Capability: This capability is used to reflect how digitally oriented firms sense and shape opportunities. The sub-clusters include: Customer Relationship Management, Market Analysis, Social Media, Online Marketing And Web Analytics. This provides firm with the ability to look from an outside-in perspective, drawing rich customer data that will help detect opportunities that firms can exploit. Customer relationship management can help firms improve existing relationships and form new ones through the use of technologies (Wind & Rangwaswamy, 2001). These relationships are improved through the use of customer data retrieved from digital technologies and also present new opportunities by doing market analysis (Leeflang et al., 2014). Additionally, social media and online marketing can generate rich customer data that will also help marketeers sense opportunities (Royle & Laing, 2014; Leeflang et al., 2014). Lastly, web analytics is part of digital orientation as it analyses data retrieved from online platforms that will help detect patterns that can be used for discovering new opportunities (Stone & Woodcock, 2013).

(23)

on to cloud computing, which is a model that enables on-demand, convenient network access to a shared pool of resources (Garrison, Kim & Wakefeld, 2012). This has inherently changed the way organizations manage computing resources and allows firms to lower their IT expenditures and operation cost, as these cloud services are low in cost and they can dispose outdated equipment (Garriosn, Kim & Wakefield, 2012). Making use of cloud computing allows firms to harness additional benefits as it also increases agility (Garrison, Kim & Wakefield, 2012) and thus better responding to new opportunities. Finally, Internet of Things (IoT), which is defined as a technological system that interrelates computing devices that are able to transfer data through a network without any human interaction (Li, Xu & Zhao, 2015). IoT can be used to exploit opportunities as it transforms the speed, pattern and information exchange between two actors, fostering IoT skills can help firms be superior and achieve a sustainable competitive advantage (Li, Xu & Zhao, 2015). IoT is effective when firms have information sharing mechanisms that can be used to strengthen IoT capabilities (Li, Xu & Zhao, 2015).

(24)

Table 2. Measurements of Variables

Variable Sub-Clusters Scholars

Digital Marketing Capability Customer-relationship Management Market Analysis Social Media Online Marketing Web Analytics

(Wind & Rangaswamy, 2001) (Leeflang et al., 2014) (Royle & Laing, 2014) (Leeflang et al., 2014) (Stone & Woodcock, 2013)

Technological Capability Artificial Intelligence Augmented Reality Big Data Cloud Computing Internet of Things (Ransbotham et al., 2017) (Kim et al., 2015) (Wamba et al., 2017)

(Garrison, Kim & Wakefield, 2012) (Li et al., 2015) Cross-Functional Integration Information Security Data Warehousing Business Intelligence Data Management Database Management

(Anttila, Kajava & Varonen, 2004) (Nelson, 2001)

(Meredith et al., 2012)

(Sherman, Berkowitz & Souder, 2005) (Joshi, 1998)

Moderator Variable

Presence of Chief Digital Officer: As explained in the theoretical background, there is an assumption that when firms are digitally oriented and have a higher level IT manager, innovation performance will increase. As mentioned by Teece (2018), top management’s skills and knowledge are vital to sustainable performance outcomes and therefore, is of importance when looking at the broader picture of digital orientation and its impact on innovation. I have chosen to collect this variable and apply dummy coding. This is because it is important to test the condition: if the presence of a Chief Digital Officer exerts influence on the proposed relationship between digital orientation and innovation performance. Hence, firms that do have this position filled will be awarded with a “1”, in contrast to those firms that do not, which will be demonstrated with a “0”.

Control Variables

To control for potential confounding variables, I selected the following variables as controls. Additionally, similar to the independent variables, skewness seemed to be a problem, this too was counteracted through a log transform.

(25)

(Vahlne & Jonsson, 2017). Hence, firm size is included in my control variables, which is measured by transforming total assets of each firms per year.

R&D Intensity: In a study done by Bhattacharya & Bloch (2004), it is illustrated that R&D intensity can further induces innovation activity. A high R&D intensity is not a sole indicator for innovation performance, but can stimulate innovation by dedicating more resources opposed to lower R&D intense firms (Bhattacharya & Bloch, 2004; Sayrul & Incekara, 2015). Thus R&D intensity is assumed to hold influence over innovation performance and therefore included in my controls. To control for this, I have divided research and development expense over total assets for each firms and each year.

Industry Sector: Generally, innovation performance may be higher in high-tech sectors in comparison to low-tech sectors (Baesu, Albulsecu, Farkas & Drãghici, 2015). Thus, this is included in my controls by using dummy variables. The data set includes 5 industries; manufacturing, telecommunication, finance, service and retail. Dummies were created by awarding a “1” if a firm belongs to an industry and a “0” if it did not based on industry sector codes (SIC). Applying a dummy code entails: n-1, thereby excluding the financial sector as those appears to be least applicable in this study.

Year: Similar to the sector codes, I used year dummies to control for time specific effects. Therefore, we took value of “1” for a specific year and a “0” for a different year.

3.3 Analysis

(26)

lagged performance effect. Especially in panel data, autocorrelation can be an issue, which causes problems for the true estimation of the other variables (Achen, 2000). This presence of such autocorrelation can lead to a bias that may lead to wrong inferences (Achen, 2000). Hence, introducing a lagged variable in the regression model reduces this chance by capturing the dynamic effects that relate to panel data (Keele & Kelly, 2005).

4. Results

4.1 Descriptive Statistics and Correlations

Table 3 illustrates the descriptive statistics including means, standard deviations, minimum and maximum value for each variable construct. This summary consist of the final sample size of 605 firms alongside 5584 observations. There is a slight difference in observations as the control variables firm size and R&D intensity only include 5562 observations. Furthermore, skewedness seemed to be an issue and therefore all the data is log-transformed. Based on Table 3, it can be concluded that innovation performance is high in variation amongst firms based on the standard deviation (446.791). This means that there are significant differences in innovation performance between the sampled firms. While for the independent variables, the standard deviation is rather low which leads to the assumption that those capabilities are rather close between individual firms. It also is noted that the variance in the independent variables are all rather close (around 2.1) except for Chief Digital Officer whose variation is relatively low (0.339). But this may have been influenced because of the dummy coding used for this variable.

Secondly a Pearson correlation matrix was performed in which the correlation of each coefficient measures the strength and direction of the linear relationships between two given variables. To interpret these results the following conditions need to be taken into consideration, correlations can range from -1 to 1, a value of “1” indicates a perfect positive relationship and “0” indicates no relationship. This correlation matrix can help detect whether or not multicollinearity is a problem. Multicollinearity is a problem because one cannot be sure to trust p-values even though they are statistically relevant. A benchmark for detecting this issue is when there is a high correlation coefficient of 0.7 (Luger, Raisch & Schimmer, 2018). Table 4 illustrates the correlation matrix, based on those findings we can assume that multicollinearity is not problematic as none of them exceed the 0.7 benchmark.

(27)

Table 3. Descriptive Statistics

Variable N Mean S.D. Min. Max.

(1) Innovation Performance 5,584 237.344 446.791 1 1624 (2) Digital Marketing Capability 5,584 4.093 2.337 0 14.102 (3) Technological Capability 5,584 1.052 2.019 0 10.373 (4) Cross-Functional Integration 5,584 3.113 2.120 0 13 (5) Presence of Chief Digital Officer 5,584 0.132 0.339 0 1 (6) Firm Size 5,562 7.524 1.095 0.693 8.498 (7) R&D Intensity 5,562 3.135 2.736 0.693 7.626

Notes: All variables except variable 1 are log-transformed.

Table 4. Correlation Matrix

Variable (1) (2) (3) (4) (5) (6) (7) (1) Innovation Performance 1 (2) Digital Marketing Capability 0.077*** 1 (3) Technological Capability 0.169*** 0.302*** 1 (4) Cross-Functional Integration 0.067*** 0.456*** 0.397*** 1 (5) Presence of Chief Digital Officer 0.005 0.0109 0.018 0.003 1 (6) Firm Size 0.029** -0.002 -0.011 0.000 -0.014 1 (7) R&D Intensity 0.039*** -0.038*** 0.007 0.047*** 0.005 -0.156*** 1

(28)

4.2 Regression Results & Hypothesis Testing

Table 5 presents the results of two fixed effects model, one with the lagged dependent variable and one with the regular dependent variable. A total of six models are conducted to test the proposed hypotheses. The baseline model only includes the control variables and independent variable (model 1 and model 4), afterwards the independent variables are added (model 2 and model 5) and is completed by the adding the moderator and interaction effects (model 3 and model 6). The main model can be found in the first three models, which is the fixed effects model including the lagged dependent variable.

Table 5 displays the results of H1 (b = 9.561, p<0.05) and H2 (b = 28.67, p<0.01) showing the coefficients and their significance level. In this case, each additional unit of digital marketing capability is accompanied by a 10.10 increase in innovation performance. Similarly, each additional unit of technological capability contributes to a 16.96 increase in innovation performance. The results of the statistical test provide support for both hypotheses, suggesting that digital marketing capability and technological capability positively influence innovation performance. Regarding H3 (b = -0.559, p>0.1), a negative coefficient and a non-significant result was found. This was contrary to what was hypothesizes, cross-functional integration was expected to portray a positive relationship with innovation performance. However, the model did not provide statistical support.

Additionally, H4 included the moderator variable and the interaction effect with the independent variables. In regards to H4a (b = 14.02, p<0.1) a positive significant result is shown. This is in line with hypothesis 4a, thus providing statistical support. Meaning that the presence of a CDO strengthens the positive effect of having digital marketing capability with a value of 14.02 for each CDO. Followed by

(29)

Table 5. Regression Analysis Dependent Variable (Innovation Performance) Model 1 (t-1) Model 2 (t-1) Model 3 (t-1) Model 4 (t) Model 5 (t) Model 6 (t) Independent Variables

Digital Marketing Capability

Technological Capability

Cross-Functional Integration

Moderator Variable

Presence of Chief Digital Officer Interaction Effect DMC x CDO TC x CDO CFI x CDO 9.561** (4.361) 28.67*** (3.591) -0.559 (4.003) 7.827* (4.457) 29.46*** (3.769) -0.344 (8.599) -59.68* (35.24) 14.02* (7.507) -4.184 (7.926) -0.344 (8.599) 10.10** (4.107) 16.96*** (3.247) -0.650 (3.817) 9.821** (4.201) 17.00*** (3.613) -0.948 (3.959) -25.33 (32.78) 2.102 (7.118) -0.035 (7.613) 2.196 (8.214) Control Variables Firm Size R&D Intensity Year Dummies Sector Dummies Constant Number of Observations R-Squared 8.348 (6.074) 0.380 (2.978) YES YES 217.7*** (47.82) 5,432 0.005 7.575 (6.030) 1.839 (2.961) YES YES 127.2** (51.89) 5,432 0.020 7.429 (6.031) 1.908 (2.963) YES YES 136.1*** (52.10) 5,432 0.021 9.429 (5.787) 3.956 (2.824) YES YES 82.03* (45.57) 5,562 0.013 9.044 (5.769) 4.823* (2.820) YES YES 11.13 (49.45) 5,562 0.020 9.019 (5.772) 4.896* (2.526) YES YES 14.39 (49.66) 5,562 0.021

(30)

4.3 Robustness Check

A robustness check was done because of errors that can occurring due to the dynamic nature of the panel data set, meaning that the cause and effect relationship of underlying phenomena is dynamic over time, hence time-effect influences may rise over a given period (Ullah, Akhtar & Zaefarian, 2018). Panel data carries higher risks to autocorrelation and endogeneity. Autocorrelation refers to the degree of correlation between the values of the same variables across different observations in the corresponding dataset (Anderson, 1954). Autocorrelation compromises the independency of variables. Hence, you cannot estimate the correct values of parameters and can lead to a bias (Mukherjee & Laha, 2019). Endogeneity bias refers to the condition in which an explanatory variable correlates with an error term (Ullah, Akthar & Zaefarian, 2018). This can cause inconsistent estimates, which lead to wrong inferences and misleading conclusions that will affect the overall theoretical interpretations (Ullah, Akthar & Zaefarian, 2018). As endogeneity bias can occur from different origins, a variation of methods can be used to address them. In this study, the generalized method of moments model (GMM) can be used (Ullah, Akthar & Zaefarian, 2018). This is because some assumptions are fulfilled; 1) being that some regressors may be endogenously determined, 2) the nature of the relationship is dynamic, 3) the time periods in panel data is smaller than the number of observations (Roodman, 2009). The GMM model provides consistent results and takes into account different sources of endogeneity; unobserved heterogeneity, simultaneity and dynamic endogeneity (Ullah, Akthar & Zaefarian, 2018). Using a GMM model removes endogeneity bias as it transforms data internally. There are two options when performing a GMM model, first being first-difference transformation and secondly, second-order transformation (two-step GMM). The latter is preferred over the former as it prevents unnecessary data loss, hence for this study I will conduct a two-step GMM model (Ullah, Akthar & Zaefarian, 2018). In Table 6 the generated results of the GMM model can be found. Similar to the regression, two models were tested; one including the lagged dependent variable and a regular dependent variable.

The output of the GMM model includes 5432 observations distributed over 605 groups based on tickers and a time-period from 2010 to 2019. Firstly, in contrary to the fixed effects model (t-1), no statistical support is found for H1(b = 45.69, p>0.1), thus not providing statistical support for the findings I have found in former regression model. Secondly, the GMM output supports findings of H2, showing a positive significant coefficient (b = 69.62, p<0.05). Contrary to the fixed effects model (t-1), H3 (b = -66.13, p<0.1) displays a significant negative effect, which was not expected. Thus providing statistical evidence that cross-functional integration negatively affects innovation performance. Looking at the presence of a CDO, in compliance with the regression is partly supported. Where H4a (b = 308, p<0.01) provides a positive significant effect and thus supportive. In contrary Hb (b = -48.41, p>0.1) and H4c

(31)

lagged variable still is a problem. The overall takeaway of the GMM model indicates that my findings are only partially supported and thus not entirely robust.

Table 6. Two-step GMM Model

Variable Two-step GMM Model

(t-1) Two-step GMM Model (t) Independent Variables Digital Marketing Capability Technological Capability Cross-Functional Integration Moderator Variable

Presence of Chief Digital Officer Interaction Effect DMC x CDO TC x CDO CFI x CDO 45.69 (29.53) 69.62** (21.75) -66.13* (34.55) -2.175*** 308.4*** (102.9) -48.41 (102.2) 29.53 (133.6) 3.790 (21.10) 66.36** (15.19) -37.20 (24.96) -111.0 (379.9) 73.33 (62.72) -77.14 (75.62) -31.92 (91.15) Control Variables Firm Size R&D Intensity Year Dummies Sector Dummies Constant Number of Observations F Statistic Groups/Instruments AR(2) Sargan Test Hansen Test 4.195 (17.79) 7.836 (7.776) YES YES -113.4 (308.8) 5,432 4.03 605/83 0.016 0.00 0.00 10.02 (14.29) 6.017 (6.597) YES YES 71.77 (238.7) 5,562 3.9 608/83 0.00 0.00 0.00

(32)

5. Discussion

5.1 Theoretical Implications

In context of today’s competitive environment, the fast-pace of in which external market factors are changing rapidly in conjunction with the increasing importance of digital technologies, this study has built on a new strategic direction to keep up, namely digital orientation (Quinton et al., 2017). The overarching concept of digital orientation was defined as a deliberate strategic position of firm and taking advantages of opportunities presented by digital technologies and the will to acquire a substantial technological knowledge base that is used for new product development. Since these type of firms naturally operate in environments characterized by short product-life cycles, uncertainty, fast pace of innovation and rapid change in customer demands (Quinton et al., 2017; Horthina et al., 2011; Zhou et al, 2005) , a high level of flexibility is needed to stay in the game (Teece, 2007). Specifically, in digitally oriented firms the core is built around technologies, but these are quickly evolving and their value vanishes rapidly (Srinivasan, 2008). Such conditions make the dynamic capabilities perspective appropriate as it provides firms with the ability to reconfigure internal and external competences to address those rapid changes (Teece, Pisano & Shuen, 1997). Hence three elements were chosen to reflect DO and their consequent impact on innovation.

The definition of DO states that firms use digital technologies to sense opportunities, this is done through digital marketing capability which allows firms to form meaningful and deep relationships with customers via web-based communication channels, whilst leveraging customer data retrieved from those channels to create a deep customer understanding (Moncrief & Cravens, 1999; Wind & Rangaswamy, 2001; Stone & Woodcock, 2013; Chatterjee, 2001). This is believed to be beneficial for innovation as it allows firms to better align new products with what customers desire (Weerawardena, 2003; Merrilees, Rundle-Thiele & Lye, 2010; Mariadoss, Tansjuhaj & Mouri, 2011). As their knowledge on customers increases, there is room to detect patterns and guide firms into new target markets (Royle & Laing, 2014). The required-skill set for such marketing activities are more likely to align with digitally oriented firms opposed to other orientations as the presence of technology and thus their technical ability is the core way of working (Quinton et al., 2017). This is in line with the findings of the regression and empirically supporting the reasoning that digital marketing capability has positive impacts on innovation in digitally oriented firms.

(33)

push digitally oriented firms to be superior as other technologies may become obsolete (Srinivasan, 2008) due to the fast changes. Digitally oriented firms display high levels of technological capability by the presence of a strong technical knowledge base that allows them to exploit opportunities and drive innovation by harnessing emerging technologies that will push them at the front of the market (Rotolo, 2016). In coherence with previous literature, empirical support was found for the notion that firms that adopt a DO with higher levels of technological capability will positively affect innovation performance.

In DO, the integration of this knowledge throughout the firm is essential to building transformative capabilities (Teece, 2007). In digitally oriented firms technology is the core knowledge source and its presence is complex and interconnected influencing all organizational functions (Gebdhardt et al., 2006). Therefore, in digitally oriented firms cross-functional integration is not optional but a must. This is beneficial to innovation as it increases mutual understanding, achieves consistency in development stages and enhances flexibility by pooling resources and skills in such a way that resources are effectively utilizes (Sethi, 2000; Troy et al., 2008). For DO, this can be achieved through cross-functional integration by using knowledge transfer mechanisms such as databases. Storing information is particularly useful in DO as the information generated in such orientation leads to big data sets that can help recode, retrieve and review information of past performances that ensure future performance (Sherman, Berkowitz & Souder, 2005). Accompanied by their technical knowledge base, digitally oriented firms are able to incorporate technical sharing mechanisms that offer transparency, easy access and real-time adoption that make knowledge diffusion effective and thus fostering continuous renewal (Shaw, Subramanian, Tan & Welge, 2001). Contrary, there was no empirical support found on this positive effect of cross-functional integration. A possible explanation in the literature could be the knowledge needing to be integrated is often complex, difficult to codify and tacit in DO (Wang et al., 2006). In such conditions, even better coordination is needed to reach mutual understanding and shared standard (Gebdhardt et al., 2006). However, achieving this is rather difficult in cross-functional integration and can lead to conflicts because of the fundamental differences between functions and the tacit nature of technology (Song, Thieme & Xie, 1998; Sherman, Souder & Jenssen, 200).

(34)

Their risk-taking behavior puts emphasis on the creation of new revenue streams through digital technologies and innovations (Tumbas, Berent & Vom Brocke, 2018). Appointing a CDO is an important step towards realizing DO (Singh & Hess, 2017; Quinton et al., 2017). Transforming to DO, involvement at strategic level is needed to create a digital roadmap steering operational activities while also searching for new technologies to invest in that will leverage the existing knowledge base and consequent innovations (Chanis, Meyers & Hess, 2018). The empirical findings partially support this notion, where a CDO strengthens the positive relationship between digital marketing capability and innovation yet no empirical findings were found for the other two elements. This positive effect can be explained through the gap existing in what marketeers should do and can do (Day, 2011). Technical skills are required to perform digital marketing capabilities and their input becomes increasingly important in strategic-decision making (Royle & Laing, 2014). This link is achieved when a digital strategist is present that possess technical knowledge that helps identify which technologies complement current marketing activities and invest in such technologies (Crush, 2011). The possible explanation for the negative result of technological capability and cross-functional integration is that CDOs are useful is due the lack of specific characteristics that relate to the role of a CDO, what they contribute and how they do that (Saldanha, 2019). Since this concept has not received much attention, not much knowledge is available and therefore, a complete picture or exact contribution cannot be realized.

Important to note is that I only have found partial support for my hypothesis, hence making my findings not entirely robust. This means that I cannot be entirely sure on the conclusion of this study. Two out of three elements portrayed empirical support and one did not. This makes it hard to come to a certain conclusion on whether or not digital orientation positively influences innovation performance. The study does show support that digital marketing capability and technological capability are useful elements to achieve DO, but cross-functional integration may not be as useful. The true contribution of a CDO was also only partially supported hence it cannot be stated confidently that this presence has a moderating role.

5.2 Managerial Implications

(35)

in the transition phase but due to our limited understanding it is difficult to determine how such a role should be shaped and managed in order to have a positive effect.

5.3 Limitations and Future Research

(36)

6. References

Achen, C. H. (2000, July). Why lagged dependent variables can suppress the explanatory power of other independent variables. In annual meeting of the political methodology section of the American political science association,

UCLA (Vol. 20, No. 22, pp. 07-2000).

Afuah, A. (2002). Mapping technological capabilities into product markets and competitive advantage: the case of cholesterol drugs. Strategic Management Journal, 23(2), 171-179.

Al-Ansari, Y., Altalib, M., & Sardoh, M. (2013). Technology orientation, innovation and business performance: A study of Dubai SMEs. The International Technology Management Review, 3(1), 1-11.

Anderson, T. W., & Darling, D. A. (1954). A test of goodness of fit. Journal of the American statistical

association, 49(268), 765-769.

Andries, P., & Faems, D. (2013). Patenting activities and firm performance: does firm size matter?. Journal of Product

Innovation Management, 30(6), 1089-1098.

Anttila, J., Kajava, J., & Varonen, R. (2004, September). Balanced integration of information security into business management. In Proceedings. 30th Euromicro Conference, 2004. (pp. 558-564). IEEE.

Atuahene-Gima, K. (1996). Market orientation and innovation. Journal of Business Research, 35(2), 93-103.

Baker, W. E., & Sinkula, J. M. (2009). The complementary effects of market orientation and entrepreneurial orientation on profitability in small businesses. Journal of Small Business Management, 47(4), 443-464.

Barney, J. B. (1986). Organizational culture: can it be a source of sustained competitive advantage?. Academy of

Management Review, 11(3), 656-665.

Berkhout, G., Hartmann, D., & Trott, P. (2010). Connecting technological capabilities with market needs using a cyclic innovation model. R&d Management, 40(5), 474-490.

Bhattacharya, M., & Bloch, H. (2004). Determinants of innovation. Small Business Economics, 22(2), 155-162. Borenstein, M., Hedges, L., & Rothstein, H. (2007). Meta-analysis: Fixed effect vs. random effects. Meta-analysis. com. Brettel, M., Heinemann, F., Engelen, A., & Neubauer, S. (2011). Cross-functional integration of R&D, marketing, and manufacturing in radical and incremental product innovations and its effects on project effectiveness and efficiency.

Journal of Product Innovation Management, 28(2), 251-269.

Breznik, L., & Hisrich, R. D. (2014). Dynamic capabilities vs. innovation capability: are they related?. Journal of Small business and Enterprise Development.

Byrne, D., Oliner, S. D., & Sichel, D. E. (2017). Prices of high-tech products, mismeasurement, and the pace of innovation. Business Economics, 52(2), 103-113.

Calantone, R. J., Cavusgil, S. T., & Zhao, Y. (2002). Learning orientation, firm innovation capability, and firm performance. Industrial Marketing Management, 31(6), 515-524.

Chandler, A. D. 1990. Scale and scope. Cambridge, MA: Harvard University Press. Chatterjee, P. (2001). Online reviews: do consumers use them?

Crush, P. (2011). Filling the digital skills gap. Marketing, 13 July, 33–34.

Clark, T. S., & Linzer, D. A. (2015). Should I use fixed or random effects. Political Science Research and Methods, 3(2), 399-408.

Cooper, R. G. (1994). Third-generation new product processes. Journal of Product Innovation Management: an

international publication of the product development & management association, 11(1), 3-14.

Cui, A. S., & Wu, F. (2016). Utilizing customer knowledge in innovation: antecedents and impact of customer involvement on new product performance. Journal of the Academy of Marketing Science, 44(4), 516-538.

An Empirical Analysis on Digital Orientation: Exploring the Relationship between Digital Orientation and Innovation Performance and the Moderating Role of the Presence of a Chief Digital Officer