Research Master in Economics and Business
Digital transformation of firms: A (r)evolutionary
perspective and empirical test
Student: Nicolai Fabian (s3440001)
Supervisor: Prof. Dr. P.C. Verhoef & Dr. Q. Dong
2
Abstract: Although digital transformation is widely considered as imperative for companies to stay relevant in a vastly changing environment, many firms fail in digitally transforming their business. To better understand how digital transformation may influence firm performance and, more importantly, under what conditions such influence is more beneficial or detrimental, this paper is among the first to quantify the impact of digital transformation on firm performance. By taking a (r)evolutionary perspective, we systematically theorize and measure digital transformation and examine its impact of firms’ digital performance. Furthermore, we show how openness to ideas facilitates the impact of digital transformation on digital performance and how radicalness of changes makes this impact risky and weaker. We empirically test our research model by using survey data from 201 firms in the Northern Netherlands. We find that digital transformation, on average, has a positive impact on firm digital performance and that this relationship is strengthened when firms are more open to new ideas but weakened if firms are pushing new ideas in a radical manner.
Keywords: digital transformation, firm digital performance, openness to ideas, radicalness of changes, evolutionary theory
INTRODUCTION
3
digital technology and to digitally transform is a facilitator of competitive advantage (Verhoef et al., 2019).
Companies have realized over the past years that technology has the potential to disrupt their businesses (Kane, Palmer, Philips, Kiron, & Buckley, 2015) and thus put digital transformation among the top priorities for future success (Capgemini Research Institute, 2018; Goasduff, 2018). While companies have realized that digital transformation must be a top priority, their efforts are not always successful. Research found that up to 80% of (digital) transformation efforts fail and that billions of dollars are lost each year due to bad transformation efforts (Tabrizi, Lam, Girard, & Irvin, 2019; Zobell, 2018). Wrongfully executed transformation efforts are assumed to be among the biggest risk factors for companies in 2018/2019 (Sun, 2018; McKendrick, 2019). In consequence, knowing how to digitally transform and how to deal with the implications of digital transformation becomes crucial not only for gaining a competitive advantage but also for long term financial health of the company (Sebastian et al., 2017).
4
transformation, (2) by examining its impact on firm digital performance, and (3) by identifying key contingency factors shaping the impact of digital transformation and firm digital performance.
We posit that digital transformation is formed by (1) the digital capabilities as the firm ability to integrate, build, and reconfigure internal and external competencies to utilize digital technology in a rapidly changing environment (Teece, 2010) as well as (2) the change in business model defined as the change in how the firm creates and delivers values to its customers, and then converts payments into profits (Li, 2018). Both first-order elements have been found to be important drivers of digital transformation (e.g. Li, 2018; Liu et al., 2011; Li et al., 2017). Moreover, we have theorized that digital transformation for established firms can be reached by an incremental step-by-step approach that builds on learning new ways to utilize technology (Verhoef et al., 2019). For this study, we build on this view by relying on evolutionary theory and its assumptions of the continuance of behavior and path dependency (Nelson & Winter, 1982; 2002). We use an evolutionary perspective to assess the contingencies of digital transformation. First, we posit that openness to ideas, which is defined as the firm’s ability to be open towards new ideas and ways of working (Soto et al., 2011; Venkatesh & Bala, 2012) facilitates the development of new capabilities and (digital) business models and thus digital transformation. Moreover, we posit that radicalness of changes, defined as the advance in novel and groundbreaking ideas, technologies and ways of working to the existing business activities that are faster than the actual rate of progress (Gatignon et al., 2002; Lyytinen & Newman, 2008; Luo et al., 2012), inhibits digital transformation because it contradicts path dependency and continuance of behavior (Nelson & Winter, 2002).
5
helps us to understand why some transformation efforts succeed while most others fail. Therefore, our study provides a new way to theorize about the mechanisms of digital transformation and how it affects firms.
The rest of the paper is organized as follows. First, we discuss digital transformation research and define the key constructs based on (partly our own) previous research. Then, we provide the theoretical background to our conceptual model followed by the development of hypothesis which illustrates the relationship between digital transformation and firm digital performance. Third, we discuss our methodology. Subsequently, we discuss our findings. Finally, we conclude and discuss the potential for future research based on our findings.
LITERATURE REVIEW1
Digital transformation by nature exists at the intersection of different literature streams and should be studied with a multidisciplinary perspective (e.g. Loebbecke & Picot, 2015; Verhoef et al., 2019). We posit that to understand digital transformation and its implications, it's not enough to review one literature stream but scholars must engage in a cross-discipline exchange of knowledge (Verhoef et al., 2019). Thus, we used a scoping review approach (Paré, Trudel, Jaana, & Kitsiou, 2015) to systematically search relevant contributions to the field of digital transformation from January 2000 till February 2018 in information systems, marketing, strategic management, and innovation management both in the respective top journals of each field as well as in the web of science database(Verhoef et al., 2019). We browsed nearly 4.000 papers, read over 1000 abstracts, selected nearly 200 papers and derived a final sample of 84 papers. Through thematic analysis and open coding (Boyatzis, 1998), we derived the three phases of digital change and their respective conceptual definitions.
Reviewing the literature showed us that each of the included literature streams accounts for a (complimentary) area of digital transformation at firm-level. In the IS field, we found a strong focus on the adoption of technology and subsequent changes in business (Agarwal, Gao, DesRoches, & Jha, 2010; Frishammar, Cenamor, Cavalli-Björkman, Hernell, & Carlsson, 2018; Li et al., 2018; Lucas et al., 2013). In the innovation management literature, the focus is on the possibilities of new technologies (e.g. 3D printing) but also on new product development (Aubert-Tarby, Escobar, & Rayna, 2018; Kolloch & Dellermann,
1 The literature review section, the expressed thoughts as well as Table 1 are based on our own paper (Verhoef et al., 2019),
6
2018; Li, 2018; Rindfleisch, O’Hern, & Sachdev, 2017). In the marketing field, the focus lies on the value creation as a consequence of digital transformation (Pagani & Pardo, 2017; Vendrell-Herrero, Bustinza, Parry, & Georgantzis, 2017), platforms (Ramaswamy & Ozcan, 2016) as well as advertising and social media in general (Kannan & Li, 2017). Lastly, the strategic management field has a focus on business models and competitive advantage (Kane et al., 2015; Osterwalder & Pigneur, 2010) as well as the consequences for firm performance (Nagaraj, 2018). Thus, each of those fields has a distinct focus such as technology adoption (IS), value creation (marketing/innovation) or the development of business models (strategic management), which together forms a more complete picture of digital transformation.
Three phases of digital change
We distinguish between three phases of digital change: digitization, digitalization, and digital transformation (Verhoef et al., 2019). They describe the utilization of digital technology within companies, but their scope, as well as goals, differ. It is to mention that across and within literature streams these terms are partly used interchangeably. To avoid confusion and lay the ground for the conceptual and operational development in our paper, we will summarize what we mean with each term. Additionally, we will clarify how those stages are connected and what are the strategic imperatives of each stage (see Table 1).
The first term digitization refers to the encoding of information in digital format such that tasks can be performed by the use of digital artefacts (Dougherty & Dunne, 2012). Other scholars use the term to describe the encoding of analog information to digital format (Yoo et al., 2010), the conversion of analog to digital information (Loebbecke & Picot, 2015), or the change of analog to digital information (Tan & Pan, 2003). Digitization is commonly used at the task level (Li, Nucciarelli, Roden, & Graham, 2016; Sebastian et al., 2017) and describes for example the addition of IT to existing tasks to facilitate cost-effective resource configurations (Lai, Wong, & Cheng, 2010) and is an enabler of more efficient operations (Vendrell-Herrero et al., 2017). Thus, digitization does not affect the value-creating activities of the firm but only affects limited tasks within a process, while the process itself remains the same (Verhoef et al., 2019). It requires digital assets to store new information and can be measured with “classic” metrics such as ROI or ROE (Table 1). We refer to digitization “as the action in which analog information is converted to digital information for cost-saving purposes” (Verhoef et al., 2019).
7
online communication channels (e.g. engagement platforms), which change the process of firm-customer interaction (Ramaswamy & Ozcan, 2016). Those changes lead to the organization of new socio-technical structures, which were not possible without the availability of digital technology (Dougherty & Dunne, 2012). Digitalization relies on the use of IT, or in broader terms digital technology, as an enabler of new possibilities and to change existing processes (Leviäkangas, 2016). Thus, in digitalization new digital capabilities (Table 1) are key to use digital technology to enhance processes within a form and/or the creation of new processes by connecting formerly unconnected resources (Pagani & Pardo, 2017). Thus, digitalization has broader implications than digitization (Verhoef et al., 2019) but also remains a one-time duty because of its limited scope to isolated processes (Mettler & Pinto, 2018). It can be measured with (digital) KPIs such as ROI/ROE but also customer engagement measures (Table 1) We refer to “digitalization as the addition or creation of processes by digital technology with the goal to enhance the performance by connecting formerly unconnected parts with each other” (Verhoef et al., 2019).
8
Given the three phases of digital change, we have also defined strategic imperatives in our earlier paper. Those imperatives on the one side help clarify the phases by showing examples, resources and capabilities needed and on the other side helps to identify goals and metrics for each phase. The strategic imperatives itself rely on the narrative that established companies undergo a gradual step by step approach to reach transformation (Verhoef et al., 2019). We theorized that moving to new ways of doing business results in conflict with existing practices (Markides, 2006). Moreover, new digital business models may cannibalize existing ways of value appropriation (Teece, 2010). Thus, for incumbents, the digital journey likely starts with small changes by digitizing analog information first and follows by digitalizing distinct processes (Verhoef et al., 2019). Only if the necessary knowledge and experience are collected, the bigger scale (digital) transformation is following. In line with this argument, success should be measured with a variety of metrics.
We have also defined growth strategies based on the different phases that help firms to develop the necessary capabilities to be able to move to the next phase. These growth strategies are based upon the Ansoff matrix (Ansoff, 1957) and have been expanded by (platform) driven business growth strategies. First, we adapted the existing growth strategies (market/product development) from Ansoff to the digital era. Those are exclusively available at the digitization phase because they allow leveraging new efficiency gains. In the digitalization phase, we argue that because distinct (but isolated) processes are subject to change, the former growth strategies can be used, but also the development of platforms becomes a growth strategy for companies. An example in this phase are co-creation platforms (Ramaswamy & Ozcan, 2016). For the digital transformation phase with the goal of new digital business development, all former strategies are available. This is because, with all the former strategies, new business models can be established. Additionally, for companies that already created platforms, we argue that platform diversification as the addition of new products and services to reach new markets is a viable strategy (Verhoef et al., 2019).
9
costs (Verhoef et al., 2019). With digitalized processes, digital metrics such as customer experience become more prevalent, while in the transformation phase the breadth of digital metrics should be expanded to cover among for e.g. the digital share of value creation (Verhoef et al., 2019).
Table 1. Three phases of digital change
Phase Examples Digital Resources and Capabilities Organization al Structure Digital Growth Strategies Metrics Goal Digitization Automated routines and tasks; Conversion of analog into digital information Digital assets Standard top-down hierarchy Market penetration, (product-based) Market development , Product development Traditional KPIs: Cost-to-serve customer, ROI, ROA Cost savings & more efficient deployment of resources for existing activities. Digitalization Addition of digital components to product or service offering; Introduction of digital distribution and communication channels. [Above] + Separate, agile units [Above] + [Above] + Cost savings & increased revenues; More efficient production via business process re-engineering; Enhanced customer experience. Digital agility, Reconfiguring capability, Networking capability Platform-based market penetration, Co-creation platform Digital KPIs: User experience, Unique customers/user s, active customers/user s Digital transformation Introduction of new business models like ‘product-as-a-service’, digital platforms, and pure data-driven business models
[Above] + Separate units with flexible org. forms, internalization of IT and analytical functional areas
[Above] + [Above] + New cost-revenue model: Reconfigurat ion of assets to develop new business models. Big data capability Platform diversificatio n Digital KPIs: Digital share, magnitude and momentum, co-creator sentiment THEORETICAL FOUNDATION
10 Firm digital performance
We are interested in firm digital performance related to digital transformation. Firm performance has been used extensively in the literature as outcome variable often related to some form financial performance but also other highly context-dependent variables (e.g. Rai, Patnayakuni, & Seth, 2006; Teece, 2010). When considering the goal of digital transformation to create new business models (Table 1), we need a more holistic overview than just financial performance. For example, big data usage (capability) alone stimulates firm performance (Wang et al., 2018). The outcomes include among others the support in business growth (Wang et al., 2018) or in other words, the operating performance of the company (Wang, Zhong, Liang, Xiao, & Xue, 2012). Moreover, the capability to use digital technology has been found to enhance financial performance (Karimi & Walter, 2015; Quaadgras, Weill, & Ross, 2014), which in other words means the market performance of the firm (Wang et al., 2012). We adapt our construct from Wang et al. (2012) and see firm digital performance reflected by two items. First, the additional value generated in financial means (Wang et al., 2012). Second, the realized operational benefits as digital business models are more efficient (Wang et al., 2012; Verhoef et al., 2019). Thus, we conceptualize firm digital performance as the additional value generated with the help of digital technology in operating (creating new products or services) and market performance (creating additional revenue) (Wang et al., 2012; Verhoef et al., 2019).
Digital Transformation
11
Digital capabilities
A different way to describe the utilization of technology is as resources and capabilities following the resource-based view (RBV) by Barney (1991). Resources express a companies ownership and control over assets (Barney, 1991). Therefore, assets refer for example to physical or intellectual ownership over such as patents or IT infrastructure (Kawakami, Barczak, & Durmusoglu, 2015; Zhang & Dhaliwal, 2009). Moreover, capabilities refer to the capacity of the company to use resources in (novel ways) to create value (Barney, 1991; Prahalad & Hamid, 1990). We have theorized that digital assets like AI, IoT or big data analytics are necessary (Verhoef et al., 2019), but it is more important how these assets are used by companies. Considering that digital assets are widely available today at very low cost for firms, the use of the asset is much more important than the asset/resource (Tabrizi et al., 2019; Lucas et al., 2013). Thus, the discussion must be about firm capabilities rather than resources in case of digital transformation.
12
technology in novel and innovative ways compared to companies that do not or only partly possess these capabilities.
Therefore, based on dynamic capabilities we define digital capabilities as the firm’s ability to integrate, build, and reconfigure internal and external competencies to utilize digital technology in a rapidly changing environment (Teece, Pisano, & Shuen, 1997). Generally, the higher the level of the digital capabilities, the better and more complex use of new digital technology is possible. Hence, digital capabilities are an important (first order) element that forms digital transformation.
Digital business model change
The second part of our definition of digital transformation concerns the changes in value creation, appropriation and thus in the business model of the company (Verhoef et al., 2019). Business models have been defined in various ways (Li, 2018). A recent paper by (Massa, Tucci, Afuah, 2017) identified over 70 different conceptual definitions for business models. Among others, it has been identified as a structural template (Amit & Zott, 2001), conceptual model (Osterwalder, Pigneur, & Tucco, 2005), and set (Zott, Amit, & Massa, 2011). Moreover, business models are a way for firms to gain competitive advantage (e.g. Osterwalder & Pigneur, 2010; Teece, 2010) and are defined as “how the enterprise creates and delivers value to customers, and then converts payment received to profit” (Teece, 2010: p. 173). We follow the definition of Teece (2010), which is in line with Li (2018) who pointed out that business models explain how firms make money as well as create and appropriate value.
13
Therefore, the change in business models is defined as the change in how the enterprise creates and delivers value to its customers, and then converts payments received into profits (Teece, 2010). Hence, the change towards new digital business models is a key activity in digital transformation because it allows companies to not only “play around” with technology but earn money with the utilization of digital technology. Hence, the change towards digital business models is the second (first order) element that forms digital transformation as a construct.
A (r)evolutionary perspective
In our paper, we proposed that digital transformation for incumbent firms is reached by a step by step approach depending on the gradual acquisition of knowledge (Verhoef et al., 2019). Subsequently, firms first need to engage in digitization to create digital information, which is followed by digitalized processes. Only if the knowledge in these steps has been acquired, the full-scale transformation of the business model in favor of a new digital one is possible. Thus, another way to look at digital transformation is in terms of an evolutionary process (Nelson & Winter, 1982).
Evolutionary theory is grounded in economic theory and argues that firms develop and acquire new knowledge over time (Nelson & Winter, 1982). Thus, it follows four basic premises: variety, behavioral continuity, profit induced balance and limited path dependency (Nelson & Winter, 2002). The first premise refers to the variety of firms and approaches in the marketplace such that a necessary condition for evolution to occur in the presence of different conditions and approaches. The second premise assumes that firms that succeeded with certain actions in the past are likely to apply the same actions in the future. The third premise follows this logic and argues that firms will likely spend more money and further grow with actions that were rewarded in the past rather than to diversify too broadly. Lastly, the third premise of (limited) path dependency follows and posits that firms have put resources into certain actions, are constrained in their future behavior by past investments.
14
technology and lead to firms being trapped in old technology despite new and better ones are available (Zhu, Kraemer, & Gurbaxani, 2006). Moreover, these arguments in IS literature follow the basic premise that firms learn over time (Nelson & Winter, 2002) and that future action is related to past action.
For digital transformation these assumptions also hold. First, investing in digital transformation and thus in the capabilities of the firm to handle (new) digital technology (e.g. Teece, 2010) and generate value from it (e.g. Osterwalder & Pigneur, 2010) puts the company on a trajectory that is not easy to change once it has been established. In other words, the investment in human capital and technology lead to path-dependent behavior of the firm because costs to change are high once a (substantial) commitment has been made (e.g. Keil, 1995; Zhu, Kraemer, & Gurbaxani, 2006; Nelson & Winter, 1982; 2002). Thus, firms that engage in digital transformation tend to follow established paths i.e. rely on technology and capabilities that are familiar to already existing knowledge (Cohen and Levinthal, 1990). Second, knowledge is accrued over time and firms gradually develop their knowledge base. Thus, firms don’t adapt fast i.e. in the short term (path dependency) but more slowly and gradually in the mid to long term (Nelson & Winter, 2002).
Openness to ideas
We have defined the gradual acquisition of knowledge, in other words learning, as key requisite for digital transformation (Verhoef et al., 2019). Because digital transformation requires the acquisition of new capabilities and change in the business model, the openness towards new ideas as a facilitator of learning, is a key contingency factor in the relationship between digital transformation and firm digital performance.
15
technology and/or innovation. Therefore, we conceptualize openness to ideas as the firm's ability to be open towards new ideas and ways of working (Soto et al., 2011; Venkatesh & Bala, 2012).
Radicalness of change
The goal of digital transformation is a change in the core value creation mechanisms of the company (Schallmo et al., 2017) by gradually acquiring new knowledge (Verhoef et al., 2019). We have proposed that digital transformation depends on previously acquired knowledge from digitization and digitalization (Verhoef et al., 2019). Following innovation terminology, digital transformation is incremental (evolutionary) in nature. The counterargument would be that digital transformation can also be reached by revolutionary action. Revolution is defined as a radical shift in the current paradigm, for e.g. the photography industry, which changed dramatically due to a combination of ICT technology and cameras (Lucas & Goh, 2009). Thus, the radicalness of changes is the second important contingency.
Radicalness has been defined in innovation literature as an advance in price/performance that is faster than the actual rate of progress (Gatignon, Tushman, Smith, & Anderson, 2002). Moreover, radical innovation has been identified as a key driver of disruption (Zhou, 2013). For example, in IS literature, radicalness has been studied to understand how knowledge drives (radical) innovation (Luo, Lyytinen, & Rose, 2012). Radicalness is defined as “when innovators need to acquire extensively unique and novel technological and process-related know-what, know-why- and know-why” (Luo et al., 2012). Additionally, “change (of information systems) covers the generation, implementation, and adoption of new elements in an organizational system […]” (Lyytinen & Newman, 2008). Thus, change covers the addition of new elements (e.g. business models) to an existing system, while radicalness describes the pace of change. Therefore, we conceptualize radicalness of change as the advance in the novel and groundbreaking ideas, technologies and ways of working to the existing business activities that are faster than the actual rate of progress (Luo et al., 2012; Lyytinen & Newman, 2008).
Conceptual model
16
relationship between digital transformation and digital firm performance as well as contingencies and control variables.
Table 2. Construct Definitions
Construct Definition Source
Digital Transformation
as a firm’s change in the utilization of digital technologies to create new ways of value creation and
appropriation and, thus, to develop a new digital business model.
(Verhoef et al., 2019)
Digital Capabilities*
as the firm’s ability to integrate, build, and reconfigure internal and external competencies to utilize digital
technology in a rapidly changing environment.
(Teece, Pisano, & Shuen, 1997; Sambamurthy, Bharadwaj, & Grover, 2003; Karimi & Walter, 2015; Wang, et al., 2018; Verhoef
et al., 2019)
Digital Business Models*
as the change in how the enterprise creates and delivers value to its customers, and then converts
payments received into profits.
(Teece, 2010; Osterwalder & Pigneur, 2010; Liu et al., 2011; Li,
2018)
Firm Digital Performance
as the additional value generated with the help of digital technology in operating, creating new products
or services, and market performance as creating additional revenue.
(Wang et al., 2012; Verhoef et al., 2019)
Openness to
Ideas as the firm’s ability to be open towards new ideas and ways of working.
(Soto et al., 2011; Venkatesh & Bala, 2012)
Radicalness of Change
as the advance in the novel and groundbreaking ideas, technologies and ways of working to the existing business activities that are faster than the actual rate of
progress.
(Gatignon et al., 2002; Lyytinen & Newman, 2008; Luo et al., 2012)
17
HYPOTHESIE DEVELOPMENT Digital transformation and firm digital performance
Digital transformation is defined as a firm’s change in the utilization of digital technologies to create new ways of value creation and appropriation and, thus, to develop a new digital business model (Verhoef et al., 2019). Without new digital technology, new business models would have not been possible (Lokuge, Sedera, Grover, & Dongming, 2019) but the successful utilization of these technologies is key to determine their impact on firm performance. We define firm digital performance as the additional value generated with the help of digital technology in operating, creating new products or services, and market performance as creating additional revenue (Wang et al., 2012; Verhoef et al., 2019). Thus, the link between digital transformation and firm digital performance depends on the degree to which companies can both develop the capabilities to utilize digital technology and how well these changes are translated in actual changes in the business (model).
The digital transformation causes changes in the way how value is created through for e.g. improving decision making performance (Ghasemaghaei, Ebrahimi, & Hassanein, 2018) or stimulating growth opportunities (Wang et al., 2018). More importantly, digital business models are generally more profitable than non-digital business models (Verhoef et al., 2019). When comparing the selected company’s EBIT and net profit with each other, we found that digital business models are on average around 10x more profitable than non-digital business models (Verhoef et al., 2019). Multiple companies have emerged that were unknown 25 years ago such as Amazon, Facebook, eBay or Uber that serve as examples (Kannan & Li, 2017). A common business model of those digital players is the platform-based model, in which the company only serves as a mediator between supply and demand such as Uber or Airbnb (Sorescu, 2017). Given that those business models require much less human labor, not many physical structures such as buildings and the costs of adding more people to the platform are very small, those business models are highly profitable (Verhoef et al., 2019). Even platform-based business models are highly profitable because they benefit from the same underlying principles such as less required human labor, less to no physical locations, easy scalability and low distribution cost. For these changes to occur, firms are required to make profound changes to their existing ways of value creation.
18
(Verhoef et al., 2019). Moreover, following the assumptions from evolutionary theory, firms are expected to continue with their efforts of digital change and digital transformation (Nelson & Winter, 2002). Thus, despite the fact that some firms will fail on the way or will face high costs of transformation. Considering the benefits of digital business models compared to non-digital business models, we can expect that more non-digital transformation will lead to more digital firm performance across all firms in the market.
In sum, digital transformation stimulates new ways of value creation and appropriation, which increases firms operating performance by creating new products and services that are less cost intensive. And because new digital business models are more profitable in terms of market performance than physical ones, digital transformation on average is expected to have a positive influence on digital financial firm performance. Hence, we expect the following hypothesis:
Hypothesis 1: On average, there is a positive relationship between digital
transformation and firm digital performance. The moderating role of openness to ideas
Digital transformation requires firms to utilize digital technology in new ways to create and appropriate value (Verhoef et al., 2019). As digital transformation is more than only providing services in new ways (Agarwal et al., 2010) or entering of new markets (Li, 2018), organization-wide learning of new capabilities and ways of value creation is required to change the organization. However, a necessary precondition for learning is the openness towards new ideas and innovations (e.g. Quinton et al., 2018; Loguke et al., 2019). We define openness to ideas as the firm’s ability to be open towards new ideas and ways of working (Soto et al., 2011; Venkatesh & Bala., 2012).
19
news industry, these business models can mean for example the replacement of old paper-based business models in favor of online platforms (Aubert-Tarby et al., 2018).
A prerequisite for learning and development is the openness of the organization towards new ideas (Karimi & Walter, 2015). Previous research has identified openness as a key driver of innovation and change (e.g. Abraham & Junglas, 2011; Lokuge et al., 2019; Quinton et al., 2018). Similar concepts are learning orientation or in other words, openness to ideas and thus the questioning of beliefs and procedures (e.g. Quinton et al., 2018). In more open organizations established practices are more often questioned and subsequently changed for more efficient and/or effective practices (Day & Schoemaker, 2006; Quinton et al., 2018). Moreover, in these organizations adoption of technology is more likely (Quinton et al., 2018). Thus, openness facilitates learning as a key prerequisite for digital transformation.
As previous research has shown, openness is positively linked to digital performance (Quinton et al., 2018). Organizations that are more open are more likely to embrace in change, adopt new procedures and thus capture value from them (Quinton et al., 2018). In order to digitally transform, firms are required to adopt new ways of value creation, learn new capabilities, change their organizational structure and finally create new digital business models. (Verhoef et al., 2019). Hence, openness facilitates learning (e.g. Abraham & Junglas, 2011; Lokuge et al., 2019; Quinton et al., 2018). Given the benefits of digital business models and the theorized relationship between digital transformation and firm digital performance, we can expect that firms that score higher on openness to ideas will easier learn to facilitate change. According to evolutionary theory (Nelson & Winter, 2002), past decisions shape future outcomes. Thus, firms that are more open will likely better learn to utilize technology and will better create value with them. As new business models are not perfect from their beginning, the willingness to learn and adapt them, in other words, the openness to new ideas and change, is positively linked to the impact of that business model (Teece, 2010). Subsequently, we can assume that organizations that are more open to ideas will have an easier time to adjust to the necessities of a vastly changing digital environment (Lokuge et al., 2019). Thus, the relationship between digital transformation and digital firm performance will be strengthened for firms with higher levels of openness as a key facilitator of change. Hence, we expect the following hypothesis:
Hypothesis 2: The positive relationship between digital transformation and firm
20
openness strengthen the relationship between digital transformation and firm digital performance.
The moderating role of radicalness of changes
Digital transformation as the last form of digital change for incumbent firms can be reached by a step by step approach in which new knowledge is gradually developed (Verhoef et al., 2019). Thus, we follow basic evolutionary assumptions such as continued behavior and path dependency that pose constraints on how firms develop new knowledge (Nelson & Winter, 2002; Cohen & Levinthal, 1990). The digital transformation is a highly complex topic and of the established firms that embark on the digital transformation journey, many fail (Tabrizi et al., 2019). In the executive literature, we found evidence that strategy and not technology is driving digital transformation (Kane et al., 2015). However, a common problem in transformation efforts is that firms often fail to take the necessary time to incrementally develop ways of working. In fact, they underestimate the time gradual development takes and shift too radically towards new organizational forms such as agile units to facilitate digital transformation (e.g. Sebastian et al., 2017). We define radicalness of change as the advance in novel and groundbreaking ideas, technologies and ways of working to the existing business activities that are faster than the actual rate of progress (Gatignon et al., 2002; Lyytinen & Newman, 2008; Luo et al., 2012).
21
specific path (Zhu, Kraemer, & Gurbaxani, 2006). Thus, advancing radically beyond the normal pace of learning, or even switching paths likely causes harm to the organization because previously learned knowledge becomes (partly) obsolete.
The argument of path dependency and continued behavior poses significant constraints on companies’ behavior in the short term. Radically imposing new ways of working, without prior experience likely causes high (switching) costs (Zhu, Kraemer, & Gurbaxani, 2006). Subsequently, product or service development might be delayed. Hence, radicalness in pursuing a new path to revolutionize the company rather than following an evolutionary pathway causes harm to the development of new knowledge and more importantly, the establishment of new business models. Hence, the relationship between digital transformation and firm digital performance is harmed by higher levels of radical change because the value of prior knowledge depreciates. Hence, we expect the following hypothesis:
Hypothesis 3: The positive relationship between digital transformation and firm
digital performance is negatively moderated by radicalness of changes, such that high levels of radicalness weaken the relationship between digital transformation and firm digital performance.
METHODS Data collection
22
leaving us with 201 respondents that fulfill all criteria and can be used for hypothesis testing. The firms are on average 41 years old (mean =40.94; SD = 63.73) and have on average 260 employees (mean = 260.83; SD = 1380.09).
While we are not the original creators of the survey, it has several advantages to use. First, insights into (digital) innovation activities are hard to collect on a larger scale especially with sufficiently high response rates. Second, the sample is representative of the business structure in the North of the Netherlands, which provides us with enough confidence to have generalizability of our findings.
Development of measures
Given that there are no measures in the literature for digital transformation and firm digital performance, we follow established best practices for instrument development and validation by MacKenzie, Podsakoff, & Podsakoff (2011). In the first phase, we used qualitative methods to conceptualize and operationalize digital transformation and firm digital performance based on a literature review and expert interviews to subsequently develop measurement items (Rossiter, 2002). The second phase then employs quantitative methods for instrument testing and validation.
Conceptually we have defined digital transformation as 2nd order formative construct. Formative measures do not necessarily exist as “real entity” in the world and are theoretically constructed (MacKenzie et al., 2011). We choose a formative way of theorizing because the construct of interest does not exist in the real world but is artificially constructed. In line with this argument, the forming elements have to represent all aspects of digital transformation (MacKenzie et al., 2011). The 1st order elements that form digital transformation are (1) digital capabilities and (2) business model change. We have found both in the literature to be important drivers of digital transformation. In line with formative theorizing, both items can exist on itself and in different nomological networks at the same. Both first-order elements are designed as reflective items.
Scale development
23
both first-order elements. By these interviews, we ensure the content validity of the operationalization phase and assure that the construct objects and attributes are correctly specified (Rossiter, 2002). This qualitative interim step between conceptualization and item generation (MacKenzie et al., 2011) is warranted if items are adapted to a new context, or in our case as there are no existing measures and ensure conceptual and empirical relevance for the target population (Lokuge et al., 2019).
We selected experts in organizations that are representative for the sample we want to study based on criteria such as firm size, the number of employees and sector to have a broad overview about digital transformation and its implications (e.g. Dubé & Paré, 2003; Seawright, Gerring, & Seawright, 2008). We used our network to find and select in a total of 10 organizations. We used semi-structured interviews with executives responsible for (digital) innovation/technology/change in case of bigger organizations or the general manager of the organization in case of smaller organizations. The questions were based on the previous literature review guided by the main research question “What is the digital transformation in your opinion and how is it affecting your company?”
The information obtained in during the interviews serves, in combination with the literature, as the basis for the development of measures for digital transformation and firm performance. We used the findings from the interviews as a basis to validate the previous findings from the literature and as additional proof that our conceptual findings are also reflected in the real world. Based on the interviews and literature we operationalize digital transformation as a second-order formative construct, which describes the utilization of digital technology to create and appropriate value for the enterprise. It is formed by reflective first order digital capabilities and change in the business model of the company. Both first-order elements will be assessed using six item 5-point Likert scales. Furthermore, we operationalize firm digital performance based on (Wang et al., 2012) as reflective construct measuring the two dimensions (1) additional revenue generated as well as (2) the additional operating benefits generated measured by one question each on a 5-point Likert scale.
24
based on a panel of respondents (MacKenzie et al., 2011). For both phases, we used postgraduate students from the business domain familiar with business research. The choice for the sample was that business managers who must fill in the survey later likely have received a comparable type of education. In the first phase, we showed the items and asked the raters to group them into constructs. As expected, we did not saw a great degree of fit to our theorized model. Given that digital transformation is conceptualized as a second-order construct and does not exist as “real entity” the outcome although makes sense. Based on the comments of the raters we have rewritten two ambiguous items because the question was not well understood. In the second phase, we presented the overall concepts to the raters and gave them the corresponding items without telling them which item belongs to which concept. In total, we had an interrater agreement (alpha) of 0.72. Given that this value is above the commonly used threshold of 0.7 we concluded that we can proceed with the quantitative face. Lastly, we have made slight adjustments to the clarity of the questions based on the comments of the raters.
Pilot test
After the items have been generated, we put them into an online survey to evaluate the scale and refine it if necessary (MacKenzie et al., 2011). We have created an online survey and distributed it in the wider network of our center. We used this approach to again have access to a varied sample including firms with different size and industry. Therefore, we can ensure that the pilot sample corresponds to the goal of the study and later to the bigger sample. In total, we collected 101 responses to our survey including 61 respondents that filled in the entire survey. We have collected firm names of approximately half the organizations and used it to check if there is enough variation in the sample as well as if there are industries or areas that dominate the sample.
25
For composite reliability, we first checked Cronbach’s Alpha values. For all three constructs, alpha was above .8 and therefore sufficiently high. Furthermore, we checked the composite reliability scores which are all above 0.8 as well. Thus, for composite reliability, we have enough evidence to proceed with the next tests. We continued to check the significance level of the factors loadings to determine how well the factors each load on our formative constructs. In this case, the t-statistic values were all above 5 (mostly over 10) indicating high statistical significance. We then checked the construct validity by using the average variance extracted (AVE). In this case, all constructs showed significant high convergent and discriminant validity with the AVE for all constructs measuring above 0.5 (Fornell & Larcker, 1981). The AVE of each individual construct was higher than the shared variance between the constructs indicating strong discriminant validity. We also checked the cross loading of the factors using an unrotated component analysis. The analysis reveals that there are two factors with an eigenvalue above 1 and that the loadings are according to the theoretical rationale.
Following established best practices (Li et al., 2013), we tested for multicollinearity among the measures using the variance inflation factors (VIF). The VIF scores in PLS ranged from 1.8 to 3.2, which provides evidence that multicollinearity is not an issue in our sample. Additionally, we checked the significance of the indicators in the overall measurement model at the 0.05 alpha level. Both of our constructs are highly significant and explain above 50% of the variance. Lastly, we checked the nomological validity of our constructs in the nomological network. Thus, we used our model to predict firm digital performance. It is an essential part of testing the validity of new constructs (Lokuge et al., 2019). For the dependent variable, firm digital performance, as explained before we adapted it from existing literature and tested it as a reflective construct in this study with factor loadings for a single construct of 0.861 and 0.850. The path coefficient is t=0.540 with an R² of 0.291. Thus, we have strong evidence for nomological validity of the construct. Moreover, we have enough evidence that the developed scale is useful for predictions in the real world and can be used for the subsequent steps.
Measures
Table 3. Measurement Items
Construct Measurement Items Scale
Digital Transformation
2nd order (formative) item, which is formed by digital capabilities and digital business model change
Digital Capabilities*
1. The ability and knowledge of the employees to use digital technologies
26
2. The ability to manage and analyze data of clients, competitors and other market data to capture value from it
extent 3. The ability to find the best and most suitable digital
technologies
4. The ability to integrate new digital technologies into our business activities
5. The ability to change business activities to create additional value for clients with new digital technology 6. The ability, knowledge, and awareness to handle digital technology ethically (privacy, security)
Digital Business Models*
1. Digital technology has changed the way how we do business
5-point scale; 1= not at all; 5= to a great
extent 2. Digital technology has changed our internal operations
3. Digital technology has changed the collaboration with partners
4. The knowledge to use digital technology to improve the customer's journey (analyzing and collecting data) 5. We have entered new geographical markets and serve new customers with the help of digital technology 6. Our role in the value system has changed (e.g. extending markets via online channels, entering new business fields)
Firm Digital Performance
1. we have been able to create additional revenue with the
help of digital technology and data 5-point scale; 1= not at all; 5= to a great
extent 2. We have increased the efficiency of our company with
the help of digital technology and data Openness to
Ideas
1. The use of creative ideas in the company is a problem 4-point scale: 1= High, 2= moderate, 3=
low, 4 = none 2. We have difficulties to translate creative ideas into
concrete innovations
Radicalness of Change
1. Our workers have groundbreaking ideas about new procedures, processes, and products
5-point scale: 1= strongly disagree, 5=
strongly agree 2. Our workers have pioneering ideas about practices and
routines
3. Our workers have ideas to handle work in fully new ways
Age When was your organization found?
indication of number Size How many employees does your organization have in the
year 2018?
Industry What is the SBI code of the main business activity of your business?
27
Digital transformation
We measure digital transformation as a second-order formative construct (Table 3). It is formed by two first order items digital capabilities and digital business model change, which are both measured with six survey items each as reflective indicators. For digital capabilities, we ask if the company possesses for e.g. the capability to manage data ethically or to find new digital technology that is useful for the business. In contrast to business model change, we measure for e.g. the ability of the company to connect with other parties or to the ability to find reach new markets and customers. The items are all measured on a five-point Likert scales (not at all – to a great extent).
Firm digital performance
We measure firm digital performance with two survey questions asking for the extra revenue generated by digital technology as well as the business improvement generated. Both questions were adapted from Wang et al. (2012) for the purposes of this study (Table 3). For both questions, we use a five point Likert scale (not at all – to a great extent).
Openness to ideas
Openness to ideas is measured with two questions in the survey and was adapted from Soto et al. (2011) and Venkatesh & Bala (2012). While the original measure of openness to ideas stems from the Big5 personality inventory, the openness to innovation construct measures the same idea at the organizational level. We have adapted our questions to the study setting. Both questions use a 4-point Likert scale in which higher scores equal more openness to ideas (Table 3).
Radicalness of changes
Similar to the former construct, radicalness of changes was adapted from Gatignon et al. (2002) and Luo et al. (2012). It originally referred to the radical innovation potential of the individual workforce. We have adapted it and use it at the firm level construct since the capabilities of the workforce can be regarded as a resource under the control of the firm (Barney, 1991). It is measured with three questions on a five-point Likert scale in which higher scores indicate more radicalness of changes.
Control variables
28
Second, we use the number of employees to control for firm size (e.g. Rai et al., 2006), which is also used as a natural logarithm. Lastly, we control for industry using the Dutch SBI industry codes (e.g. Yu et al., 2018). We used them to create four broad groups: manufacturing, retail, services, and public, which are based on the Central Bureau of Statistics (CBS) classification. In this case, we created four binary dummy variables to control for industry effects (we only use three in the analysis and services as the biggest category as a reference point).
Non-response bias analysis
We checked for non-response bias by making a median split in the dataset for response data. Based on the two groups we conducted a test for difference in digital transformation scores (t = 0.445; p = 0.667) which indicates that there is no evidence to believe in a difference in answering time. Thus, we conclude that nonresponse bias is no serious problem in our data collection.
Safeguard against common method bias
We also tried to avoid and checked for common method bias (CMB). Mixed scales (e.g., binary, 5-point, etc.) are used in survey design, which helped reducing CMB (Podsakoff et al., 2003). We conducted Harman’s single factor test followed by a marker variable test (Malhotra & Kim, 2006). First, for Harman’s single-factor test, we found 3 factors with eigenvalues greater than 1 of which none explained more than 39% of the variance. Second, for the marker variable test, we followed Lindell & Whitney (2001) to use the smallest (0.011) and second smallest correlation (0.02) as the proxies for CMV (Table 4). We found that partial correlations for the variables that were significant remained significant. Thus, we have enough evidence to conclude that common method bias did not seriously affect the validity of this study.
Table 4
Assessment of Common Method Bias
Antecedents of firm digital performance
Zero-order correlation
First smallest correlation as
CMV proxy
29 Openness to ideas 0.078 12.82 0.068 24.36 0.059 Size 0.106 9.430 0.096 16.98 0.088 Age 0.020 55.00 0.009 100.00 0.000 Note: * p < 0.05; ** p < 0.01; *** p < 0.001 RESULTS
We used a two-folded analysis strategy because of the nature of the research design. First, we used partial least squares (PLS) structural equation modeling (SEM) to estimate the basic measurement model between digital transformation and firm digital performance. The use of PLS is widely recognized in the IS field and mainly used to test (complex) predictive models and supports formative rather than only reflective models (Li et al., 2013; Lokuge et al., 2019). More importantly, PLS is useful if the research model is at an early stage (Zhu, Kraemer, & Gurbaxani, 2006). Thus, because our study is the first effort to empirically measure the impact of digital transformation on firm digital performance, PLS is the right statistical technique to estimate the basic measurement model. Based on this initial estimation of the measurement model. We use the latent variable scores (LVS) for digital transformation, firm digital performance, the moderators and control variables in ordinary least squares (OLS) regression. While we could estimate the model also in PLS, with the interdisciplinary element of this study and with regards to the fields we have based our conceptual development on, we want to rely on a more conservative technique such as OLS regression.
Measurement properties
30
performed additional checks such as unrotated factor analysis and all constructs load according to their theoretical properties. The last step was to test the path coefficients to determine if digital transformation influences firm digital performance. The PLS results were significant (t = 19.007, p < 0.000) indicating a significant positive relationship between digital transformation and firm digital performance. Thus, we have enough evidence to proceed with hypothesis testing based on the LVS scores for digital transformation and firm digital performance.
Table 5
Descriptive Statistics and correlations Mean SD 1 2 3 4 5 (1) Digital Performance 0.000 1.002 (2) Digital Transformation 0.000 1.002 0.699** (3) Radicalness of changes 3.208 1.018 0.198** 0.185** (4) Openness to ideas 3.051 0.801 0.078 0.020 0.522** (5) Size 1.078 0.873 0.106 0.175** -0.061 -0.353** (6) Firm Age 1.219 0.528 0.020 0.011 -0.301** -0.339** 0.528**
**. Correlation is significant at the 0.01 level (2-tailed). Correlations in bold are proxy for CMV.
To the test the hypothesis we used the LVS scores for DT and DP and included them in the original dataset. As specified before, we perform the hypothesis testing on a sample of 201 firms. We have also tested the internal consistency for the sum variables of openness to ideas (alpha = 0.692) and radicalness of changes (alpha = 0.888). The scores for openness to ideas are slightly below the recommended value. However, because the value is only slightly below the recommended range and because of good theoretical reasons, we decided to keep this item. The correlations among variables and descriptive are reported above in table 5. Hypothesis testing
31
for hypothesis one (b= 0.818, p < 0.000). Thus, confirming our hypothesis that digital transformation has a positive impact on firm digital performance.
Table 6
OLS regression results
(1) (2) (3) Digital transformation (H1: +) 0.712*** 0.818*** (0.054) (0.187) Digital Transformation x Openness to ideas (H2: +) 0.198*** (0.073) Digital Transformation x Radicalness of changes (H3: –) -0.220** (0.061) Openness to ideas 0.055 0.016 0.008 (0.104) (0.075) (0.073) Radicalness of changes 0.145 0.076 0.091 (0.083) (0.061) (0.059) Size -0.027 -0.176* -0.178* (0.110) (0.081) (0.079) Age -0.103 0.057 0.044 0.162 (0.118) (0.115) Manufacturing sector -0.720 0.015 0.052 (0.178) (0.129) (0.126) Retail sector 0.254 0.076 0.051 (0.178) (0.143) (0.139) Public sector -0.180 0.138 0.149 (0.221) (0.162) (0.158) Constant -0.336 -0.134 -0.129 (0.410) (0.298) (0.290) R² 0,064 0.509 0.541 Adj. R² 0,030 0.489 0.517 Delta R² - 0.459 0.023 F 1.871 24.925*** 22.296*** AIC -9.42 -137 -147 BIC 17.01 -107.64 -110.40 n 201 201 201
Note: * p < 0.05; ** p < 0.01; ***p < = 0.001. Standard errors are in parentheses. Dependent variable is firm digital performance (DP)
32
digital performance. For this, we calculated the interaction term between digital transformation and openness to ideas. We estimated a partial model with only the main effect of openness on firm digital performance which is not significant (2). However, in the full model the interaction term of digital transformation and openness to ideas is statistically highly significant (b = 0.198, p < 0.000). Additionally, we also performed a slope test following Aiken & West (1991), in which we plotted the significant interaction effect (Fig. 2a). We found that the effect of digital transformation on digital firm performance more positive for higher levels openness (high openness: b=1.609, p < 0.05; low openness: b=1.016, p < 0.000). Thus, we have evidence for the second hypothesis that higher levels of openness to ideas strengthen the relationship between digital transformation and firm digital performance.
33 Endogeneity test
Given the cross-sectional nature of our data, it should be interpreted as an association rather than causation. Hence, we can expect that endogeneity caused by reverse causality or simultaneity makes the OLS results cross inconsistent. Endogeneity could be caused because firms with higher digital performance may have more financial resources that can be invested in the development of digital capabilities, suggesting that correlation could be driven by reverse causality. Additionally, it could be that important variables were not collected that simultaneously affect digital capabilities and digital business model change, thus leading to an illusory correlation that is observed. To test for endogeneity issues, we follow Bharadwaj, Bharadwaj, & Bendoly (2007) & Dong, McCarthy & Schoenmakers (2017) to estimate a two-stage Heckman model.
In the first step, we took the mean value for digital transformation and created a dummy variable indicating whether a firm had a digital transformation value greater than the mean of the sample. Subsequently, we estimated a Probit model. The model demonstrated a good fit (Table 7). Based on that model we calculated the inverse Mills ratio (IMR), that represents the propensity for digital transformation, that is endogenously determined. We found that IMR was statistically significant (p < 0.05), providing evidence that endogeneity exists. After controlling for endogeneity, the results are still consistent with the earlier OLS regression. Especially, the moderation of radicalness and openness are both still significant and with the expected sign. Therefore, we can conclude that endogeneity does not bias the conclusion of the study, although the digital transformation is endogenous.
Table 7
Heckman Model Results
First Stage Second Stage
Digital transformation 0.922***
(0.191)
Digital Transformation x Openness to ideas 0.191**
(0.073) Digital Transformation x Radicalness of
34 (0.236) (0.332) Manufacturing dummy 0.607 0.377 (0.472) (0.199) Retail dummy 0.734 0.266 (0.496) (0.172) Public dummy 0.305 -0.674 (0.292) (0.422)
Inverse mills ratio 2.84**
(1.354) Constant 0.226 -4.208** (0.303) (1.960) Pseudo R² 0.0366 Chi-square 9.97*** R² 0.5515 Adj. R² 0.5254 F 21.13*** n 201 201
Note: ** p < 0.05; *** p < 0.01; p < = 0.001. Standard errors are in parentheses. Dependent variable in stage one is a dummy variable indicating whether the sum of digital transformation is greater than the mean. Dependent variable in second stage is market performance
DISCUSSION Theoretical implications
In this paper, we built on our earlier work (Verhoef et al., 2019) and take a (r)evolutionary perspective to investigate the impact of digital transformation on firms as well as discuss important contingencies of that relationship. With this approach, we make several contributions to the emerging research on digital transformation.
35
empirical study yet published about the impact of digital transformation on firms. Therefore, our study provides a new perspective on the mechanisms of digital transformation and is among the first that shows quantitative and thus more generalizable insights about its impact on firms.
Additionally, we elaborate more closely on the contingencies of digital transformation and deepen our theorizing by considering openness as a facilitator of the development of new digital knowledge. We take an evolutionary perspective (Nelson & Winter, 2002) based on our earlier argument that digital transformation for incumbent firms is reached by the incremental accumulation of knowledge and capabilities (Verhoef et al., 2019). We explain how digital transformation depends on learning to use new technology (e.g. Karimi & Walter 2015; Verhoef et al., 2019) and how this process is contingent to evolutionary assumptions such as continued behavior (Nelson & Winter, 2002). In line with the argument that digital transformation depends on the gradual development of (digital) capabilities (Verhoef et al., 2019), we explain how openness to ideas (Soto et al., 2011; Venkatesh & Bala, 2012) functions as a facilitator of learning. We find that openness to ideas positively moderates the relationship between digital transformation and firm digital performance. To the best of our knowledge, little is known so far in the literature about positive contingencies for digital transformation efforts except appointing chief digital officers (e.g. Singh & Hess, 2017). Compared to earlier studies, we provide empirical insights based on an evolutionary perspective under what conditions digital transformation efforts are more successful.
36
stream of literature by providing evidence for our earlier proposition that established firms should transform stepwise (Verhoef et al., 2019).
Managerial implications
Our study also offers relevant practical implications for managers and has broader societal value. Although digital transformation is considered to be among the top strategical threats (Kane et al., 2015), companies are not yet sufficiently able to react to the upcoming disruptions. In fact, most digital transformation efforts fail (Tabrizi et al., 2019). Our study can help managers to determine some of the factors driving success.
First, our study provides managers with a fresh perspective on the upcoming digital transformation journey by showing them that digital transformation works as an evolving process that builds on the gradual development of new knowledge. Our study reveals that managers should see a digital transformation not as a quick revolution in which they sprint to the goal. In contrast, we show that quickly changing the business harms performance. We provide timely guidance for managers that digital transformation can only be reached by taking small incremental steps and learn to use new technology rather than radically changing the business at once. Managers need to consider that digital transformation is driven by strategy (Kane et al., 2015) and such a strategy takes time to evolve and to generate value for the enterprise. Thus, digital transformation efforts should be planned over longer time horizons.
Second, we provide an actionable tool to measure digital transformation at the firm level and thus to determine the current state of digital transformation. Managers can use our tool to detect improvement areas for example for capabilities or business model change and take corrective action. The executive literature already paid a lot of attention as shown by multiple articles dealing with the process of digital transformation (Dery, Sebastian, & van der Meulen, 2017; Kane et al., 2015; Peppard, Edwards, & Lambert, 2011; Singh & Hess, 2017). Hence, our study does offer important practical insights for firms to assess and actively steer their digital transformation efforts.
