How to increase data-driven maturity : a review of the dynamic capability framework for organizational change

(1)

How to increase data-driven maturity:

a review of the dynamic capability framework for

organizational change

Alexa Binnendijk —

(2)

How to increase data-driven maturity:

a review of the dynamic capability framework for

organizational change

Name: Alexa Binnendijk Student number: 10379134 Date of submission: 27-01-2017 Institution: University of Amsterdam

Master’s Thesis: Msc. in Business Administration – International management track Thesis supervisor UvA: Dr. M. Paukku

(3)

Statement of originality

This document is written by Student Alexa Binnendijk who declares to take full responsibility

for the contents of this document. I declare that the text and the work presented in this document

is original and that no sources other than those mentioned in the text and its references have been

used in creating it. The Faculty of Economics and Business is responsible solely for the

(4)

Abstract

This study investigates how dynamic capabilities increase a company’s data driven maturity. The

objective of this study is to establish a newly explored field with more information.It will give

insights in the organizational dynamic capabilities and its foundations that are necessary to

increase a companies’ data-driven maturity.To answer the research question, semi-structured

interviews are held with consultants and data analytics employees of various companies. The

results of this study show that a company goes through organizational change corresponding to

the dynamic capabilities framework developed by Teece (2007). A company needs a sensing

capability to reach a low data-driven maturity level. To further develop to a medium data-driven

maturity level a company needs to develop a seizing capability. And finally to be able to reach a

high data-driven maturity level a company needs a transforming capability. For the sensing

capability, the microfoundations ‘awareness’ and ‘experimenting opportunities’ were mainly

important. For seizing capability, the microfoundations ‘structure of the organization’,

‘individual capabilities’ and ‘experience’ were mainly important. And last, for transforming

capability, ‘data governance’ and ‘training’ came forward as two most important

microfoundations. These results can help managers of companies, who want to increase their

(7)

1 Introduction

With the rise of the Internet in the 1970s and the adoption of the World Wide Web since the

1990s the quantity of data generated by business, science and government has increased

immensely (Chen, Chiang & Storey, 2012). According to Tan, Blake, Saleh & Dustdar (2013, p.

62) this increase in data is also known as the data deluge. It is estimated that more than four

zettabytes (or 1021 bytes) of digital data are generated each year (Tien, 2013). Figure 1 shows where some of these enormous amounts of data come from.

Figure 1. The continuously increasing big data. Reprinted from Big data: a survey. Mobile Networks and Applications, 19(2), 171-209. By Chen, M., Mao, S., & Liu, Y. (2014).

(8)

In today’s fast changing business environment, that is characterized by uncertainty and eroding

customer loyalty, new forms of technology and competition arise (Kiron, Shockley, Kruschwitz,

Finch & Haydock, 2011). Lavalle et al. (2010) state, that even though some executives remain

overwhelmed by “data deluge”, more and more leaders of large organizations have moved past

the “overwhelmed” phase and are already taking advantage of increased information richness

and analytics to gain a measurable competitive advantage.

Organizations are using different forms of analytics, such as dashboards, Business

Intelligence (BI), web analytics, Big Data analytics and so on to capture data and enhance the

value of information underlying decisions (Kiron et al., 2011). Organizational decision making is

in the midst of a fundamental shift from the reliance on a leader’s “gut instinct” to increasingly

data-based analytics (Brynjolfsson, Hitt, & Kim, 2011). According to Amir Orad, CEO of

Sisense, a BI software provider: “an organization where every person, who can use data to make

better decisions, has access to the data they need when they need it, is a data-driven organization.

Being data-driven is not about seeing a few canned reports at the beginning of every day or

week; it's about giving the business decision makers the power to explore data independently,

even if they're working with big or disparate data sources.” (Mitzner, 2016). Research shows that

data-driven organizations perform better than their industry peers (Kiron et al., 2011; McAfee,

Brynjolfsson, Davenport, Patil & Barton, 2012).

In 2011 about 58% of organizations were applying analytics to create competitive

advantage (Kiron et al., 2011). Part of this trend is due to the widespread diffusion of enterprise

information technology, such as Enterprise Resource Planning (ERP), Supply Chain

Management (SCM), and Customer Relationship Management (CRM) systems, which capture

(9)

operational data there are opportunities to collect data. Think of, for example, mobile phones,

factory automation, vehicles and other devices that are routinely instrumented to generate

streams of data on their activities. With the massive amounts of data now available, companies in

almost every industry, are focused on exploiting data for a competitive advantage (Provost &

Fawcett, 2013). It is questionable, however, to what extent organizations are able to survive,

grow and actually leverage data to a positive effect (Nair, 2014). Still very few companies have truly adopted data analytics. There are many ways to assess whether an organization is data

driven (Patil, 2011). According to Patil (2011, p. 2), some like to talk about how much data they

generate and others about the sophistication of the data they use. In this study we look at what

data-driven maturity level a company is.

In the literature the term maturity is described by Lahrmann, Marx, Winter, & Wortmann

(2011, p. 3) as a “state of being complete, perfect or ready. To reach this desired state of

maturity, an evolutionary transformation path from an initial to a target stage needs to be

progressed”. To get to this mature stage in today’s fast changing technological environment,

organizations must develop “dynamic capabilities” to create, extend and modify the ways in

which they make their living (Teece, Pisano and Shuen, 1997). Teece et al. (1997, p. 516) define

dynamic capabilities as: “the firm's ability to integrate, build and reconfigure internal and

external competences to address rapidly changing environments. Dynamic capabilities reflect an

organization's ability to achieve new and innovative forms of competitive advantage, given path

dependencies and market positions.”. Thus, this would mean that a company that is changing and

evolving into a more data-driven mature company needs dynamic capabilities to do this.

Although the concept of data & analytics is acknowledged in many organizations, no

(10)

Kumta and Shah (2002) the major problems are managerial, not technical. From a management

perspective this study could be of great value because it can help manager increase their

company’s data-driven maturity and data-driven organizations have many benefits. The use of

data analytics can support companies by stimulating efficiency, companies are better able to

provide valuable decision-making knowledge, which can result in fewer operating costs

(Hedgebeth, 2007). Analytics keeps a company updated, even though changes in the market can

occur at a very rapid pace, analytics gives insights for a more accurate forecast of market trends

(Hedgebeth, 2007). However, Kwon, Lee, & Shin (2014) find that many companies are still in

the early stage of the adoption curve when it comes to using data in, for example, their

decision-making. There is a lack of understanding, lack of experience and a lack of capabilities to deal

with data analytics (Kumta & Shah, 2002). This study will help managers focus on what

dynamic capabilities to develop when they want to increase their data-driven maturity.

This study can contribute to the literature a lot too. In the literature topics like: how to

monitor (Big) data (Rabl et al., 2012) and challenges with methodologies for data analysis and design (Kaisler et al., 2013; Zaslavsky, Perera, & Georgakopoulos, 2013) are recurring themes.

But something that has not been talked about much in the data analytics literature is the question:

how do dynamic capabilities increase a company’s data-driven maturity? As well as to the data

analytics literature, this study can also have a great contribution to the dynamic capabilities

literature. Ambrosini and Bowman (2009) state that, much more research is needed before we

can have a full understanding of what dynamic capabilities are and how they work. According to

them the priorities for the future would be to clarify some of the concepts that seem to be open to

differing interpretations, for example, scholars should be encouraged to look into integrating the

(11)

2009). This study will do just that, it will try to further develop the dynamic capabilities

framework in the field of organizational change. It will give insights in the organizational

dynamic capabilities and its foundations that are necessary to increase a companies’ data-driven

maturity. Also Helfat and Peteraf (2009) agree that the broad and integrative foundation of

dynamic capabilities provides a ready platform for further theoretical development from a variety

of perspectives.

Ten semi-structured interviews are conducted with employees of various companies. Half

the interviews were done with consultants and the other half with employees that work at various

companies at the data analytics department. This has led to five company cases, which are

combined into three cases, low driven maturity, medium driven maturity and high

data-driven maturity. The interviews with the consultants had a supportive role, this will be further

explained in the methods section.

In the next section the existing literature will be discussed in more detail. Subsequently

the theoretical foundation of this study will be discussed. Multiple working propositions are

established in order to investigate the research question. The study will continue by discussing the methodology of this qualitative research. In the results section the validity of the propositions

is presented. These results are discussed in the discussion section. And finally, this study will

conclude with a summary of the key findings, the research limitations and the recommendations

(12)

2 Literature review

2.1 Data-driven organization

As mentioned in the introduction, the rise of the Internet has increased the amount of data

generated by business, science and government immensely (Chen, Chiang & Storey, 2012).

Senior leaders are wondering whether they are getting full value from the massive amounts of

information they have within their companies (LaValle, Lesser, Shockley, Hopkins, &

Kruschwitz, 2011). This has sparked their interest in analytics (Liberatore & Luo, 2010). Many

leaders now aspire their company to become a data-driven organization. To be data-driven means

cultivating a mindset throughout the fabric of the business to continually use analytics to make

fact-based business decisions (Spotfire Blogging Team, 2015). Another similar definition used

by Patil (2011, p. 1) is: “A data-driven organization acquires, processes, and leverages data in a

timely fashion to create efficiencies, iterate on and develop new products and navigate the

competitive landscape”. McAfee et al. (2012) researched with a team at the MIT center for

digital business and a team of McKinsey’s business technology consultants, if data-driven

companies perform better than companies that are not data-driven. They concluded that

companies who figure out how to combine being a data-driven organization with domain

expertise will be able to gain a substantial competitive advantage (McAfee et al., 2012). This study will focus on how an organization can increase its data-driven maturity.

2.1.1 Data analytics

As mentioned, if a company wants to become data-driven it has to start using data analytics.

(13)

Subsequently, Evans and Lindner (2012) enrich this definition. They state that analytics is: “the

use of data, information technology, statistical analysis, quantitative methods and mathematical

or computer-based models to help managers gain improved insight about their business

operations and make better, fact-based decisions". The two definitions both support the statement

that Hedgebeth (2007) makes. He states that in today’s competitive and knowledge-based

economy, (advanced) analytics tools are used to assist firms with collecting, analyzing and

distributing information, so that managers are able to make informed decisions (Hedgebeth,

2007).

Analytics is often viewed from three major perspectives (Evans & Lindner, 2012). The

first perspective is descriptive analytics. Descriptive analytics is the most commonly used type of

analytics (Evans & Lidner, 2012) and as such tries to answer the question of “What happened

and/or what is happening?” (Delen & Demirkan, 2013). It summarizes data into meaningful

charts and reports, which allows managers to make queries to, for example, understand the

impact of an advertising campaign, find opportunities or problems or review business

performance (Evans & Lidner, 2012).

The second perspective is predictive analytics. Predictive analytics answers the question:

“What will happen and/or why will it happen?” (Delen & Demirkan, 2013). In this type of

analytics data and mathematical techniques are used to discover trends, associations, affinities

etc. between data inputs and outputs (Delen & Demirkan, 2013). Historical data is examined to

detect patterns or relationships and then extrapolate these relationships forward in time (Evans &

Lidner, 2012). Managers can use predictive analytics, for example, to predict risks or identify the

most profitable costumer. It can find relationships in the data that might not be apparent in

(14)

hidden patterns in large quantities of data, which is also known as Big Data analytics (Evans &

Lidner, 2012).

Finally, the third perspective is prescriptive analytics. According to Delen and Demirkan

(2013), in prescriptive analytics, data and mathematical algorithms are used to determine a set of

high-value alternative courses-of-action or decisions given a complex set of objectives,

requirements and constraints with the goal of improving business performance. Prescriptive

analytics tries to answer the question: “What should I do and why should I do it?” (Delen &

Demirkan, 2013). It deals with the optimization of business decisions. With prescriptive

analytics companies can answer questions like: “How much should we produce to maximize

profit?” or “What is the best way of shipping goods for our factories to minimize costs?” (Evans

& Lidner, 2012).

Within organizations, decision makers at all levels use these data analytics perspectives to

improve decision making and optimize business processes (Cosic, Shanks & Maynard, 2012).

Companies that implement data analytics as part of their business processes are achieving great

benefits (Cosic, Shanks & Maynard, 2012). Descriptive analytics is the most simple perspective

and prescriptive analytics is the most advanced perspective (Mesgarpour & Dickinson, 2014).

The more advanced the analytics perspective the higher the degree of intelligence and

complexity of the analytics perspective (Mesgarpour & Dickinson, 2014). This subsequently,

increases the value that can be gained. Companies should strive to gain value from higher levels

of analytics maturity by moving from information and hindsight (descriptive analytics) to

optimization and foresight (prescriptive analytics) (Mesgarpour & Dickinson, 2014). However,

(15)

2.1.2 Data science

The new wave of data analytics imposes new challenges (Rabl et al, 2012). The volume, variety,

velocity and ‘smartness’ of data have far surpassed the capacity of manual analysis and, in some

cases, have surpassed the capacity of conventional databases (Kitchin & McArdle, 2016; Haas &

Pentland, 2014; Provost and Fawcett, 2013). At the same time, computers have become much

more powerful, networking is universal (think of social network sites for example) and

algorithms have been developed that can connect datasets to enable deeper and broader analyses

than previously possible (Provost & Fawcett, 2013). Companies have realized they need to hire

more data scientists if they want to gain a competitive advantage. Data science is promoted

nowadays as a “sexy” career choice (Provost & Fawcett, 2013). But what is data science?

Provost & Fawcett (2013) argue that it is hard to illustrate exactly what data science is. First,

data science is complexly intertwined with other important concepts also of growing importance,

(16)

Figure 2. Data Science in the context of closely related processes. Reprinted from Data science and its relationship to big data and data-driven decision making. Big Data, 1(1), 51-59. By Provost, F., & Fawcett, T. (2013).

Second, there is the natural tendency to associate what a practitioner does with the definition of

the practitioner’s field, which can lead to overlooking the fundamentals of the field itself.

Provost and Fawcett (2013) believe that it is not of great importance to try to define the

boundaries of data science precisely. But broadly taken, data science involves techniques,

models, principles and processes to make sense of the (automated) data analyses (Provost &

Fawcett, 2013). In practice data science is defined slightly different in almost every article, blog

and company. In the companies (that were researched in this study) that have data scientists in

house, they often resort under the (advanced) data analytics department. In this study we use data

analytics as a unified term, which includes: descriptive, predictive, prescriptive analytics,

business intelligence and also data science.

2.1.3 Data-driven decision making

Tien (2013) argues that data is worthless from a decision-making perspective, unless the data is

processed or analyzed properly. When one knows how to deal with datapredictions and

-analyses, data can effectively help improve a company’s decision making (Provost & Fawcett,

2013). Data-driven decision-making, as mentioned earlier, refers to the practice of basing

decisions on the analysis of data rather than purely on intuition (Provost & Fawcett, 2013). For

example, a top manager could decide to open a new branch of a store at a specific location based

(17)

could base his/her location selection on the analysis of data regarding which location is

economically most interesting and will possibly have the highest success rate. She/he could also

use a combination of these approaches. Data-driven decision making is not an all-or-nothing

practice. Organizations engage in data-driven decision making to greater or lesser degrees

(Provost & Fawcett, 2013). McAfee et al. (2012) argue that when companies use data to base

their decisions on, instead of intuition, these data-driven decisions tend to be better decisions.

And because the decisions are based on evidence, they have the potential to revolutionize

management (McAfee et al, 2012). According to Tan et al (2013) there are two different types of

data-driven decisions: “(1) decisions for which ‘‘discoveries’’ need to be made within data, and

(2) decisions that repeat, especially on a massive scale, and so decision making can benefit from

even small increases in accuracy based on data analysis” (Tan et al, 2013).

Adding to that Watson, Wixom, Hoffer, Anderson-Lehman & Reynolds (2006, p. 1)

argue that: “Data management for decision support has moved through three generations, with

the latest being real-time data warehousing”. In most cases the value of data decreases rapidly,

because of the data’s potential for affecting tactical decision making and business processes the

latest generation is most significant (Watson et al., 2006). In Watson et al.’s (2006) research they

find that organizational changes must be made to enable this third generation of real-time data

for decision making. Companies that recognize the potential of real-time data-driven decision

making early on, may be able to arrange their organization in such a way that they will be able

facilitate a comfortable transition to real-time decision making (Watson et al., 2006). To be able

to make this change companies need organizational dynamic capabilities (Teece, Pisano and

Shuen, 1997). Summarizing, a more data mature company can gain more value from data-driven

(18)

making. This study will examine how dynamic capabilities influence a company’s data-driven

maturity at an organizational level. But, what are organizational dynamic capabilities? This will

be further explained in the next section.

2.2 Capabilities

As the value of data-driven decisions and the amount of data are growing, the complexity of

enterprise systems increases (Kiron et al., 2011; McAfee et al., 2012). Which, as mentioned,

means that the capabilities necessary for monitoring and analyzing such systems also grows

(Rabl et al, 2012). Some companies have already built sophisticated monitoring tools, which

allow them to, for example, trace individual transactions across geographically distributed

systems (Rabl et al, 2012). But what capabilities do companies need and how do these

capabilities increase a firms data driven maturity?

Existing literature typically discusses topics like what systems and models should be used

to store and analyze data sets (Rabl et al, 2012; Provost & Fawcett, 2013; Tien, 2013; Assunção

et al, 2015; Kraska, 2013; George, Haas & Pentland, 2014). But, as mentioned, the major

problems are managerial, not technical (Kumta & Shah, 2002). Even though there is not a lot of

literature that systematically explains how organizational dynamic capabilities can help a

company change and evolve, researchers do agree that dynamic capabilities can be seen as an

extension of the resource-based view (RBV) (Ambrosini & Bowman, 2009; Kwon, Lee & Shin,

2014; Teece, Pisano & Shuen, 1997). This section will start with shortly explaining the RBV,

continue with explaining what capabilities are and will finish with an elaboration on dynamic

(19)

2.2.1 Resources

According to the resource-based view, if a firm wants to achieve and sustain a competitive

advantage a firm needs to have resources (Barney, 1991). Resources are firm-specific assets that

are difficult, if not impossible to imitate (Teece, Pisano & Shuen, 1997). Examples of resources

are trade secrets and certain specialized production facilities. Such resources are difficult to

transfer among firms because of transaction costs and transfer costs and because the assets may

contain tacit knowledge. Tacit knowledge is knowledge that is difficult to transfer to another

person by means of writing it down or verbalizing it. In the literature researchers have theorized

that firms can achieve a competitive advantage when they possess resources that are: valuable,

rare, inimitable and non-substitutable (VRIN) (Teece, Pisano & Shuen, 1997). However, in the

information technology (IT) resources have become increasingly commoditized and often do not

satisfy the VRIN criteria (Cosic, Shanks & Maynard, 2012). But non-VRIN resources can be

combined with existing organizational resources to form VRIN resources (Cosic, Shanks &

Maynard, 2012). According to Cosic, Shanks and Maynard (2012) dynamic capabilities were

conceptualized in response to the criticism on IT resources. Dynamic capabilities focus on

‘resource renewal’: transforming and renewing resources into new organizational capabilities

(Teece et al. 1997).

2.2.2 Capabilities and dynamic capabilities

Resources form the basis of achieving and sustaining a competitive advantage (Eisenhardt &

Martin, 2000). Often forgotten is that a company also needs capabilities to be able to achieve a competitive advantage from their resources. (Teece & Pisano, 1994). Capabilities are usually

(20)

spend money on (Borwich, 2013). Strategic management’s key role in appropriately adopting,

integrating and transforming internal and external organizational skills, resources and functional

competences to match the requirements of a changing environment are emphasized by the term

‘capabilities’ (Teece, Pisano & Shuen, 1997). According to Teece (2014) an enterprise capability

is a set of current or potential activities that use the firm’s productive resources to make and/or

deliver products and services. There are two important classes of capabilities: ordinary and

dynamic (Teece, 2014).

Ordinary capabilities include the performance of administrative, operational, and

governance-related functions that are necessary to accomplish tasks (Teece, 2014). Winter

(2003) proposes organizational ordinary capabilities as: zero-level capabilities, which he defines

as those that permit the firm to earn a living in the present.

Subsequently Winter (2003) explains that there are also first-level capabilities. First level

capabilities modify and change zero-level capabilities. These capabilities are also referred to as

dynamic capabilities. Dynamic capabilities are the antecedent of organizational and strategic

routines by which managers alter their resource base. Looking at different definitions in the

literature it can be concluded that there is some consensus about the dynamic capability construct

(Ambrosini & Bowman, 2009). All the definitions reflect that dynamic capabilities are: in the

most general sense organizational processes, their role is to change the company’s resource base,

they are built instead of bought, they are path dependent and they are embedded in the firm

(Ambrosini & Bowman, 2009). Dynamic capabilities are the only perspective in the literature

that focuses on how companies can change their valuable resources over time (Ambrosini &

Bowman, 2009). Dynamic capabilities are the firm’s processes that use resources, also known as

(21)

competitive advantage (Teece et al., 1997; Eisenhardt & Martin, 2000). For some differences

between ordinary and dynamic capabilities see table 1.

Table 1. Some differences between ordinary and dynamic capabilities. The foundations of enterprise performance: Dynamic and ordinary capabilities in an (economic) theory of firms. The Academy of Management Perspectives, 28(4), 328-352. By Teece, D. J. (2014).

2.3 Change model

2.3.1 Dynamic capabilities to support change

As mentioned earlier, even though many companies are exploiting data for a competitive

advantage, still very few companies have truly adopted data analytics. As of 2010 many

organizations began to carefully examine the possibilities of using data analytics and started to

actively consider its adoption (Bughin, Chui & Manyika, 2010). As these new forms of

technology and competition arise, companies must adapt to and exploit changes while seeking

opportunities through technological and organizational innovation (Helfat et al., 2009). Markets

emerge, collide, split, evolve and die continually. To survive and grow under these conditions of

change, dynamic capabilities are the organizational and strategic processes by which firms can

(22)

capabilities are especially important in rapidly technological changing environments like the data

analytics environment (Teece et al, 1997). Teece et al. (1997) state that the dynamic capabilities

framework is used to analyze methods and sources of wealth creation by MNEs operating in

these rapidly technological changing environments. Dynamic capabilities are able to support

change because, managers can generate new value-creating strategies through acquiring and

shedding resources. They integrate resources together and recombine them again (Grant, 1996;

Pisano, 1994). Resources can lead to a competitive advantage but resources do not specifically

address how future valuable resources could be created or how the current stock of VRIN

resources can be refreshed in changing environments: this is the concern of the dynamic

capability perspective (Ambrosini & Bowman, 2009). How much value is created would depend

on how these resources are organized, or how they are combined within the company (Ambrosini

& Bowman, 2009). If an enterprise possesses resources but lacks dynamic capabilities, it has a

chance to make a competitive return for a short period, but it will not be able to sustain these

returns for a long time, except due to chance (Teece, 2007).

Dynamic capabilities can be broken down for applied purposes into three clusters:

sensing, seizing and transforming (Teece, 2007). Sensing is the identification, co-development,

development and assessment of technological opportunities within the company and in the

external market (Teece et al., 1997). Seizing is the mobilization of resources to address needs

and opportunities and to capture value from doing so. Finally, transforming is the continuous

renewal and recombining of resources. Engagement in continuous or semi-continuous sensing,

seizing and transforming is essential if the firm is to sustain itself as customers, competitors and

technologies change (Teece, 2007). This can be seen at the top part of figure 3. This dynamic

(23)

sensing, seizing and transforming capabilities in a technological changing environment. It starts

with sensing which is concerned with how a business enterprise and its management can first

spot the opportunity to earn economic profits. Then seizing which is concerned with making the

decisions and institute the disciplines to execute on that opportunity. And then transforming

which is concerned with continuously refreshing the foundations of its early success (Teece,

2007).

In his article Teece (2007) also developed the microfoundations which undergird

sensing, seizing and transforming capabilities. He defines microfoundations as distinct skills,

processes, procedures, organizational structures, decision rules and disciplines that are difficult

to develop and deploy. Teece (2007, p. 1319) attempts to identify the microfoundations of the

capabilities necessary to sustain superior enterprise performance in an open economy with rapid

innovation and globally dispersed sources of invention, innovation, and manufacturing

capability. Similar Kindström, Kowalkowski, and Sandberg (2013) tried to identify the key

microfoundations that are forming the basis of successful realignment of a firm’s dynamic

capabilities that are necessary to achieve a better fit with service innovation activities. This study

also tries to identify the microfoundations of dynamic capabilities but in another environment

then Kindström, Kowalkowski, and Sandberg (2013) and Teece (2007) have done previously.

Some of the microfoundations Teece (2007) has established will be tested to see if they are also

relevant in a fast changing technological environment where a company is trying to increase its

data-drive maturity. Furthermore, possibly additional microfoundations will be explored. In the

theoretical framework, the relevant microfoundation will be defined and supported by literature.

(24)

Figure 3. Foundations of dynamic capabilities and business performance. Explicating dynamic capabilities: the nature and microfoundations of (sustainable) enterprise performance. Strategic management journal, 28(13), 1319-1350. By Teece D. J. (2007)

2.4 Data-driven maturity

2.4.1 Maturity model

In this study we use a maturity model to look at organizational change. Many maturity models

embrace the notion that successful organizational change can be triggered and/or achieved by an

assessment of practices (Maier, Moultrie & Clarkson, 2012). To reach earlier mentioned

Lahrmann, et al.’s (2011) state of being complete, perfect or ready an evolutionary

transformation path from an initial to a target stage needs to be progressed. Maturity models are

(25)

(2015) also states that the main goal of a maturity model is to provide a capability assessment

tool that focuses on specific data key areas in organizations, to help guide development and to

avoid pitfalls. Hillson (2003) adds to that, that the current position of an company can be

assessed objectively in a maturity model because of a number of defined levels of capability and

it defines the next level of capability which a company can aspire.

According to Halper and Krishnan (2013, p. 5) Big Data maturity can be defined as “the

evolution of an organization to integrate, manage, and leverage all relevant internal and external

data sources”. As Big Data is part of data analytics, in this study this definition is used for the

data-driven maturity as well. An assumption made in the literature is that a high level of (Big)

Data maturity regarding their data-driven decisions correlates to the increase of top-line revenues

and reduction of operational expenses (El-Darwiche et al. 2014; McAfee et al. 2012). Paulk et al.

(1993) also found in their research that as maturity increases, the difference between targeted

results and actual results decreases across projects. An example they give in their research: “if

ten projects of the same size were targeted to be delivered on May 1, then the average date of

their delivery would move closer to May 1 as the organization matures. Level 1 organizations

often miss their originally scheduled delivery dates by a wide margin, whereas Level 5

organizations should be able to meet targeted dates with significant accuracy” (Paulk et al., 1993,

p. 22). This is because mature companies are carefully constructed (Paulk et al., 1993).

Paulk et al. (1993) state that, maturity is based on many small, evolutionary steps rather

than revolutionary innovations. As mentioned in the previous section according to Teece (2007)

sensing, seizing and transforming capabilities are steps a company needs to go through to be able

to sustain itself in a fast changing environment. To capture the evolutionary steps of maturity

(26)

developed consists of five maturity levels, which provide organizations with guidance on how to

gain control of their processes for developing and maintaining software and how to evolve

towards a culture of software engineering and management excellence (Paulk et al., 1993).

In this study the CMM in combination with the dynamic capabilities framework is chosen

as a theoretical foundation to explain how a company’s organizational changes can lead to a

higher data-driven maturity. There are three reasons why the CMM is chosen. First, the model

can be related to data-driven maturity, because instead of software that has to be developed and

maintained, in the case of the data-driven maturity model, data analytics has to be developed and

maintained. Software and data analytics are both fairly new concepts, that are both IT-related.

Secondly the CMM is chosen, because it deals with capabilities and its underlying processes.

This study also deals with capabilities and its underlying processes. And thirdly, in the literature

maturity models are often derived from CMM because it is generally acknowledged and

recognized, and it is one of the few maturity models that is well documented (Hribar Rajterič, 2010). Cosic, Shanks and Maynard (2012) studied Business analytics maturity models in their

research-in-progress paper. They also found that, in general, BA maturity models lack strong

theoretical foundations and focus too much on data warehousing.

Cosic, Shanks and Maynard (2012) also mention that maturity models can have several

purposes, they can be descriptive, prescriptive and comparative. A descriptive maturity model

assesses the maturity situation within an company as it is at that moment. A prescriptive model

includes, in addition to the as-is situation, guidelines for improving maturity at each level, and

allows companies to identify desirable future levels of maturity. And last, a comparative maturity

model is a prescriptive model that has been used in a large number of companies so that

(27)

CMM is a prescriptive maturity model (Cosic, Shanks & Maynard, 2012). Figure 4 shows the

five stages of the CMM.

Figure 4. The five levels of the Capabilities Maturity Model. Reprinted from Key practices of the Capability Maturity Model version 1.1. By Paulk et al., (1993).

The literature and the conversations at KPMG show that there is no very thorough

research done on dynamic capabilities in combination with the fast changing technological data

analytics environment. Kwon, Lee, and Shin (2014) suggest, that with the growing potential of

Big Data, organizational variables and other conditions important when wanting to use data,

have to be further identified. The lack of research and the interest showed by the business in this

topic have led to the following research question of this study:

(28)

2.4.2 Consultants role

Huge developments in information technology have led to a central role for consulting

companies (Sarvary, 1999). It used to be the case that consulting companies were providing

clients with a resource namely ‘smart people’ who could solve the clients problem. But clients

started hiring these ‘smart people’ themselves, explicitly they started hiring MBAs from top

business schools (Sarvary, 1999). Finding ‘smart people’ was not a problem anymore, instead the

clients wanted to benefit from consulting company’s broad experience, expertise and knowledge.

Companies that want to increase their data-driven maturity often rely on the knowledge and

experience of consulting firms. As mentioned in increasing data-driven maturity, technology is

often not the problem but the problems are more managerial (Kumta & Shah, 2002). This is why

companies who want to increase their data-driven maturity often turn to management consultants

for help. Following the theory of the dynamic capability framework that is used for

organizational change in this study. Consultants could be of help at every step of change, they

can help develop sensing, seizing and transforming capabilities. The level of data-driven

maturity a company already has, defines what type of dynamic capabilities the consultants

should focus on. Consultants have a supporting role (Sarvary, 1999).

KPMG is one of those companies with a lot of experience and knowledge in management

consulting, they have specific departments for IT advisory. A topic KPMG IT advisory advises

on is data analytics and data-driven maturity. I was able to write my thesis at KPMG’s

management consulting IT advisory department which gave me access to their knowledge and

best practices. Because I wrote my thesis at KPMG, I’ve spent approximately 560 hours at their

offices where besides working on my thesis, I observed, talked to many employees, and as part

(29)

KPMG is fully aware of the possibilities data analytics has for companies, and the

importance for companies to increase data-driven maturity. They have developed a data-driven

maturity scan, which is a questionnaire that assesses at what level of data-driven maturity a

company is at the moment. Along with the driven maturity scan they developed a

data-driven maturity model. See figure 5. The data-data-driven maturity model is used as a tool in this

study in combination with Paulk et al.’s (1993) CMM to set up propositions and define the

microfoundations necessary to increase data-driven maturity.

Figure 5. The five levels of the Data-driven Maturity Model. Reprinted from Data Driven maturity scan By KPMG (2016). By KPMG IT Advisory

(30)

3 Theoretical framework

3.1 Dynamic capabilities & data-driven maturity for change

In this chapter the dynamic capabilities framework will be linked to Paulk et al.’s (1993) CMM

and to KPMG’s data-driven maturity model. As mentioned, some of the microfoundations that

Teece (2007) has defined will be further developed and possible new ones will be explored. The

CMM levels will be linked to KPMG’s data-driven maturity levels, and according to those the

possibly relevant microfoundations will be set up. A visual summary of the information

mentioned in this section can be found in appendix 1.

KPMG’s first level is the awareness level, which can be related to Paulk et al’s (1993)

CMM initial level. Paulk et al. (1993) states that in the initial level, the organization does not

provide a stable environment for developing and maintaining software. In KPMG’s awareness

level, instead of the absence of a stable environment for developing and maintaining software

they talk about data analytics. In both maturity models the first level is mainly related to creating

awareness. No dedicated expertise, platform and processes related to data analytics are in place

yet. As mentioned, according to Kumta and Shah (2002) the major problems are managerial, not

technical. This is in line with the sensing capabilities, which is also concerned with managerial

capabilities and not with technical capabilities. As mentioned in the literature review, sensing

capabilities are concerned with identification, co-development, development and assessment of

technological opportunities within the company and in the external market (Teece et al., 1997).

Awareness is an important aspect of sensing capabilities (Teece, 2007).

KPMG’s second level is the experimental phase. This phase is related to the second level

(31)

establish effective project management. To be able to repeat best practices of earlier successful

projects they are required to be documented adequately. The focus is primarily on projects

(Kumta & Shah, 2002). This is similar to KPMG’s experimental phase, in which a company uses

cases and hypotheses that are tested through pilots. During these pilots, expertise is built and a

dedicated platform is made available. The data analytics initiatives are, however, ad hoc and per

project. An organization that has a couple of projects going is usually on level two. In this

entrepreneurial phase, in agreement with the dynamic capabilities change framework, the sensing

capabilities seem to be mainly important. Entrepreneurship is about sensing and understanding

opportunities, getting things started, and finding new and better ways of putting things together

(Teece, 2007). In specific, similar to Teece’s (2007) microfoundations the experimenting and

R&D aspect are important. Experimenting is a process through which a company can sense

possible new opportunities, to be able to do this there needs to be a budget for R&D. According

to the literature and to KPMG, the microfoundations of the sensing capability are ‘awareness,

‘experimenting opportunities’ and ‘budget for R&D’. If a company is in the first or second

maturity stage this company has a ‘low data-driven maturity’. These first two levels have led to

the following working proposition:

WP1: A company’s data-driven maturity can only start to increase through a sensing capability.

The third level defined by KPMG is the cohesiveness level. In Paulk et al’s (1993) CMM

this level is called the defined level. At their defined level, the emphasis shifts to the

organization. Best practices are gathered across the organization and organization standard

(32)

by establishing common processes and measurements. The process capability is based on a

common understanding of the activities, roles and responsibilities. At this level, capacities have

been defined and are collected systematically (Kumta & Shah, 2002). This is similar to KPMG’s

cohesiveness level in which, according to KPMG, an organization coordinates projects through a

program and uses a structured yet agile approach. According to the dynamic capabilities

framework in this stage developing a seizing capability is mainly important (Teece, 2007). As

mentioned in the literature review, seizing is the mobilization of resources to address needs and

opportunities and to capture value from doing so. Once a new technological opportunity is

sensed, it can be addressed through new products, processes or services. According to Teece

(2007) this involves deciding which business model to build. According to the KPMG model for

data-driven maturity it is mostly important that this business model is a good fit for an agile

approach. That is why in this study, the first microfoundation of a seizing capability is ‘the

ability to work agile’. Besides building a business model, Paulk et al. (1993) mentions that in this

third phase, there is a clear understanding of roles and responsibilities. To achieve this a

company has to have the skills and be able build loyalty and commitment (Teece, 2007). With

the help of the information from the literature and exploratory conversations the

microfoundations ‘experience’ and ‘individual capabilities’ were set up. Individual capabilities

are important because the right capabilities help assure clear roles and responsibilities across the

business and a well-defined prioritization and execution model ensure minimal task duplication.

Experience was mentioned in the exploratory conversations to be very important because of

multiple reasons, it increased the ability to gain value from sensed opportunities, but it also

increases the commitment and loyalty in a firm. The third phase is defined in this study as the

(33)

WP2: A company’s data-driven maturity can increase to a medium data-driven maturity through sensing and seizing capabilities.

KPMG’s fourth level is the business driven level. Paulk et al. (1993) define the fourth

level as the managed level. Both maturity models are very similar. According to Kumta and Shah

(2002), in the managed maturity level of the CMM decisions are based on data collected. The

process performance and the project progress are controlled quantitatively. All organizational

processes are mapped to a common measurement and assessed using a base line. According to

KPMG, in the fourth maturity level data analytics is an increasing part of the strategy for

business units with maturing guidelines and processes for handling supply and demand. Valuable

analyses, enriched data and development resources are recycled and shared among units, which

is also mentioned by Kumta and Shah (2002) as controlled quantitatively. In this level an

organization has data scientists and business owners working together.

Finally, the highest level and fifth maturity level is defined by KPMG as the embedded

level. In this level most operational decisions are driven by near real-time data and analytics.

Rich data is able to be shared across business units through a self-service model and analytics is

an integral part of all business units operations, thus allowing the company to continuously

improve. Depending on the organization, the focus may be heavily customer centric and an

ecosystem of suppliers and downstream data users share added value, created through advanced

analytics. Paulk et al. (1993) defines this final level as the optimizing level, in which continuous

process improvement is the main focus. The organization is constantly trying to prevent the

(34)

analytics. In mature organizations, everyone is responsible for process improvement at any given

point in time. Continuous process improvement means controlled change and a measured

improvement in process capability. Transforming capabilities are mainly important in these two

phases. Transforming is, as mentioned, the continuous renewal and recombining of resources.

The transforming capability is according to Teece (2007) concerned with ‘data governance’ and

‘training’, which are both mentioned as important in this stage. When looking at the transforming

capabilities, the ‘knowledge management’ microfoundation is important as well. As mentioned,

organizational processes are mapped to a common measurement and assessed using a base line.

In order to be able to create this base line and make everyone benefit from the knowledge,

knowledge management has to be good. Companies in the fourth and fifth level are defined in

this study as companies with a high data-driven maturity.

WP3: A company’s data-driven maturity can increase to a high data-driven maturity level through sensing, seizing and transforming capabilities

The dynamic capabilities sensing, seizing and transforming and their microfoundations are all

interrelated (Teece, 2007). According to Teece (2007) successful companies must build and

utilize all three classes of capabilities and employ them.

3.2 Conceptual framework

The propositions for this study are set up through a combination of literature and practical

expertise, namely: the CMM, the dynamic capabilities framework, exploratory conversations at

(35)

propositions that are studied in this research:

RQ: How do organizational dynamic capabilities increase a company’s data-driven maturity? WP1: A company’s data-driven maturity can only start to increase through a sensing capability. WP2: A company’s data-driven maturity can increase to a medium data-driven maturity through sensing and seizing capabilities.

WP3: a company’s data-driven maturity can increase to a high data-driven maturity level through sensing, seizing and transforming capabilities

In addition, the microfoundations underlying the dynamic capabilities are also defined

through the literature and KPMG’s expertise. For sensing capabilities the microfoundations:

“awareness”, “R&D budget” and “experimenting opportunities” are developed. For seizing

capabilities the microfoundations: “agile working”, “experience” and “individual capabilities”

are developed. And finally for transforming: “data governance”, “knowledge management” and

“trainings” are developed.

The conceptual framework as shown in figure 6 is a reflection of the research question

and the working propositions. The conceptual model shows the dynamic capabilities, sensing,

seizing and transforming, that are necessary for a company to be able to change into a more

data-driven mature company. This study researches how these dynamic capabilities exactly influence

(36)

Figure 6. Conceptual model

4 Research methodology

4.1 Research approach

This section discusses the methodology of this study. It is an descriptive study, with the objective

to establish a newly explored field with more information (Kowalczyk, 2014). As mentioned,

concepts of data & analytics are acknowledged in many organizations, but no dedicated

expertise, platform and processes are in place for a structured approach. Questions are being

raised in the literature about the major problems in this field being managerial, instead of

technical. This study attempts to explore and explain the data analytics field further while

providing it with more information. A descriptive study is effective when analyzing

non-quantified topics and issues (Kowalczyk, 2014). It looks at the ‘how’ and the ‘what’ rather than

the ‘why’ (Yin, 1984; Sandelowski, 2000). The research question in this study is: “How do

dynamic capabilities increase a company’s data-driven maturity?”. The choice is made for a

(37)

Furthermore, rather than following the deductive or inductive approach, this research

process had, a more abductive approach. The abductive approach has elements of both the

deductive and inductive approach. A general theory (dynamic capabilities framework) is being

developed further (Dubois & Gadde, 2002). According to Teece (2007) the dynamic capabilities

framework is not yet a grounded theory, more research is needed. Working propositions are set

up to help develop the dynamic capabilities framework. The working propositions are tested

through multiple case studies. According to Eisenhardt (1989, p. 543) a case study is: “a research

strategy which focuses on understanding the dynamics present within single settings”. A case

study can be singular but it can also include numerous cases, and multiple levels of analysis

(Yin, 1984). In the next section the research philosophy will be discussed.

4.1.1 Ontology, epistemology and methodology

There is an ongoing debate, for many years now, regarding the suitability of research methods in

the social sciences (Morgan & Smircich, 1980). In the 1960s and 1970s organizational research

was dominated by the use of quantitative methods. Only later qualitative methods started to

become more and more in favor (Morgan & Smircich, 1980). In the beginning of the 1980s there

was a so called “paradigm war” between quantitative and qualitative research (Soini, Kronqvist,

& Huber, 2011). But what is a research paradigm? In the literature multiple slightly different

definitions for a paradigm are used. In this study we use the definition of Kuhn (1962). He states:

“a paradigm is a set of common beliefs and agreements shared between scientists about how

problems should be understood and addressed”. A paradigm is characterized by the following

(38)

Scotland (2012) describes ontology as the study of being. It is related to the nature of

reality. Researchers need to take a position regarding their perceptions of how things are and

how things work (Scotland, 2012). They should realize that there might be multiple realities and

that they can explore these realties through multiple forms of evidence from different

individuals’ perspectives and experiences. The question they try to figure out is: ‘What is

reality?’. Epistemology is concerned with the question: ‘How do you know something?’.

According to Scotland (2012) epistemological assumptions are about how knowledge can be

created, acquired and communicated. The last characteristic, methodology, deals with the

question: ‘How do you go about finding it out?’. In other words, methodology is concerned with

why, what, from where and how is the data collected and analyzed (Scotland, 2012).

This study uses, as mentioned, a qualitative methodology. A qualitative methodology

differs from a quantitative methodology in various ways (Bryman, 1984). Qualitative research is

believed to be more flexible than quantitative research, in that it emphasizes discovering

unexpected findings and the possibility of altering research plans. In contrast, a quantitative

method emphasizes hypothesis testing, fixed measures and has a less protracted form of

fieldwork involvement (Bryman, 1984). In a qualitative approach there is a commitment to

seeing the social world from the actors point of view. Because of this, there is a preference for a

contextual understanding so that behavior, for example, is understood in the context of systems

employed by a particular group or society (Bryman, 1984). During the 1990s the view that

quantitative and qualitative methods are each tied to a specific set of ontological and

epistemological assumptions gained broad acceptance (Soini, Kronqvist & Huber, 2011).

(39)

that qualitative research can be informed by a variety of epistemological views instead of just a

specific set of ontological and epistemological assumptions.

In the literature two paradigms are mainly used. The first paradigm is the positivism

paradigm, this paradigm states that the ontological position is one of realism. There is a single

reality. The positivist epistemology is one of objectivism. Phenomena have an independent

existence, which can be discovered via research (Crotty, 1998). Positivist methodologies usually

have a deductive approach and are directed at explaining relationships through, for example,

closed ended questions (Scotland, 2012). The one aspect from the positivist paradigm that has

some similarity with this study is the fact that this research starts with a theory, but this theory is

developed and explored further in this study.

The second paradigm is the interpretive paradigm. In the interpretive paradigm the

ontological position is relativism, which means that reality is subjective and differs from person

to person (Scotland, 2012). The interpretive epistemology is subjectivism. The world does not

exist independently of our knowledge. It is based on real world phenomena. Examples of

interpretive methodology are: case studies, phenomenology, hermeneutics and ethnography

(Scotland, 2012; Creswell, 2013). The interpretive paradigm is mainly in line with this study

because, through semi-structured interviews with multiple and different interviewees (different

points of view are taken; the company perspective and the consultants perspective), multiple

cases are compared to look at the different realities of the individuals that are interviewed. Also

the survey that is taken to measure at what level of data-driven maturity a company is leaves

room for interpretation. Depending on, for example, the company size, sector and industry,

(40)

4.2 Research design

4.2.1 KPMG data-driven maturity model

To be able to further develop the dynamic capabilities framework and thus, answer the research

question KPMG’s data-driven maturity model is used as a tool. KPMG has used its experience

and best practices to develop the data-driven maturity model. In this study the model is used to

help explain how the dynamic capabilities influence a company’s data-driven maturity. In the

literature no data-driven maturity model exists yet, which is why the model was linked to the

CMM to show the similarities and overlap with a theoretically acknowledged maturity model. As

mentioned, along with the maturity model, KPMG developed a data-driven maturity scan. A

questionnaire that measures at what level of data-driven maturity a company is. KPMG’s

maturity model and scan were both developed prior to my internship.

4.2.2 Multiple case study

As mentioned earlier, in this study the choice is made for a multiple case study. This approach

suits best because, a multiple case study enables the researcher to explore differences and

similarities within and between cases (Eisenhardt, 1989; Yin, 2003; Scotland, 2012). This study

explores which dynamic capabilities and which microfoundation of dynamic capabilities are

present in which level of data-driven maturity and which ones are necessary to increase the

company’s data-driven maturity. Since the epistemology, as mentioned, in this study is

subjectivism, companies from different industries might have different realities and

interpretations of data-driven maturity. With a multiple case study the similarities and

(41)

and procedures a case study can withstand, and still generate the same findings the more external

validity there is (Soy, 2015).

4.2.3 Case selection

This multiple case study consists of five company cases that are, as mentioned, combined to

three cases: low data-driven maturity, medium data-driven maturity and high data-driven

maturity. The cases are set up through 10 semi-structured interviews with individuals at different

companies. The cases were selected with the help of KPMG’s and my own network. Through my

unique negotiated access to KPMG I was able to select cases that I wouldn’t have been able to

select without KPMG. As mentioned KPMG’s network was especially useful because KPMG

gives advice to data analytics departments at a large variety of companies, and because KPMG is

one of the leaders in the data analytics field they have a good overview of what is going on in the

market. Another helpful aspect of KPMG’s network is that it is common that former employees

of KPMG transfer to data analytics departments of large companies. They are very sought after

by large companies because they have a lot of experience and knowledge in the data analytics

field.

The interviewees that represent the cases were selected through snowball sampling

(Browne, 2005). This is a commonly used method when investigating hard to reach groups.

Existing subjects (in this case, for example, the KPMG consultants) were asked to reach into

their networks and nominate other subjects. This way the sample grows just like a rolling

snowball does. For this study the goal was to get insight into a variety of companies’

organizational capabilities and their data-driven maturity. To be able to get this insight, the

(42)

staff of a large firm will probably not be very valuable for my study. So with the help of KPMG

consultants, the selected target group were employees that work at data analytics departments

and IT consultants that give advice to data analytics departments. Most of these people have full

schedules and do not have a lot of time and/or interest in doing an interview. That is why

snowball sampling was the best strategy.

The data analytics employees were chosen because they can explain what dynamic

capabilities they have at the moment and what they have learned from previous increases in

data-driven maturity (if some data-data-driven maturity is already achieved at least).

IT consultants were selected because of their experience. They have seen many

companies with data analytics departments and/or companies that are trying to set up data

analytics activities. Whenever help is needed by a company that wants to increase its data-driven

maturity, they can reach out to IT consultants, being the experts in the business. Besides

consultants at KPMG, I also interviewed consultants at two other large consulting companies, to

decrease the chance of one biased KPMG view. This is a form of triangulation. According to Jick

(1979) triangulation is: ‘the combination of methodologies in the study of the same

phenomenon’. I did not use two different methodologies but, according to Jick (1979)

organizational research triangulation is also often used to improve accuracy of their judgements

by collecting different kinds of data bearing on the same phenomenon. This is what is done in

this study, employees at data analytics departments from companies that were at different stages

of data-driven maturity and IT consultants from three different firms were interviewed.

The cases had to meet a couple of important criteria: 1) the company had to have a data

analytics department or at least a department that works with the basics of data analytics; 2) the

(43)

companies at different stages of data-driven maturity. This was done to make sure that low,

medium and high data-driven mature companies’ view was covered; 4) and an employee at their

data analytics department had to be willing to sit down for an interview of 45 minutes in the

month of November or the beginning of December. No distinction was made between industries.

4.3 Data collection

4.3.1 Semi-structured interviews

Data collection was conducted between the first week of November 2016 and the first week of

December 2016. The data was collected through semi-structured interviews, that were

constructed based on the literature review. Before the interview took place, the data analytics

employees were asked to fill out KPMG’s data-driven scan to determine at what stage they were.

Two interviews where done through Skype. The remaining eight were conducted face to face.

The answers to the data-driven scan helped to define the semi-structured interview questions.

The interviews with the consultants had a slightly different approach. The same questions were

posed but they were altered slightly. Instead of finding out what dynamic capabilities their

company has, they were asked about their expert opinion on which dynamic capabilities a

company should have to increase its data-driven maturity. The interviews with Dutch native

speaking interviewees were done in Dutch, because this allowed them to speak more freely, and

prevented a language barrier. The other interviews were done in English. The quotes of Dutch

interviewees are translated to English in this study. The interview questions generated primary

qualitative data.

Six of the interviews were done with data analytics employees of companies, that are, as

(44)

were done with consultants of three different large consultant companies, which all give advice

to data analytics departments. It was agreed with the interviewees, that their names would not be

disclosed in this study. See table 2. It is important to note, that because I wrote my thesis at

KPMG, I have spent an extra 560 hours at their offices. Besides working on my thesis, I thus had

the opportunity to observe and talk to many employees. Being part of the Enterprise solutions

Talent Pool, I furthermore was able to follow multiple trainings.

Table 2. Overview of interviewees, industry, job role, company size and data-drive maturity.

Company Industry Employees # Job role Data-driven

maturity

Airline company Transport logistics 20,000 + Big data program manager 4

Airline company Transport logistics 20,000 + Data officer 4

Governmental institution Government 1000 - 5000 Data analyst 1,6

Financial services company Financial services 20,000 + Data physicist/scientist 4,5 International sportswear (Online) retail 1000 - 5000 Manager supply chain info & analytics 3

Consulting firm 2 Consultancy 20,000 + Manager advanced analytics 4,1

Consultancy Industry Employees # Job role

Consulting firm 1 Consultancy 20,000 + Manager data insights and analytics Consulting firm 1 Consultancy 20,000 + Senior consultant technology advisory Consulting firm 3 Consultancy 20,000 + Advanced analytics senior consultant Consulting firm 2 Consultancy 20,000 + Manager advanced analytics

4.4 Data analysis

All the interviews were recorded with permission and transcribed afterwards. The transcripts

were systematically analyzed and coded. Computerized data analysis was done to identify

emerging patterns or common themes. The computer program Atlas.ti was used. For the coding a

combination of a deductive and inductive approach was taken. Sensing capabilities, seizing

capabilities and transforming capabilities were used as “starter codes”. Some of the underlying

How to increase data-driven maturity : a review of the dynamic capability framework for organizational change

How to increase data-driven maturity:

a review of the dynamic capability framework for

organizational change

How to increase data-driven maturity:

a review of the dynamic capability framework for

organizational change

Contents

Abstract

1 Introduction

2 Literature review

3 Theoretical framework

4 Research methodology