Digital business model innovation : a multiple case study

(1)

1

MASTER’S THESIS

Digital Business Model Innovation: A multiple case study

MSc in Business Administration: Digital Business Track

Student: Daniel Oldach

Student Number: 11372087

Date of Submission: 23rd of June 2017

Version: Final Version

Institution: Universiteit van Amsterdam

Supervisor: Carsten Gelhard

(2)

2

Statement of originality

This document is written by Student Daniel Oldach 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 supervision of completion of the work, not for the contents.

(3)

3

List of Figures

Figure 1 – Activity System Perspective Framework... 10

Figure 2 – Blockchain’s impact on supply chains ... 18

Figure 3 – Activity System Perspective combined with NICE framework ... 20

Figure 4 – Proposition derived from literature review ... 23

Figure 5 – Data overview for Internet of Things cases ... 29

Figure 6 – Data overview for Cloud Computing cases ... 32

Figure 7 – Data overview for Blockchain cases... 34

Abstract

This thesis paper investigates the effect of three technologies; internet of things, cloud computing and blockchain; on firms’ business models. Guided by the business model framework by Zott & Amit (2010), propositions are derived from literature which describe the expected impact of each technology. Following the approach of a directed content analysis, the documents which are collected through the means of companies’ websites and databases, are analyzed. The dataset comprises of nine companies in total, three per technology. Findings are reported descriptively per technology and the discussion of the findings of each case individually is performed. The holistic impact of each technology is discussed and the propositions are analyzed based on the findings. Theoretical and practical implications of the findings are presented. The value driver efficiency is the most affected part of the business model, with each technology affecting it. Other elements impacted are the value driver complementarities and novelty as well as the structure of the activity system. Finally, directions for future research are delineated, building on the limitations of this study.

(5)

5

Introduction

This paper is designed to research the effect of three technologies on business models. New, emerging technologies are redefining the way whole industries operate nowadays. Three technologies in particular are having a major impact on multiple industries and the way companies do business. These are, specifically, the Internet of Things (IoT), cloud computing and Blockchain.

“The Internet of Things (IoT) is the network of physical objects that contain embedded technology to communicate and sense or interact with their internal states or the external environment” (Gartner.com, 2017). These kinds of connected devices can be used for a wide variety of purposes, such as in smart homes. Here lights, thermostats and locks for example can all become connected, enabling remote access, control and maintenance for the user. The scale of implication possibilities is only bound by imagination. From the small, wearable smart device on your wrist to whole smart cities.

“Cloud computing, often referred to as simply “the cloud,” is the delivery of on-demand computing resources—everything from applications to data centers—over the internet”

(IBM.com, 2017). This entails the different services, be it Software as a Service (SaaS), Platform as a Service (PaaS) or Infrastructure as a Service (IaaS). These enable the user to take advantage of applications, resources to develop applications and storage, all while not owning any

hardware. The services are accessed via the internet, and can be accessed from any device anywhere.

Blockchain is the technology behind cryptocurrencies, most famously Bitcoin. It is part of a peer-to-peer system and is mainly a database in which the records, called blocks, of all

(6)

6 transactions are stored. Blockchain is an interesting technology when considering the fact that it has the ability to become a mode of decentralization, in a multitude of areas, more so than only economic. Possible application areas in the realm of businesses can be smart contracts,

international payments and the improvement of relational databases (Medium.com).

All three of those were, or in the case of Blockchain, are depicted on the Gartner Technology Hype Cycle (Forbes.com, 2016). Technologies are positioned on the Hype cycle based on their maturity, implementation and societal impact. IoT and cloud computing can therefore be considered mature and valuable regarding their potential impact on society and business, while blockchain is still positioned at the Peak of Inflated Expectations (Gartner’s hype cycle, 2016). Evidently, in 2015, 15 billion objects in the IoT world existed and by 2020 it is projected to reach 200 billion, making it an average of 26 devices per human on earth. More importantly, over 40% of those objects are used for Business and Manufacturing purposes, adding value in terms of decreasing costs, increased efficiency and improved inventory and maintenance management (Intel.com, 2017). Simultaneously, cloud computing is expected to be another grand opportunity. “In 2016, spending on public cloud Infrastructure as a Service hardware and software is forecast to reach $38B, growing to $173B in 2026” (Forbes.com, 2016).

Blockchain’s impact on the other hand is not yet fully known. It’s varied application

possibilities, the hesitance of incumbents to implement it and its rather immature nature, make the technology difficult to evaluate. This however only refers to quantitative impact

measurements. Ernst & Young for example, urges all businesses to think about the technology and “companies should identify new blockchain-enabled opportunities”, before “it will be too late” (EY.com, 2017).

(7)

7 Nonetheless, all three technologies possess the potential to revolutionize whole industries, if adopted correctly into a business model. Companies such as Amazon, Uber, Skype and Google have gained immense success and have established themselves not only as pioneers but also as incumbents in a rather short matter of time. Their way to success has in no doubt found origin in the way that these companies were able to leverage new technologies and incorporated them into their business models.

With the emergence of these technologies, it becomes interesting, in which manner other firms will be able to leverage the benefits. This study will therefore analyze the business models of three cases per technology in order to see which elements of the respective business models are important, more dominantly defined and valued, compared to the other business models. As an expected outcome, stereotypes for each technology will be constructed, indicating which parts of the business model are particularly influenced by these technologies and what benefits this carries. Based on these circumstances, a research question can be synthesized as follows: “How do IoT, Cloud Computing and Blockchain impact Digital Business Models and which elements of the business models are characteristically impacted by each technology?”

Literature Review

Business Model Innovation

The literature around the topic of Business Model Innovation is quite recent and broadly discussed. The theoretical concept of a Business Model is still vaguely and abstractly defined. Over the past decade, the field has grown in importance and relevance, for practitioners as well

(8)

8 as scholars. The research in this area is still diverse and some many schools of thoughts exist as to how to conceptualize the theoretical construct (Gassmann et. al, 2016).

The relevance of the field has increased due to continuous innovation in the field of Business Models. Companies like Amazon, Skype and Uber have radically changed the way of doing business in their industries. Business Model Innovation has become a new way to achieve a sustainable competitive advantage. In 2002, Chesbrough and Rosenbloom were among the first to introduce a concrete classification of a business model. According to them, a business model comprises of six functions: value proposition, market segment, value chain, cost structure and profit potential, value network and competitive strategy (Chesbrough and Rosenbloom, 2002).

Across the different literary streams, a few elements occur frequently. Business models are regarded as boundary-breaking concepts which define how the business is rooted in and interrelates with its surroundings (Teece, 2010; Zott & Amit 2007). They also show a firm by illustrating the different parts and combining it into an ensemble (Chesbrough & Rosenbloom, 2002). The key components of a business model however have not been agreed upon in the literature. Nonetheless, three elements have been mentioned by many authors, more specifically the value proposition, value creation and value capture (Tecee, 2010).

Furthermore, academics have also recognized that a business model can function as a basis of competitive advantage (Venkatraman & Henderson, 2008).

Despite the lack of unity in theoretical perspectives, simply based on the commonalities, the Business model as a unit of analysis proves to be worthwhile. Important however for this research is the choice of one theoretical viewpoint in order to use it as a defined concept in the

(9)

9 analysis, rather than just an abstract idea. One, clear conceptualization is also needed to compare cases within as well as across the technologies studied.

The conceptual view of Amit and Zott in this case has been judged to be most relevant and applicable, given the circumstances of the research. Unlike other Business Model schools of thought, which argue from cognitive standpoints or consider strategic choices, the model by Amit and Zott revolves around tangible elements and design themes. These are more suitable. It is further explained that the activity system perspective they take, enables and forces firms to think about the business model holistically, rather than an isolated number of choices.

Concerning the purpose of this research, this perspective seems viable, as the implication of one of the technologies will likely affect multiple parts of the business model, rather than having an secluded effect. They argue that a business model can be defined as “structure, content and governance of transactions” (Zott & Amit, 2007). Content comprises of the activities that are completed to distribute the value proposition. The structure discusses the way that these activities interact, meaning how they combine in order to deliver the value proposition. Hence, it refers to the value chain activities and the architecture of the organization. Lastly, the governance states which player performs certain activities.

This theory relates theories of economics to activities which create value in businesses. Four methods to create value in businesses are described. Efficiency is the first one and it is linked to the field of transaction cost economics. It refers to the cost savings realized due to the

connections in the activity system as well as savings through reductions of transaction costs. Novelty is related to Schumpeterian innovation and alludes to the degree of innovation. This can occur through the delivery of new content, new structuring of activities and the way these are governed. Complementarities, the third source of value creation discussed, is based in the

(10)

10 resource based view and touches on the synergies between business model elements as well as activities and components within the firm. Lastly, lock-in, is rooted in strategic networks and relates to those business model activities which create switching cost. Combined, they make up the NICE framework (novelty, lock-in, complementarities, efficiency). While they individually can create value in businesses, they are also interdependent on each other, as depicted in figure 3. An increase in one value driver can affect another value driver consequently, which means that the effects between them need to be considered as well, rather than viewing them as secluded elements. (Amit & Zott, 2001; Amit & Zott, 2012).

As part of the Long-Range Planning Special Issue in 2010 about the different integral elements of a business model, Zott and Amit further advanced their research. Using previous work, an activity system perspective was created, as shown below. The model consists of design elements and design themes. It combines the previous research insights, namely the content, structure and governance elements of an activity system in combination with the four drivers of value creation, mentioned in the NICE framework, depicted in figure 1 (Zott & Amit, 2010).

(11)

11

Business Models and Information Technology

Business Models in the past years have deviated largely from traditional ones, as mentioned above, most particularly shown by companies such as Amazon, Skype, Uber and Netflix. These Business Models have all been influenced by Information Technology. Certain patterns emerge, which can also be categorized by technologies. However, three dominant trends emerge,

regardless of which technology has an influence. These three, according to Fleisch et. al (2015) are as follows:

Integration of users and customers: IT has brought the users and customers closer to the firm, to the point in which they are being involved into the company’s value chain. Examples are Open source, E-commerce, Mass Customization.

Service Orientation: IT enables firms to be closer to the customer, even after sale. Relationships can be managed continuously. Examples in this case are Freemium, Rent instead of buy and Razor and Blade

Core Competence Analytics: The emergence of data and the analysis of such, be it transaction or use data, is valuable to a firm, helping it in processes like product design, pricing and sales structuring. Examples include Flat Rate, Pay per Use, Performance-based contracting.

As evident by these three trends, Business model patterns find themselves heavily influenced by technology, regardless of its nature. Given that IT in general influences business models a great deal, the influence of specific technologies becomes highly interesting. The impact of a singular technology can be studied, applied and leveraged in order to optimize business models.

(12)

12

Internet of Things

The internet of things connects the devices in our environment. Its range of application is seemingly limitless. Businesses can benefit from it in a variety of way. Especially in a

manufacturing environment, IoT devices can have a huge impact. On the one hand, it reduces uncertainty by exposing information and data from the manufacturing process (IoTP1). A resource which has been so far very scarcely used. This allows for information-based innovation for end users as well as producers (Ehret and Wirtz, 2016). Especially maintenance, repairs and performance improvements of machinery can be drastically improved by using IoT data

effectively (Fleisch et al. 2015) (IoTP2). Referring to the range of applicability, IoT devices stand to add value in homes, offices, factories, worksites, vehicles and even cities. Not only performance can be improved; based on IoT data, innovation can be radically improved, using the data to invent optimally designed products and parts. Retail stores, for example, can use IoT technology to automate check out and purchase processes (McKinsey.com, 2015).

Given the insights from the literature, the IoT technology is presumed to influence Business Models in a certain way. In regard to the methodology by Amit & Zott, especially the Design Themes are affected. Efficiency is the first value driver that is impacted. IoT devices will generate a large amount of data, which can consequently be used to optimize processes and products, making certain activities more transparent and therefore more easily adjustable and open for efficiency improvements. Fleisch et al. (2015) illustrate this well by stating that the marginal costs of measuring are close to zero and that adjustments can be made instantaneously (IoTP3). Greater efficiency is also achieved through predictive maintenance and remote repair, which together enable greater transparency and control of the process (Wortmann & Flüchter, 2015; Chen et al. 2014) (IoTP4 & IoTP7). Additionally, the reduction of measurement costs of

(13)

13 production processes, output and quality, combined with the lower governance costs, allows companies that are positioned downstream on the supply chain to divert from asset ownership, as it is no longer necessary. Hence, the IoT grants firms the opportunity to more easily outsource manufacturing and make use of service contracts for the production output (Ehret & Wirtz, 2016).

The other driver affected are the complementarities which play a role in relation to IoT

technologies. Certain software and hardware are necessary to make use of the technology, and more hardware can also mean more data that is available for use and analysis. Ju et al. (2016) use the example of home appliances to delineate this effect. Combining various appliances creates a more personalized and unified experience for the user. Cohesive services create synergies between the products and offer greater value to customers (IoTP5). Li et al. (2013) take the discussion a step further and not only limit the synergies of the technology to components within the IoT but elaborate on the possibility of combining the IoT with cloud services which in turn create the “Web of Things”, which permits the efficient delivery of new solutions by “leveraging computing resources and platform services, such as domain mediation, application context management and metering on cloud.” (IoTP6)

Cloud Computing

Cloud computing does not only entail benefits for private users. Business can too leverage this technology in their business model. The most obvious benefit is the obsolescence of software as well as hardware. This saves costs, time it takes to figure out which software/ hardware is necessary and enables the firm to be a lot more flexible and adaptable since no large fixed assets need to be acquired and maintained.

(14)

14 Berman et al. (2012) refer to ways that cloud computing can add value to how businesses better meet customers’ demands and drive future growth. Firstly, cost flexibility is increased, as firms can make use of the pay-per-use model. This means that a shift from capital expenses to

operational costs takes place. There is no more need to construct hardware, pay for license fees or develop software (CCP1). This allows for greater flexibility and enables more rapid

innovation. Secondly, business scalability benefits from cloud computing (CCP3). The

computing power and capabilities can be increased quickly and easily and can also be adapted to meet different demands at different times. Efficient growth can therefore be achieved. Third, market adaptability is another benefit of cloud computing to firms. Today’s market environment along with customer demands are changing rapidly. In order to meet customer needs better, cloud computing can be used to innovate and prototype rapidly due to the ability to swiftly modify processes, products and services. Masked complexity, the fourth benefit, refers to the opportunity that businesses have to disguise some complex, difficult parts of their operations from the end user, by for example performing upgrades or maintenance without the involvement of the user. Context-driven variability implies that a higher degree of personalization can be achieved. The high computational power and ability to store user information, enabled through cloud services, grants businesses the chance to create personal user experiences and address even the smallest market segments. Lastly, ecosystem connectivity enhances the collaboration with external partners. Cloud-based platforms enable various groups to share knowledge and information more easily and efficiently.

Maresova et. al (2017) conducted a cost benefit analysis on the use of cloud computing in companies. Benefits listed by them included cost savings in terms of saved energy, saved staff and allowing the IT staff to focus on key activities as well as increased flexibility and response to

(15)

15 business needs and increased collaboration with the customer. Costs were considered as risk of transition and data loss, purchase of software and technical equipment and costs of training employees. They concluded that cloud computing can significantly help in dealing with today’s uncertain market environment and using resources efficiently in the face of the still lingering effects of the economic crisis.

The elements of Business Models that are influenced by cloud computing are three of the four value drivers. Novelty is increased, as firms are able to more rapidly innovate and adapt to market conditions, due to less investments into fixed assets. Xu (2012) supports this by stating that cloud computing helps to “align product innovation with business strategy and creates intelligent factory networks that encourage effective collaboration”. Moreover, the “flexibility in deploying and customizing solutions”, allows firms to excel in the innovation of products and services (CCP4). According to Marston et al. (2011) many businesses profit from the

circumstance that no large investments into hardware need to be made, which leads to a faster time to market (CCP2). This is enhanced by the ability to use “compute-intensive business analytics and mobile interactive applications that respond in real time to user requirements”, granting businesses the capabilities to quickly react to changes in customer preferences and quickly incorporate those into the products or services (CCP5).

Complementarities are all the different and useful applications and functions that can be accessed

via the cloud, creating synergies between the modules and allowing the firm to benefit more from a multitude of products and services. Cloud architectures facilitate the utilization of plug-and-play software modules, which can allow for new combinations of capabilities and create new competences and activities to complement current activities (Xu, 2012) (CCP7). Specific

(16)

16 are location-, environment- and context-aware and respond in real time to data from a variety of sources, […] parallel batch processing that allows users to take advantage of huge amounts of data [and] business analytics that can use the vast amount of computer resources to understand customers, buying habits, supply chains and so on from voluminous amounts of data” (Marston et al., 2011). These characteristics of cloud computing enable firms to create data-driven

solutions and make use of a combination of software programs, which linked can create powerful tools.

Finally, efficiency refers to the cost savings of those who make use of cloud computing, as expensive fixed assets, requiring large capital investments, such as hardware and different software licenses, can now be eliminated. Efficiency can also be achieved due to the fact that fluctuating demand can be dealt with a lot more proficiently, as cloud computing often is employed based on a pay-per-use basis. If an own data center was utilized, its capacity must be able to deal with peak demand, resulting in underutilizations for all the times that peak demand is not occurring. Web Start-ups can also benefit in the regard that unknown demand might occur unexpectedly. Cloud computing can be easily adjusted to deal with unexpected upsurges

(Armbrust et al., 2010) (CCP6). The problem of underutilization is also pointed out by Marston et al. (2011), who state that “substantial capital investments in information technology are grossly underutilized”, with corporate servers using only 10-30% of their processing power. It is further added that maintenance costs debilitate large amounts of resources (up to two-thirds of average IT staffing budget), which are consequently not available for value adding activities. Manufacturing companies can profit from cloud computing in terms of more fluid

implementation of new functions (CCP8). Many customizations and alterations in the process can now be dealt with by the IT stuff seamlessly and quickly, rather than hardcoding,

(17)

17 implementing and adjusting it so the prospective process evolution. This would also require the need to involve new contracts with more outside vendors. Cloud based solutions eliminate that necessity (Xu, 2012).

Blockchain

Blockchain has the opportunity to increase trust in a collaborative environment, or even more so, abolish the notion of trust altogether (Beck et al., 2016; Zyskind et al., 2015) (BP1). While the primary purpose of blockchain is linked to the cryptocurrency Bitcoin, second generation blockchain applications have become increasingly popular. These involve asset ownership, intellectual property and smart contracts. Smart contracts in particular have the chance to disrupt the way contracts are being written and enforced. Smart contracts can be self-enforcing and simultaneously monitor data from external sources. Furthermore, blockchain can immensely improve the way distributed databases have performed. Updates and modifications to the databases can easily cause difficulties and errors in regard to how they are proliferated to the different nodes that contain the data. With blockchain technology, much like when a transaction in the bitcoin network is rejected when the balance has already been spent, certain modifications to data can be rejected, if the modification has already taken place (Peters & Efstathios, 2016). Additionally, smart contracts can evade the need for centralization in terms of a trusted party in cooperative process execution (Weber et. al, 2016).

Elements of Business Models affected by the blockchain technology are especially found in the structure of the Activity System. Blockchain reduces the need of trust between parties, making contracts and transactions safer and more transparent (BP2 & BP8) (Kosba et al., 2016). This can be seen by the graphic representation (Figure 2, skuchain.com, 2017) of relationships between

(18)

18 players in a supply chain, created by one the case companies, Skuchain. They display today’s supply chain, in comparison with tomorrow’s supply chain, which incorporates blockchain technology as a method of transaction between all links in a supply chain. The relationships are maintained, yet amplified with trust and transparency with respect to the flow and distribution of goods and money. Bheemaiah (2015) discusses that today’s collaborative digital economy profits from blockchain technology, as its decentralized model “removes the fallacy of having a single failure point that is inherent of today’s client-server based business models”. It is further argued that a “fundamental change in economic and societal infrastructure” (BP4) takes place and provides the chance of “creating a sharing economy that is decentralized, distributed and democratic” (BP3).

Today’s supply chain Tomorrow’s supply chain

Figure 2 – Blockchain’s impact on supply chains

The value driver efficiency is affected by the blockchain technology, in a way that trust and mistrust are taken out of the equation. Transaction cost are significantly reduced as contracts can be now be self-enforcing and transaction are safe and secure, which can remove payment errors

(19)

19 for example. Further, decentralized structures allow the network to carry the main load of

keeping privacy and keeping a consistent service level, which reduces costs for all individual players, leading to better resource use and efficiency (Doctorow, 2012) (BP9). Additionally, the removal of involvement of third parties in transactions, decreases the time spent and increases efficiency of said transactions, while simultaneously leading to lower transaction costs (Brito & Castillo, 2016; Zahng & Jiangtao, 2016) (BP6 & BP5). Blockchain’s use of 2-factor

authentication mitigates the risk of fraud and counterfeit (Bheemaiah, 2015) (BP7). Especially the pharma industry can benefit from this, as fake drugs cost pharmaceutical firms up to $200 billion per year (Economist.com, 2010).

Framework and propositions

Depicted below is the framework which is used to analyze the impact of the three technologies on the business model framework by Zott & Amit (2010). Given the purpose of this study, to research the effect each technology has on business models, the framework by Zott & Amit was adopted and used as basis for the analysis. It is hence studied in a different context than what it was originally created for. This will add value in regard to the verification and applicability of the model in other areas and industries. Since the impact of technologies on the business model is studied, no altercations to the framework itself have been made, as the impact of the

technologies cannot be pre-determined to only a few areas. While the theory gives indications of which elements might be more affected than others, excluding parts of the model prematurely may restrict the impact of this study and the extent of the results. The framework is depicted in figure 3. It consists of the elements of the business model framework as well as the propositions

(20)

20 and their expected impact respectively. Value drivers are linked to the propositions by color and lines.

Figure 3 – Activity System Perspective combined with NICE framework, as shown in Zott & Amit 2010

(21)

21 The propositions are derived from the literature review above. Each proposition has a unique code. Propositions for the internet of things are labeled IoTP1 – IoTP7. Cloud computing propositions are categorized by the codes CCP1 – CCP8. Lastly, propositions regarding the expected impact of blockchain technology are classified by the codes BP1 – BP9. The table in figure 4 shows the propositions, combined with the affected element of the business model by that particular proposition and is completed by the theoretical grounding for that proposition. All codes and propositions can be located in the literature review above, where the context of the code and the information which it was derived from can be found. This table provides a more organized and clearer overview. Furthermore, these propositions can be found in the framework depiction in figure 3, where each proposition is linked to the value driver or business model element that it individually affects.

Technology Proposition Affected Value driver/ Element of business model

Theoretical Grounding

IoT IoTP1: IoT devices will

generate large amounts of data, leading to more insight into the process

Efficiency Ehret and Wirtz,2016

IoTP2: Data gathered from IoT devices will lead to process improvements, through more control and insight into the process

Efficiency Fleisch et al., 2015

IoTP3: IoT devices will reduce measuring costs

Efficiency Fleisch et al.,2015 IoTP4: IoT devices will enable

predictive maintenance, which leads to cost savings

Efficiency Wortmann & Flüchter, 2015

IoTP5: Combining IoT

devices, will lead to synergies

Complementarities Ju et al., 2016 IoTP6: Combining IoT devices

with cloud computing, will create the “web of things”

Complementarities Li et al., 2013

IoTP7: IoT technology will increase efficiency by enabling

Efficiency Fleisch et al., 2015; Chen et al., 2014

(22)

22 remote access and remote

intervention into the devices Cloud

Computing

CCP1: Employing cloud computing, will lead to a decrease of local hardware required

Efficiency Berman et al., 2012

CCP2: Employing cloud computing technology will lead to more rapid innovation and a faster time to market

Novelty Marston et al., 2011

CCP3: Cloud computing technology will improve business scalability

Efficiency Berman et al., 2012

CCP4: Due to the flexibility of cloud computing, firms will be able to excel in the innovation of products

Novelty Xu, 2012; Berman et al., 2012

CCP5: Customer preferences are better understood due to advanced analytics provided through cloud services, enabling firms to create tailored innovation

Complementarities & Novelty

Marston et al., 2011

CCP6: Cloud computing enables firms to more efficiently deal with fluctuating demand

Efficiency Armbrust et al., 2010

CCP7: Cloud computing facilitates the implementation of a variety of modules, which create synergies between them

Complementarities Xu, 2012

CCP8: Cloud Computing enables firms to implement new functions more efficiently

Efficiency Xu, 2012

Blockchain BP1: Blockchain redefines the notion of trust in a transaction environment

Activity System Structure

Zyskind et al., 2015; Beck et al., 2016 BP2: Overall network security

is improved due to blockchain technology

Kosba et al., 2016

BP3: Decentralization in the system will increase due to blockchain technology

Bheemaiah, 2015; Doctorow, 2012 BP4: The activity system

structure surrounding firms will change

(23)

23 BP5: Blockchain technology

reduces the cost of transactions

Efficiency Brito & Castillo, 2016 BP6: Blockchain technology

makes transactions more efficient

Efficiency Zahng & Jiangtao, 2016

BP7: Blockchain technology mitigates the risk of fraud and counterfeit, leading to cost savings for firms

Efficiency Bheemaiah, 2015

BP8: Blockchain technology increases the visibility and transparency of transactions, which adds value to firms and customers alike

Kosba et al., 2016

BP9: In a blockchain based system, costs of security are carried by the system, rather than by players, lowering security costs for all players involved

Efficiency Doctorow, 2012

Figure 4 – Proposition derived from literature review

Research Design

The research will be a multiple case study of exploratory nature. Case studies are useful research methods when a holistic, in-depth examination is required (Zainal, 2007). As the impact of certain technologies on business models is the underlying objective, a holistic and detailed approach is necessary to uncover the wide range of effects. In general, studying multiple cases is more compelling as it provides more insight from more and different scenarios. In this case, the effect will be measured across as well as within industries. Hence, selecting multiple cases is the logical choice. The Business Models, the unit of analysis, of three significant, important firms will be studied for each technology. That means, a total of nine cases will be selected. All nine

(24)

24 cases will be studied in parallel. The underlying arguments for the choice of firms will be further explained in the latter parts of this section. Firms chosen will have implemented at least one type of the three technologies and will use it in some way to add value to their business model. Each case will be studied in-depth and per case the elements of the business model will be analyzed in detail. The data for the cases will be gathered as secondary data through the means of the firm’s websites, annual reports and newspaper articles collected through databases like Lexis Nexis, meaning the data sources are text and document analyses. Document analysis is especially useful as a research method for qualitative case studies (Johnson & Stake, 1996). Advantages of

document analysis include the availability of data, meaning data can be accessed easily without permission by the authors and the stability of the data, meaning that the researchers presence does not affect the data in any way. Further, the coverage in document analysis is very wide, as documents can cover many events and settings (Bowen, 2009, Yin, 1994). The second part of this research will contain three expert interviews, one per technology. These interviews will provide primary data. Interview partners are experts who are working in a company that does business in the field of one of these technologies. The interviews will resolve around the topic of implementing the studied technologies within the business model of the representative’s

company. Interesting here, are possible inhibitors and restrictions to implement the technologies. Information found during the research will be attempted to be validated by these experts as well as new information revealed. Construct validity, which is also related to confirmability, is provided using multiple sources of evidence (Riege, 2003). Triangulation, so the use of multiple methods to study the same phenomenon (Bowen, 2009), can enhance confirmability and at the same time grant credibility, also known as internal validity. Internal validity, aside from

(25)

25 support explanation building (Miles & Huberman, 1994). Transferability, also referred to as external validity, is established through the accurate description of the settings of this study and concurrently through the selection of multiple cases in different settings, which allows readers to apply a replication logic for the transfer of the findings into other circumstances (Eisenhardt, 1989). Lastly, reliability, or dependability, was achieved by trying to document observations and actions as accurately as possible (Lecompte & Goetz, 1982).

The following Cases are selected for each technology:

Internet of Things

Due to the wide range of application possibilities of IoT technologies, each case will be selected from a different industry, in order to attempt to display a rather varied impact of the technology on business models. The industries that are selected are Healthcare, Manufacturing and Home applications. Representing the Healthcare sector will be the Ireland based Medtronic, which operates worldwide. Products such as their ZephyrLIFE Hospital Remote Patient Monitoring and their Vital Sync Platform are prime examples how the Internet of Things can add significant value in the Healthcare sector (Medtronic.com, 2017). The home applications will be represented by Nest, which have greatly impacted the market with their Nest thermostat as well as smoke detectors and cameras. Furthermore, Nest displays a wide variety of combinable products, leveraging the internet of things technology across the whole household. (Nest.com, 2017) Regarding the industrial IoT use, Bosch is selected as the representative company. This case is especially interesting, as Bosch is manufacturer of IoT technologies for the Industry 4.0 as well as Lead user, allowing it to pool expertise from a multitude of sources (Bosch.com, 2017).

(26)

26

Cloud Computing

Cloud computing, similar to IoT, can be used for a wide variety of purposes. The following cases are selected. Google offers a plethora of cloud computing services, starting at its commercially available Google Drive, apps like Google Docs and its cloud platform, offering capabilities of development tools and storage & databases (Google.com, 2017). Secondly, Apple will be used as a case. Their iCloud is a representation of a well-integrated function into its business model, benefitting from complementarities. Lastly, Netflix is a case which perfectly shows how the migration to the cloud can add value to a business model. Multiple aspects of Netflix’s service have hugely advanced and improved since its switch to Amazon’s web services. For example, Netflix’s service availability has greatly improved (Netflix.com, 2017).

Blockchain

In comparison to the other two technologies, Blockchain is a relatively immature one. It’s use cases are still relatively undeveloped and if companies implement the technology into their business model, it is still in its early stages. One interesting company is Blockstream. It’s primary innovation is sidechains, which is a technology invented to spread applicability of the blockchain technology, for purposes such as smart contracts and native asset issuance

(Blockstream.com, 2017). Shocard uses the blockchain technology to create personal, digital ID cards. These are secure, easy to use and can be useful in many regards, such as identity

verification, password-less logins or proof of age (Shocard.com, 2017). Lastly, Skuchain is a company which builds products for B2B trade and supply chain finance based on blockchain technology. This allows for more transparency and trust in collaborative commerce and connects each supply chain element more than ever before (Skuchain.com, 2017). Based on this selection, it is assumed that an array of possible uses of the blockchain technology is covered.

(27)

27

Data Analysis

As part of the document analysis, more than 300 documents were skimmed and considered as possible sources. As researcher, it is required to consider the raw data set and make a choice on what is significant and relevant in context of this research and then convert it into a format that can be used in regard to the research question (Krathwohl, 1988). Eventually, a selection was made on 87 files. These files were rated as more relevant, based on the relatedness to the topic and further the occurrence of the initial codes. The articles and other text documents are spread equally over the three technologies and the cases within them. The data collection process included searching data bases like Lexis Nexis and Questia, subtracting information and press articles from the companies’ websites and looking at interviews as well as presentations by the firms, or representatives of those, themselves. Additionally, white papers published by the case companies were used for the analysis. Due to the variety of different types of sources, it is believed that an accurate representation of multiple views and subjectivities are achieved. This is also important, as a wide range of documents is generally important in “providing preponderance of evidence”, especially when the research is heavily based upon text analysis (Bowen, 2009). Nonetheless, the content of the documents is not necessarily perfect, as some topics and themes are touched upon more frequently and in more detailed. A critical analysis of the sources and the content itself will follow in a later section, complemented by the discussion of the general drawbacks of secondary data and the method with which it was collected and analyzed. The data analysis process was performed via a directed content analysis. This approach’s goal is “to validate or extend conceptually a theoretical framework or theory. It can provide predictions about the variables of interest or about the relationships among variables, thus helping to

(28)

28 determine the initial coding scheme or relationships between codes” (Hsieh & Shannon, 2005). Existing theory and prior research provide the background for recognizing key variables for the creation of initial codes. This simultaneously grants validity to the method. Coding schemes based on theory, which point coders to the central concepts are considered “valid coding schemes” (Potter & Levine-Donnerstein, 1999). Using these initial codes, the coding process is commenced. Data which cannot be coded with these, is marked and later analyzed whether it represents an entirely new code or a subcategory of the pre-determined ones. The results of this analysis approach can be used descriptively in terms of frequency of codes used, but not

examined with statistical analyses of difference as the coded data will most likely not provide meaningful outcomes. This method is especially useful for this study, as the main strength is that existing theory will be reinforced and expanded (Hsieh & Shannon, 2005).

Hence, the theoretical background developed in this study has been the basis for the formulation of starter codes. Any new insights gained during the data analysis process were added as

additional codes. One special case exists, more specifically, the code called “lock-in” for the codes concerning cloud computing technology. It was considered as an important insight, that some firms create negative lock-in effects by binding themselves to a certain cloud service provider. The problem however is, that none of the value drivers of the framework were regarded as suitable categories for said code, which is why it stands alone.

The following sections will descriptively show the data that was collected for each technology.

Internet of Things

The data set of the Internet of Things companies comprises of 28 documents, which are equally dispersed over the three case companies. These documents were analyzed based on the coding

(29)

29 structure depicted in Appendix A. Shown in figure 5 below is an overview of the data, depicting the frequency of codes used, in total and per case.

Figure 5 – Data overview for Internet of Things cases

Clear patterns can be observed when looking at the codes within the technology across cases as well as within cases. The three overarching themes with the most code occurrences are

Synergies between modules, Remote access to data and Reduction of costs. Also frequently mentioned were the themes of Network creation and Increase of data available. The remainder of the codes were mentioned at least once, but not as frequently as the previously mentioned

0 2 4 6 8 10 12 14 16 18 20 22 Reduction of costs

Reduction of Measurment costs Lower failure rate Increase of data available Advanced Analytics Process Improvement Predictive Analytics Predictive Maintenance Remote access to data Remote Intervention Complementarities Personalization Synergies between modules Network creation Web of things

Data overview IoT

(30)

30 ones. The only code, which has not been used to categorize content is the “web of things”. Quite possibly it was too specific since it originated as a starter code which was synthesized from the literature. Furthermore, the “web of things” was discussed as being a concept created by the combination of the internet of things and cloud computing. None of the content analyzed focused primarily on the combination of the two technologies. This perhaps explains the lack of usage of said code.

Each case, so each company studied, differs in terms of predominant themes. The data analyzed for Bosch, a manufacturing company, was mainly coded with Synergies between modules, Remote access to data and Remote intervention. This is to a certain point comprehensible. IoT in manufacturing adds value when lots of machines are connected, providing data on how to improve the production process and allowing for adjustments of the process remotely, if necessary or beneficial. Related codes which also were used for the content on Bosch were Predictive Maintenance, Process Improvement, Cost reduction and Lower failure rate, although with less frequency than the aforementioned ones. In this case, the use of these codes can be justified, as all of these characteristics and actions are value-adding for a manufacturing company. The impact of the technology however, will be discussed in detail in a later section. Medtronic’s analyzed content was most recurrently coded with Remote access to data,

Reduction of costs and Predictive Analytics. Additionally, the codes Predictive Maintenance and Reduction of Measurement costs were assigned above-average. Similarly, to the previous case, these codes seem consistent with the nature of the business and how the technology can add value. Healthcare can become very expensive quickly, as capable personnel, such as doctors, are rare and high in demand. Using technology to enable the doctor patient-relationship to be more

(31)

31 efficient and making the provision of feedback easier, in both directions (doctor to patient, patient to doctor) can reduce costs significantly and increase the quality of care.

Lastly, the remaining case, Nest, differs greatly from the other two in terms of the use of codes. The content was predominantly associated with the codes Network Creation, Synergies between modules and Remote access to data as well as increase of data available.

Considering the business model of Nest, these codes are reasonable. The more IoT devices are utilized in a household, the more their effect can be leveraged, by letting the devices

communicate with each other and optimize the energy consumption. The ultimate goal is the creation of an entire network, a platform, in which all devices can be connected and enjoy interoperability.

Cloud Computing

The 29 articles in total containing content about the cases studied for the cloud computing technology were analyzed according to the coding scheme depicted in Appendix B. The

aggregated data which was analyzed for the cloud computing cases is illustrated below in figure 6.

(32)

32

Figure 6 – Data overview for Cloud Computing cases

The data for this technology appears to be spread more equally over the codes, compared to the IoT data. Ultimately, only one code has not been used to associate data with, namely the

Efficiency Implementation of new functions. The remainder of the codes were used rather uniformly. Outstanding are however the ability to deal better with fluctuating demand, Synergies between modules and Data-intensive business analytics. Aside from that, the codes for cost savings were all used quite frequently. The same goes for the Complementarities driver, where the delivery of new services also scored highly, besides synergies and data-intensive business analytics. The novelty driver saw the lowest amount of codes utilized. Here, the customization of products/services scored highest, with availability/reliability ranking second. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 Cost savings Fluctuating demand Scalability Hardware reduction Implementation of new functions Delivery of new services Synergies between modules Better customer understanding Interactive applications Data-intensive business analytics Flexibility of products/services Faster time to market Availability/reliablity Supplier Lock-in

Data overview Cloud computing

(33)

33 The first case company studied for this technology, Apple, was mainly associated with the complementarities and efficiency codes. Synergies between modules as well as cost savings scored highest, while the novelty driver was not utilized at all. Apple however, was, looking back, not the most appropriate choice as a case company. While the company uses cloud computing as part of their services offered, they do not possess their own data centers. Hence, Apple’s dependence on other firms’ cloud services, created a Lock-in effect, which was added as a new code.

Google, the 4th_{largest cloud provider in the market (Harvey, 2017), was a special case. Here, it}

was not studied how cloud computing affects their business model, but rather what Google offers other companies who use their cloud services, and how that affects their respective business models. It is studied from the supplier perspective so to say. Firms can benefit especially from Data-intensive Business Analytics as well as the flexibility and customization of products. By offering a wide range of services and options, firms can choose and personalize which services best suit their needs. Further, Google’s services allow firms to deliver new services. These were the most used codes for the analysis of Google’s content. Further, Google’s cloud service allows firm to better deal with fluctuating demand as well as eliminate the need to invest in expensive hardware, as indicated by the data.

Netflix was an example of a firm who moved all their data from their own data centers to the cloud. Data on the move was analyzed and the motivation to do so was considered as the benefits that the cloud computing technology provides for their business model. Data included statements of the firm, as well as objective articles. Clear patterns are noticeable when looking at the data. Prominent are the cost savings. While often mentioned generally, specific mechanisms of saving costs were also mentioned. Cloud computing allow Netflix to deal better with fluctuating

(34)

34

demand and provided more scalability. Another benefit of the technology is that it allowed Netflix to reach new levels of Availability and Reliability of their streaming service.

Blockchain

The last technology and the corresponding cases are represented by 30 articles, which were investigated using the coding structure illustrated in Appendix C. An overview of the aggregated data and the codes used can be seen in figure 7.

Figure 7 – Data overview for Blockchain cases

0 2 4 6 8 10 12 14 16 18 20 22 24 Increase in transaction efficiency

Reduction of transaction costs Costs of privacy carried by system Mitigation of fraud Change of structure of ativity system Decentralization Decrease of chance of failure Network Security Abolishment of trust Creation of new platform Wider reach Better visibility Added value for customer due to visibility

Data overview Blockchain

(35)

35 The data for the Blockchain technology is heavily two sided. While some codes are used often, other codes were only associated with a few passages of text. The three most dominant codes are network security, combined with the abolishment of trust, the reduction of transaction costs and the increase in transaction efficiency. On the other hand, creation of new platform and cost of privacy now carried by system rather than players, were used fewer times.

All three case firms have the objective to revolutionize certain industries with their respective application of the blockchain technology. Firstly, Skuchain plans to change the way global supply chain perform transactions. Dominating codes for the data are the change of the structure of the activity system, reduction of transaction costs, better visibility and the mitigation of fraud and counterfeit. Considering their objective, these seem understandable and in line.

Blockstream’s objective is to deploy so called sidechains, an addendum to the main bitcoin blockchain, for different purposes, which do not need be handled under the same circumstances as the cryptocurrency, for example differing in need for security or transaction speed. The content analysis of the Blockstream data clearly focused around the codes network security, abolishment of trust and decentralization. Furthermore, however with lower frequency, the codes creation of new platform and increase in transaction efficiency were associated with the data.

A completely digital ID is the idea behind the Shocard business. The company’s presentation along with articles on their product and process were mostly linked to the codes of network security, increase in transaction efficiency and mitigation of fraud. Additional codes utilized are reduction of transaction costs and costs of privacy carried by the system rather than by players.

(36)

36

Discussion

In this section, the discussion of the data analysis results and what these signify in regard to the research questions and the theoretical construct business model, takes place. Further, the collected data is discussed.

Internet of Things

The internet of things technology, the inter-connectivity of physical devices, can add value to a firm’s business model on a number of levels. The two main value drivers affected by it are Efficiency and Complementarities, as part of the Amit & Zott terminology (Zott & Amit, 2010), which is according to the expected value drivers effected, as outlined in figure 3. This section will revolve around the theoretical and managerial implications that the research and the results of this study will deliver.

An evaluation needs to take place in regard to the propositions. Since mostly all of them have received some support in the data, a clear differentiation needs to be made concerning which ones turned out to be true and which ones did not. IoTP1 is supported by the fact that an increase in the availability of data takes place. This supports the claim by Ehret and Wirtz (2016) and implies that firms who find themselves lacking control and facing high uncertainties in their production process can benefit from employing IoT technology, granting them the ability to excel in information-based innovation. IoTP2 strongly relates to the previous proposition, but is not as strongly verified by the data. While some evidence exists that IoT technology may lead to process improvements, Fleisch et al. (2015) added that this may happen if data is used

(37)

37 effectively. An increase in data and more insight into the process does not automatically

guarantee more efficiency after all. Firms cannot automatically expect to improve their whole production by using IoT devices. Effective implementation and correct use of data is still important and necessary. IoTP3, relating to the reduction of measurement costs, enjoys little evidence in the data. While this is fairly logical given the nature of the technology, that measurement costs are reduced, the data analyzed did not mention it specifically. The

terminology used was rather centered around cost savings in general, instead of directly relating to measurement costs. As indicated by Fleisch et al. (2015), measurement costs are close to zero, and firms can expect cost savings in general when using IoT technology in their measurement process. IoTP4 is also slightly supported through the findings. Firms who face immense risks and repercussion in case of process failure, can add a layer of security by employing IoT devices which can be controlled and checked via predictive analytics and maintenance. (Wortmann & Flüchter, 2015). The proposition that combining IoT devices will lead to synergies (IoTP5) is one of the most frequently mentioned ones in the coded data. Although illustrated by Ju et al. (2016) from the perspective of home appliances, the creation of synergies by combining a multitude of IoT devices is evident in the other two studied industries as well. This can be seen as an extension of the theoretical claims made in the paper and additionally aids businesses in the decision to adopt IoT technology. It is advised to rather implement it fully, or at least in a close local context, where many IoT devices are operating in a coherent process, rather than

implementing them dispersedly over the firm. If integrated well, they can reap huge benefits. The “web of things” (IoTP6) has not been mentioned in the data once. Combining IoT devices with cloud computing may very well theoretically create the so called “web of things” (Li et al. 2013), but no evidence for it exists in the cases studied. Quite possibly, the terminology is too specific

(38)

38 or the firms studied did simply not combine the two technologies. Finally, IoT devices enable remote access and remote repair for the IoT devices (IoTP7). Strongly supported by the data collected, this proposition has a wide impact on businesses. Remote access into machines may not only make outsourcing production a lot easier (Ehret & Wirtz, 2016), but also grants firms more room in the planning of production facilities. Close proximity to machines is no longer required, as data can be retrieved easily and even remote intervention may grant control over some process steps and changing of settings from a distance.

In resume, the efficiency value driver of the business model framework is effected by the generation of large amounts of data, which can, if used effectively, lead to process

improvements, a reduction of costs, the ability to detect failure before it happens and accessing data and the machinery remotely. Theoretically, some of the findings from the papers studied are verified and in some cases reproduced in a different setting, adding further value to these

findings. The complementarities driver shows that creating an integrated network of IoT devices allows firms to benefit from synergies. These can manifest themselves in terms of increased control and insight into processes and a more cohesive and connected experience in general. Combining these two findings, the interrelatedness of the value drivers, as shown in figure 3, is evident. The more IoT devices are connected, the more control and insight into the process is achieved. In consequence, more efficiency is achieved due to an increase of data and a wider reach in regard to access to machinery. This further verifies the business model framework and how value is created in firms, now shown in a new and previously unapplied realm of technology and business.

A possible thread to such a development of a shared network lies in the problem of interoperability. Many of the providers of IoT devices, sensors, software and even whole

(39)

39 platform understandably want to create customer lock-in mechanisms by building upon their own standards and compatibility. This may seriously restrict the growth and advancement of the technology. As a countermeasure, a large-scale standardization process needs to occur in order to create unity across norms (Blackstock & Lea, 2014). Furthermore, security risks are a looming concern around the technology. Having many devices within companies as well as in our daily lives connected at all time creates immense security risks, which include “confidentiality, authenticity, and integrity of data sensed and exchanged by ‘things’” (Babar et al., 2010).

Cloud Computing

As for cloud computing in general, several benefits have been identified that affect firms’ business models. CCP1, which states that by deploying cloud computing, firms can expect a reduction in the hardware they need to invest in, is supported by the findings from the cases. Cost flexibility is achieved as expenses move from capital expenses to operational expenses (Berman et al., 2012). This is mostly applicable for smaller firms and start-ups, who often do not possess sufficient funds to invest in the required hardware from the start. If funds are limited in the short-term, cloud computing provides a great opportunity to scale the operations without large upfront investments. Although a somewhat logical consequence out of the previous benefit, the proposition that more rapid innovation and a faster time to market is enabled (CCP2), has not shown up in the data. A possible explanation lies in the maturity of the firms studied, which were all large incumbents, which do not necessarily need to innovate rapidly in order to stay

competitive. The theoretical view, provided by Marston et al. (2011), is thus not influenced by the findings. Given the fact that cloud computing is usually offered on a pay-per-use basis (Berman et al., 2012), business scalability is improved (CCP3). Firms can simply increase their

(40)

40 computing power if necessary and scale their operations up and down, at the same cost ratio. If large, consistent growth is expected in a business, investing in hardware is most likely going to be very cost intensive, as new hardware needs to be added continuously. Cloud computing enables firms to not have to lock up resources into these hardware investments. Although mostly supported by the findings for the Google case, the flexibility of cloud computing, does enable firms to excel in innovation (CCP4). For example, Google offers services ranging from storage, over data management to data analytics and application development capabilities (Townsend, 2017). This leads to flexibility in employing these services. Tailored to one’s needs,

customizations can help to ideally fine-tune the way these are used and support and enhance other business activities as well as the creation of new products. Firms can choose which services are needed to add value to their innovation process and simply add or remove these in case they are no longer needed. Customer understanding is brought to new dimensions with the use of cloud computing (CCP5). The specifically mentioned “compute-intensive business analytics” (Marston et al., 2011) have been frequently mentioned in the data and shown to enable firms to create tailored innovations. Obviously, this new ability to understand customer preferences needs to be utilized correctly and be used as the grounding for customized products. Theoretically, this shows that the value creation in businesses is intertwined, as one benefit may heavily impact another, as is here the case with the value driver complementarities influencing the novelty value driver (Zott & Amit, 2010). Relating to the propositions CCP1 and CCP3, cloud computing does not only add value by eliminating the cost intensive in hardware, but also allows firms to more efficiently deal with demand variations (CCP6). Practically, this is useful for firms facing inconsistent demand, may it be in the long term or short term. Seasonal products and firms that produce them may expect lots of traffic for example in winter, or in summer, but comparably low

(41)

41 amounts during the off season. Cloud computing prevents them from investing in expensive data centers which will be suffering from underutilization for half a year, leading to cost savings. Especially the Netflix data supports Armbrust et al.’s (2010) findings, by illustrating the benefits Netflix has enjoyed since moving to the cloud. The creation of synergies between the modules of cloud computing is backed by the data (CCP7). Simply considering the fact that firms benefit from cost savings and simultaneously receive more insight into their customers’ demands shows the synergies the technology entails. If firms make use of cloud services, a critical discussion needs to take place beforehand in order to ensure that all services which may add value to their respective business are utilized in order to fully enjoy the synergies of the technology. The final proposition of the cloud computing technology states that it will allow firms to more efficiently implement new functions (CCP8). This has not been the case in the sample that was studied. Granted, Xu’s (2012) study referred to manufacturing companies, which were not studied for this technology. Despite Apple manufacturing phones and other hardware, the use of cloud

technology in that case was not related to manufacturing. So, while the theory by Xu (2012) cannot be supported, not enough evidence exists to refute it.

Regarding the business model framework, the interrelatedness of the value drivers can be excellently witnessed. The efficiency increase in terms of more cost flexibility and variability in computing power, combined with the increased understanding of customer preferences, provides firms with more funds to create tailored and customized innovation.

As previously mentioned, during the data analysis process, the term “lock-in” was repeatedly encountered. This deals with the risk that firms face when receiving cloud computing capacity from one supplier. Due to non-compatible API’s between different cloud providers, customers may confront data extraction and data transfer issues. Consequently, it is considered as an

(42)

42 inhibitor, as firms may be hesitant to adopt cloud computing. However, similarly to the IoT technology, an active standardization process across the main suppliers can reduce said risk massively (Armbrust et al., 2010).

Blockchain

As the least mature one of the technologies studied in this thesis paper, blockchain technology and its effect on business models is surrounded by the most uncertainty. An accurate description of that phenomenon was given by Louis de Bruin, Head of IBM’s European Blockchain

Department, at the UvA Room for Discussion on the impact of blockchain, which he formulated as:” Blockchain poses as the solution to yet unknown problems. We know the properties of the technology but its potential and usability is not fully understood” (de Bruin, 2017). Therefore, firms have not adopted the technology on a wide scale yet, as the technology is not fully developed and not mature enough.

Due to that, the selection of case companies to study was rather complicated. No firm has really adopted the technology yet as a supplement to their business model. Hence, the firms studied here are completely centered around it and they have either used it for their products or have an adapted application of the technology. Nevertheless, it was attempted to extricate the effects that blockchain can have on business models.

Blockchain, out of the three technologies studied, is the only one who affects the structure of the activity system. This effect is seen in the propositions BP1-BP4 & BP8. The element trust is experiencing a drastic change (BP1). Whether it is referred to as an abolishment of trust or a now trust-free system (Beck et al., 2016; Zyskind et al., 2015), strong evidence in the data exists that

Digital business model innovation : a multiple case study

MASTER’S THESIS