Capitalising on Big Data from Space

(1)

IN

DEGREE PROJECT INDUSTRIAL MANAGEMENT, SECOND CYCLE, 30 CREDITS

STOCKHOLM SWEDEN 2019,

Capitalising on Big Data from Space

How Novel Data Utilisation Can Drive Business Model Innovation

MARIA BREMSTRÖM SUSANNE STIPIC

KTH ROYAL INSTITUTE OF TECHNOLOGY

SCHOOL OF INDUSTRIAL ENGINEERING AND MANAGEMENT

(2)

www.kth.se

(3)

Capitalising on Big Data from Space

How Novel Data Utilisation Can Drive Business Model Innovation

Maria Bremstr¨ om Susanne Stipic

Master of Science Thesis TRITA-ITM-EX 2019:380 KTH Industrial Engineering and Management

Industrial Management SE-100 44 STOCKHOLM

(4)

Kapitalisera p˚ a stora datam¨ angder fr˚ an rymden

Hur nya s¨ att att utnyttja data leder till innovation av aff¨ arsmodeller

Maria Bremstr¨ om Susanne Stipic

Examensarbete TRITA-ITM-EX 2019:380 KTH Industriell teknik och management

Industriell ekonomi och organisation SE-100 44 STOCKHOLM

(5)

Master of Science Thesis TRITA-ITM-EX 2019:380

Capitalising on Big Data from Space How Novel Data Utilisation Can Drive Business

Model Innovation

Maria Bremström Susanne Stipic

Approved

2019-06-03

Examiner

Matti Kaulio

Supervisor

Ebba Laurin

Commissioner

Swedish Space Corporation

Contact person

Tobias Roos

Abstract

Business model innovation has in recent year become more important for firms looking to gain competitive advantage on dynamic markets. Additionally, incorporating data into a firm’s business model has been shown to lead to improved performance. This development has led to interest in the connection between data utilisation and business model innovation.

This thesis provides an in-depth case study of a Swedish space firm active within the satellite industry. The firm operates within an increasingly dynamic market, and ongoing disruptions in the form of new market entrants and rapid technological advancements has led to a search for new business opportunities. As a result, novel ways of utilising the increased amounts of data from space are of significant importance. While the firm is still realising profits utilising their

incumbent business model, the firm must simultaneously explore new business opportunities to avoid extinction.

The findings show that novel data utilisation, in the form of data processing, leads to business model innovation. Furthermore, the degree of business model transformation is dependent on how many of the business model's underlying elements are affected by data utilisation. Furthermore, the study concludes that a lack of trial-and-error learning impedes radical innovation efforts and hinders the development of ambidextrous capabilities within the firm. Lastly, the study finds a novel connection between the introduction of large-scale projects and improved ambidextrous capabilities.

Keywords:

Business Model Innovation, Data-Driven Business Model Innovation, Organisational

Ambidexterity, Satellite Data, Big Data

(6)

Examensarbete TRITA-ITM-EX 2019:380

Kapitalisera på stora datamängder från rymden

Hur nya sätt att utnyttja data leder till innovation av affärsmodeller

Maria Bremström Susanne Stipic

Godkänt

2019-06-03

Examinator

Matti Kaulio

Handledare

Ebba Laurin

Uppdragsgivare

Swedish Space Corporation

Kontaktperson

Tobias Roos

Sammanfattning

Innovation av affärsmodeller har under senare år blivit alltmer viktigt för företag som vill uppnå ökad konkurrenskraft på dynamiska marknader. Vidare har det visat sig att företag som använder data för att förändra sin affärsmodell når bättre resultat än sina konkurrenter. Detta har lett till ett intresse för kopplingen mellan datautnyttjande och innovation av affärsmodeller.

Detta examensarbete består av en fallstudie av ett svenskt rymdföretag, som har del av sin verksamhet inom satellitbranschen. Företaget verkar på en alltmer dynamisk marknad, och pågående störningar i form av nya marknadsaktörer och tekniska framsteg har lett till att företaget nu måste söka efter nya affärsmöjligheter. Som ett resultat av detta blir nya sätt att använda de ökade mängderna data från rymden av stor betydelse. Fastän företaget fortfarande framgångsrikt nyttjar sin befintliga affärsmodell, måste företaget samtidigt undersöka nya affärsmöjligheter för att undvika att hamna efter marknadsutvecklingen.

Studiens resultat visar att nya sätt att använda data, i form av databehandling, leder till innovation av företagets affärsmodell. Dessutom beror graden av innovation på hur många av affärsmodellens underliggande byggstenar som påverkas av införandet av data. Studien drar vidare slutsatsen att en brist på lärande genom ’trial-and-error’ inom företaget hindrar radikala

innovationsinsatser och leder till begränsade förutsättningar för att hantera organisatorisk ambidexteritet. Slutligen finner studien att storskaliga innovationsprojekt kan förbättra förutsättningarna för organisatorisk ambidexteritet.

Nyckelord:

Affärsmodellsutveckling, innovation, datadriven affärsutveckling, organisatorisk

ambidexteritet, satellitdata, big data

(7)

List of Figures

1 The Three Stages of a Business Model’s Journey. Christensen, Bartman and Van Bever (2016) . . . 8 2 Double Ambidexterity Framework. Adapted from Kaulio, Thor´en and Rohrbeck (2017). . 10 3 Summary of the Business Model Elements. . . 16 4 Framework Illustrating Organisational Response and its Corresponding Business Model. . 18 5 Translation of the Business Model Elements Outlined by Sch¨uritz and Satzger (2016). . . 18 6 Current Business Model of SaMS. . . 30 7 Current Business Model of SaMS. . . 31 8 Low-Level Processing as an Organisational Response and its Corresponding Business

Model. . . 36 9 High-Level Processing as an Organisational Response and its Corresponding Business

Model. . . 37 10 SaMS’ Responses to Disruptions. . . 39 11 Transformation of SaMS’ Business Model Over Time. . . 44

(10)

List of Tables

1 Business Model Elements. Adapted from Hartmann et al. (2016). . . 4 2 Categorisation of Organisational Responses. Adapted from Kaulio, Thor´en and Rohrbeck

(2017). . . 10 3 Patterns of Data-Infused Business Models. Adapted from Sch¨uritz and Satzger (2016). . . 12 4 Levels of Data Processing. Adapted from Parkinson, Ward and King (2006). . . 14 5 Case Study Tactics for Three Relevant Tests. Adapted from Yin (1994). . . 22

(11)

Acknowledgements

Conducting this Master’s thesis has indeed been a exceptional journey, filled with intellectual challenges where we had the opportunity to immerse ourselves in highly interesting areas of research. This research is the final and concluding step of our Master of Science degree in Industrial Engineering and Management, and it would not have been possible without the help and support from several people.

First, we would like to sincerely thank our supervisor at the Royal Institute of Technology, Ph.D. Ebba Laurin. You stood by our side throughout the whole process, highly engaged in everything from thor- oughly reading our thesis to engaging in long discussions with us. Your guidance truly challenged us in a positive way, and your profound academic support enabled us to improve our research.

We would also like to express our gratefulness to our supervisors at the investigated company; Tobias Roos and Stefan Gustafsson. You have given us support, shown immense interest in our work and we thank you for the opportunity to explore the exciting space industry. Our discussions and your insights considerably assisted us in arriving at our findings and conclusions. We also thank all the informants who dedicated their time and efforts, together with other employees who made us feel welcome and guided us through practical aspects. Without you, this thesis would simply not have been possible.

Finally, we would also like to acknowledge our examiner, Associate Professor Matti Kaulio, together with our peers in the seminar group. Your participation in discussions and insights contributed to our research and guided us along the way.

Maria Bremstr¨om and Susanne Stipic Stockholm, June 2019

(12)

1 Introduction

In this section, the foundation of the thesis is presented. A brief background on the subject is given, followed by a problematisation which leads into the purpose and research questions of thesis. The delimitations of the thesis are also outlined.

1.1 Background

In today’s dynamic and fast changing markets, firms must stay competitive to survive (Wirtz, Göttel and Daiser 2016). A way of achieving competitiveness is to innovate the firm’s business model, which in recent years has gained more attention (Wirtz, Göttel and Daiser 2016). Firms engage less and less in process or product innovation due to their time consuming and resource intensive nature, and are increasingly turn- ing to business model innovation as an alternative or complement (Amit and Zott 2012). When exploring business model innovation, organisational ambidexterity comes into play. Meaning, an organisation has to manage evolutionary and revolutionary change at the same time to ensure long term survival and success (Kaulio, Thorén and Rohrbeck 2017). A common explanation for why organisations fail to innovate is the lack of organisational ambidexterity (Tushman and O’Reilly 1996). Furthermore, ambidexterity also brings to light the interplay between business model innovation, technological innovation, exploration and exploitation (Kaulio, Thorén and Rohrbeck 2017).

Many recent technological innovations are focused on utilising data. Analysing and utilising data is becoming increasingly vital for firms to stay competitive and to survive in the long-term (Brownlow et al.

2015; Hartmann et al. 2016; LaValle et al. 2011). Hunke et al. (2017) highlight the opportunity for firms to make data the central value offering of the firm’s business model. By enriching their business model with data, firms are more likely to stay competitive within their industry (LaValle et al. 2011). The main focus is on utilisation of large data sets, commonly referred to as big data (Gandomi and Haider 2015), with data analysis playing an important role in achieving competitive advantage (Morabito 2015). The space industry is characterised by advanced technology and is operating in a data-heavy environment, with increased big data generation. In line with aforementioned, data utilisation is thus an important part of future business (Hunke et al. 2017). Even industries without data-heavy settings are realising the potential of data utilisation (Sch¨uritz and Satzger 2016). Moreover, the satellite industry is facing an increasing amount of big data, contributing to data utilisation being a major market trend (Soille, Loekken and Albani 2019).

Today the space industry is not only about races to the Moon, pioneering exploration and new discover- ies; it is also about contributing to technological development and finding solutions to global challenges.

Thanks to emerging technologies, the application areas for space-enabled technology are rapidly increasing and becoming part of people’s every day life. It is a new space era, characterised by high-speed change which presents new business opportunities. Traditionally, the space industry was characterised by large government programs, funded by governmental and institutional capital, resulting in a rather slow moving industry with high entry barriers. Now, the new era brings forth innovative technologies and new commercial entrants, leading to a market shift towards a commercial and competitive global market with lower entry barriers. This is driving the market towards lower prices, new applications and a higher speed of change. In recent years, more and more satellites have been launched into orbit. With the development of a new generation of small satellites, the cost of building and launching satellites has dropped, leading to a growing commercial interest and new actors entering the market (Simonis 2019).

The development is having a disruptive impact on the satellite management service sector, resulting in

(13)

diminishing profit margins on satellite management services. New commercial customers, often called

’new space’, are pressing prices downwards which is affecting pricing within the sector as a whole. To offer services within the satellite industry is thereby becoming more challenging. The incumbents within the sector are characterised by their value offering, focused on providing high-quality services. There is currently little differentiation amongst the incumbents, with the main distinction between incumbent firms being the added services that can be offered. However, with the industry-wide disruption, the sector of satellite management services is also facing new entrances and increased competition. In 2018, a capital-heavy IT firm announced plans to build their own ground station network. In light of this development, incumbent firms offering satellite management services need to ensure their competitiveness by exploring new business opportunities. Emerging technologies for processing satellite data into a variety of different products and services open up possibilities to reach new markets. Thus, there is an emerging opportunity within the satellite management sector to move up in the value chain by utilising satellite data to expand the current value offering.

1.2 Problematisation and Scientific Contribution

The changes in the space industry are affecting even established technology firms with viable business models, within all areas of the industry. The satellite management sector, which is characterised by long-time players who operate using established business models, is highly affected by the emerging ’new space’ customer segment and the prices being pressed downwards. Thus, within the sector, existing business models need to stay competitive and new business areas must emerge in order to not risk firm obsoletion in the long term. By engaging in business model innovation, the shifting market can be met and new business opportunities seized. Additionally, within both the research community and the satellite industry, there is a novel area of data utilisation receiving increased attention. Innovating a business model by infusing data, where data is seen as the main resource, is a way for the firm to stay competitive.

A major trend in the satellite industry is data utilisation in the form of satellite data processing, which makes the area suitable to investigate as a way of innovating an existing business model within the satellite management industry. To succeed with such an endeavour, the existing business model must stay operative while the firm simultaneously explores new business model alternatives that incorporate data utilisation.

Within the literature, there is a noticeable research gap regarding data driven business models, due to its novelty as a research area. In particular, there is a lack of in-depth case studies regarding data driven business models and the transition from an existing business model into a data driven one. Moreover, there is a great deal of research done regarding the space industry and its technological aspects. However, due to the industry previously being dominated by governmental agencies, there is little research connecting the technological and business aspects of the industry. For instance, no research connecting business models with satellite data has yet been conducted within the satellite industry. Due to the increasing importance of data within the industry, research regarding data driven business models and their adaptation to satellite data is of high relevance. This area therefore needs to be examined to establish a link between business models and satellite data, furthering a academic discourse on the subject.

1.3 Purpose

The purpose of this thesis is to explore how novel ways of data utilisation can drive business model innovation, and investigate how a space firm can exploit their incumbent business model while simultaneously exploring new alternatives.

(14)

1.4 Research Questions

To fulfil the stated purpose, a main research question, as well as two sub-questions, are formulated.

MRQ: How can novel data utilisation drive business model innovation for an established technology firm still reaping benefits from its incumbent business model?

SQ1: How does novel data utilisation, in the form of data processing, contribute to business model innovation?

SQ2: How can an incumbent space firm explore new business models while simultaneously exploiting their existing business model?

1.5 Delimitations

The thesis will only explore implementation from a business model perspective, and not address the change management perspective on implementing new business practices. The produced insights will be applicable on other industries, but only the satellite data industry will be investigated. The thesis is limited to an embedded study at SSC, with emphasis on the Satellite Management Services business unit, one out of total three units. However, empirical data has also been obtained from other units, to enable a deeper understanding. Finally, the thesis will only investigate the firm from an internal perspective, not taking into account the customer perspective.

(15)

2 Theoretical Framework

In this section, an in-depth theoretical background of business models, as well as business model innovation and ambidexterity, is presented. Use of data, in relation to business models, is also outlined.

2.1 Business Models

When exploring the term ’business model’, it becomes noticeable that a commonly accepted definition does not exist. Originally associated with web-based firms, the increased importance of information and communication technology (ICT) for other types of firms has lead to the business model concept becoming more widespread (Morris, Schindehutte and Allen 2005; DaSilva and Trkman 2014). Several papers have concluded the lack of a commonly accepted definition of what a business model is (Morris, Schindehutte and Allen 2005; Huarng 2013; DaSilva and Trkman 2014; Zott, Amit and Massa 2011), with Zott, Amit and Massa (2011) noting that because literature on the subject is produced within separate fields, it hinders the development of cumulative research on the topic of business models. Out of the 103 publications on business models studied by Zott, Amit and Massa (2011), 37 % did not provide any definition of the term ’business model’ at all, which suggests that many believe the term to be self-explanatory. Nevertheless, several researchers have proposed frameworks aimed at identifying and describing the components of a business model (cf. Wirtz, G¨ottel and Daiser (2016)).

2.1.1 Business Model Frameworks

Even though business model frameworks have different perspectives on the components of a business model, there are still several common denominators. In Table 1, common elements from eight frameworks are shown and later described in more detail.

Business Model Element

Value Customer/market Cost Model Revenue Model Activities Resources Others

Chesbrough &

Rosenbloom (2002) D D D D

Value chain, network, strategy

Hedman &

Kalling (2003) D D D D Competitors,

supply

Mitchell &

Coles (2003) D D D D Pricing

Morris et al. &

(2005) D D D D D Internal advantage,

ambition

Zott & Amit

(2010) D D Design themes

Teece (2010) D D D D Strategy filter

Osterwalder &

Pigneur (2010) D D D D D D

Relationships, channels, partnerships

Mason &

Spring (2011) D D D Technology,

architecture

Table 1: Business Model Elements. Adapted from Hartmann et al. (2016).

One of the early business model frameworks is proposed by Chesbrough and Rosenbloom (2002), who define a business model as performing six functions; articulating a value proposition, identifying market

(16)

segments, defining the firm’s value chain, estimating the firm’s cost structure and profit potential, identifying the firm’s place within the value network, and formulating the firm’s strategy. The business model is developed by mapping out these functions in the described order.

Hedman and Kalling (2003) study previous research on business and strategy and build upon their findings to develop a framework consisting of six components. These components are customers, competitors, value offering, firm activities and organisation, resources, and supply of input from the capital and labour market. A seventh ’longitudinal process’ component that focuses on changes over time is also added to ensure that the framework is adapted for dynamic nature of a firm.

Mitchell and Coles (2003), when looking into business model innovation, define the business model as consisting of five distinct elements. The first one is the ’who’, as in who are the customer that the firm serves. The second is the ’what’, meaning what value proposition the firm is offering. The third is the

’where’, meaning where geographically is the firm operating. The fourth is the ’how’, as in how will the firm deliver its value (through what activities). The fifth and final is ’how much’, as in how much will it cost for the customer.

Morris, Schindehutte and Allen (2005) develop a framework that is divided into three levels, starting from a broad perspective and becoming increasingly firm-specific. At the foundation level, the basic decisions for the firm are outlined. Morris, Schindehutte and Allen (2005) identify six components that are needed to produce this set of decisions; value creation, identification of customers, internal sources of advantages, positioning in the marketplace, economic model, and ambition of the entrepreneur. At the proprietary level, the firm must focus on finding innovative ways of achieving the basic decisions outlined in the previous level. It is at this level the firm creates advantages that cannot be replicated, as the proprietary level is strategy-specific and builds upon creating unique combinations of the variables at the foundation level. At the rules level, a set of rules or guidelines are defined that will guide the strategic decisions and actions of the entire firm. These rules will ensure that all strategic decisions are linked to the foundation and proprietary level.

A concept called Business Model Canvas (BMC) was first introduced by Osterwalder and Pigneur (2010), and is derived from Osterwalder’s previous research on business models. Here, the business model is developed further which results in a tool promoting visual thinking when creating or mapping business models. The tool consists of nine building blocks; value proposition, customer segments, customer relationships, channels, revenue streams, cost structures, key activities, key resources and key partnerships.

It was created with the intent of providing a standardised approach for designing or mapping business models and resulted in a comprehensive framework. According to Cosenz (2017), the BMC has been widely recommended in recent business modelling literature, as well as by academic incubators and ven- ture capital associations worldwide. The main reason for the BMC framework’s popularity, Osterwalder and Pigneur (2010) and Sort and Nielsen (2018) argue, is that it provides a clearer understanding of how a firm creates value. Furthermore, Osterwalder and Pigneur (2010) claim that the concept has been tested and used around the world, for instance by organisations like IBM, Ericsson and Deloitte. Zott and Amit (2010) describe business models from an activity system perspective, where the business model is seen as an activity system, and the main objective to generate value by taking advantage of a business opportunity. An activity is defined as using resources, such as human or capital resources, to achieve value creation. The term activity system is defined as a ’system of independent activities that transcends the focal firm and spans its boundaries’ (Zott and Amit 2010, p. 217). To design an activity system, two

(17)

sets of parameters need to be analysed. The first set is called design elements, which consists of con- tent (selection of activities to be performed), structure (how activities are linked and their importance) and governance (who will perform the activities). The second set of parameters is called design themes, which are configurations of the design elements used for identifying what drives value creation within the activity system.

According to Teece (2010, p. 179), a business model ’articulates the logic, the data and other evidence that support a value proposition for the customer, and a viable structure of revenues and costs for the enterprise delivering that value’. Five elements of business model design are outlined. When combined, these elements will lead to value creation for the firm’s customers, which will in turn lead to payments that are converted into profits. The five elements are: identification of benefit for customers, market segmentation, confirmation of available revenue streams, design of value-capturing mechanisms, and selection of technologies and features to be embedded in the product or service. One key aspect of creating a business model that has a sustainable competitive advantage, according to Teece (2010), is strategic analysis. A business model on its own will not be enough to create competitive advantage, but when coupled with firm-specific strategic analysis, the firm’s activities becomes hard to imitate.

Mason and Spring (2011) examine the concept of business models, analysing previous literature on the subject to reach a conclusion regarding the core elements of a business model. The underlying understanding of what a business models is can be said to be a description of the way a particular business work. Mason and Spring (2011) argue that the value of a business model originates from capturing actions and the connections between them, which provides a shared understanding of the firm’s actions.

The business model framework that Mason and Spring (2011) arrive at consists of three core elements.

The first element is technology, in the form of a product, process, core or infrastructure, and its delivery and management. The second element is market offering, i.e., what really is offered to the customers and how it is offered. The offering can be in the form of artefacts, activities or access. The third element is network architecture, in the form of markets and standards, transactions, capabilities and relationships.

Hence, the network of suppliers and buyers that make the market offering possible.

2.2 Business Model Innovation

Intensified competition on the global market has lead to a growing interest into how firms can stay competitive in dynamic, fast-changing markets (Wirtz, G¨ottel and Daiser 2016). The notion of innovating a firm’s business model to adapt to shifting market conditions has therefore gained prominence in recent years (Wirtz, G¨ottel and Daiser 2016). Using examples of technology invented at Xerox that the company was unable to utilise, but later became successful spin-offs, Chesbrough and Rosenbloom (2002) provide one of the earliest links between innovation and the business model concept, concluding that a viable business model is crucial for extracting value from technological innovations. Several authors has since demonstrated the connection between business model innovation and superior performance, and shown that successful business model innovation can be linked to sustainable competitive advantages (Amit and Zott 2012; Casadesus-Masanell and Zhu 2013; Mitchell and Coles 2003).

The novelty of business model innovation means that consensus on the phenomenon is still lacking, precipitating considerable heterogeneity of the concept’s definition in published literature (Wirtz, G¨ottel and Daiser 2016). One description proposed by Amit and Zott (2012) builds upon their activity-based view of the business model (presented in the previous section), conceptualising business model innovation as changes made to one or more of the business model’s core elements. Mitchell and Coles (2003) instead

(18)

define business model innovation as the substitution of a majority of the current business model elements by completely novel ones, dubbing this ’business model replacement’. Gambardella and McGahan (2010) choose to conceptualise business model innovation as the adoption of new ways to commercialise the firm’s fundamental assets, resulting in their continued relevance.

Chesbrough (2010) highlights one aspect of the challenges associated with business model innovation:

many times, managers simply do not know what the right business model for their firm is, and are unsure of how to find a suitable model. When a firm strives for business model innovation, Chesbrough (2010) therefore suggests mapping out the business using frameworks such as the business model canvas proposed by Osterwalder (2004). By mapping out the current business model components, the firm can more easily construct experiments to test new business models and ideas. These experiments should be designed to promote cumulative learning within the organisation, a notion reiterated by Kaulio, Thor´en and Rohrbeck (2017) and Sosna, Trevinyo-Rodriguez and Velamuri (2010). In an extensive case study, Sosna, Trevinyo-Rodriguez and Velamuri (2010) illustrate how a trial-and-error approach to business model innovation leads to sustainable competitive advantage. Building an organisation that has a positive view on experimentation and sees failure as an opportunity for learning creates favourable conditions for successful business model innovation, according to Sosna, Trevinyo-Rodriguez and Velamuri (2010). The authors also highlight that the firm’s response to early failure is an important factor in how the trial- and-error process within the firm will subsequently develop.

Teece (2010) instead focus on strategy as a success factor for business model innovation, expressing that business model innovation must take into account the overall company strategy in order to be successful.

To ensure a proper fit, the business model must pass through a ’strategic filter’. The notion of strategy as an important differentiator is echoed by DaSilva and Trkman (2014) as well, who argue that a well- planned corporate strategy enables dynamic capabilities, which in turn facilitates transformation of the business model.

Christensen, Bartman and Van Bever (2016) investigate the interdependencies between business model elements, arguing that business models are not designed for change. This view is further supported by Amit and Zott (2012), who highlight interdependencies between business model elements as well as interdependencies between the business model itself and the firm’s revenue model. Additionally, Mitchell and Coles (2003) argue that companies that are efficient in reducing costs and streamlining their current business model are less likely to achieve continuous business model innovation. Improving upon the current business model means strengthening the interdependencies, which in turn leads to a lower degree of flexibility since one element cannot be changed without influencing the entire organisation (Christensen, Bartman and Van Bever 2016; Mitchell and Coles 2003). Sosna, Trevinyo-Rodriguez and Velamuri (2010) however, note that business models are frequently revisited and revised by management, and instead present a dynamic view on the business model as a product of continuous trial-and-error learning.

According to Christensen, Bartman and Van Bever (2016), the business model of any organisation follows a three-stage evolution, becoming less flexible for each stage (see figure 1). At the creation stage, the emphasis is on value creation and the organisation focuses almost exclusively on customer needs. No routine exists yet and the organisation usually consists of a small team that is in an exploratory mode. If the team is successful in creating a value proposition, the next stage is sustaining innovation. Processes are repeated and a routine begins to form, which means that the processes are no longer a flexible element.

(19)

However, there is still room for flexibility regarding how profits should be made. Finally, the organisation evolves into the efficiency stage. By standardising processes the organisation can reduce costs and gain efficiency, but by doing so the interdepencencies are strengthened and flexibility is lost. Since the focus is on the bottom-line and generating high return on investment with as little risk as possible, managers are unlikely to favour value creating innovation over cost-reduction and innovation to improve efficiency (Christensen, Bartman and Van Bever 2016).

Figure 1: The Three Stages of a Business Model’s Journey. Christensen, Bartman and Van Bever (2016)

2.3 Organisational Ambidexterity

The topic of how firms can manage their incumbent business model while simultaneously exploring potential new business models is the central theme of the academic field of organisational ambidexterity.

One early account of the phenomenon is provided by Tushman and O’Reilly (1996), who describe ambidexterity as the ability to compete in a mature market while simultaneously developing new products or services. The former requires a focus on cost efficient and incremental innovation, while the latter instead demands speed, flexibility and radical innovation. Gibson and Birkinshaw (2004) define organisational ambidexterity as the ability to achieve alignment with today’s business demands while simultaneously adapting to the future demands. Kaulio, Thor´en and Rohrbeck (2017) describe organisational ambidexterity as an organisation managing evolutionary and revolutionary change at the same time to ensure prolonged survival and success. Several researchers show that an organisation has to be ambidextrous in order to be successful when existing in a dynamic environment, exploiting the current business opportunities while still exploring potential future opportunities (Gibson and Birkinshaw 2004; He and Wong 2004; Tushman and O’Reilly 1996).

Markides (2013) proposes that organisational ambidexterity literature can be leveraged to explore business model innovation. However, there are differing views on how to achieve ambidexterity within a firm (Markides 2013). Markides (2013) identifies three distinct solutions for the ambidexterity challenge;

spatial separation, temporal separation and contextual ambidexterity. Spatial separation means that the firm separates innovation efforts that are radically different from the firm’s current operations, usually by creating separate business units. Tushman and O’Reilly (1996) argue that an organisational architecture consisting of small, autonomous business units is a vital aspect of achieving ambidexterity. Applying the same logic on the business model field, Christensen, Bartman and Van Bever (2016) proposes that new business models that do not align with the firm’s incumbent business model should be managed separately in a new business unit to avoid conflict and create the conditions necessary to develop the new business model. Temporal separation utilises a similar logic, but instead of separating conflicting activities using organisational structure, the activities are performed at different points in time (Markides 2013).

Contextual ambidexterity was first proposed by Gibson and Birkinshaw (2004), describing it as a perceived

(20)

conflict between alignment and adaptability within an organisation. Opposing the ’trade-off’ view on alignment and adaptability which leads to spatial or temporal separation, the authors instead argue that ambidexterity is best achieved by designing a context within the firm that allows for individual judgement of what activities to perform. Gibson and Birkinshaw (2004) present four dimensions (discipline, stretch, support, and trust) that are vital in creating said context. Discipline refers to the creation of clear standards, rapid feedback and consistency in rules. Stretch is the willingness of employees to strive beyond the ’bare minimum’ and towards more ambitious goals, which is achieved by creating a sense of individual contribution to a larger goal. Support refers to an environment where employees assist each other and share resources. Trust is the sense of relying on the commitment of others, and is achieved by creating a sense of ’fairness’ within the organisation. Too little discipline and stretch leads to a ’country club’ environment, while a lack of trust and support leads to overworked and disillusioned employees, so these four dimensions need to be properly balanced in order to create optimal conditions for achieving ambidexterity (Gibson and Birkinshaw 2004). Gibson and Birkinshaw (2004) further criticises temporal and spatial separation (referring to them collectively as ’structural separation’), arguing that they lead to an increase in coordination costs.

Ambidexterity has also been used investigate the balance of exploration and exploration outside a firm- level perspective. Holmqvist (2004) illustrates how the balance of exploration and exploitation within a firm is connected to exploration and exploitation between organisations. Kauppila (2010) further shows how interorganisational partnerships can aid firms in achieving radical innovation. Further advantages of interorganisational partnerships are reduced risks, access to complimentary skills and knowledge, and access to novel technologies or new markets (Mohr and Spekman 1994). However, Mohr and Spekman (1994) also highlight negative consequences in the form of lost autonomy and information asymmetry.

Drawing upon previous research on ambidexterity and business model innovation, recent research has highlighted the need for double ambidexterity (Kaulio, Thor´en and Rohrbeck 2017; Tongur and Engwall 2014). The concept is described by Tongur and Engwall (2014, p. 534) as ’not just the ambidexterity to simultaneously foster incremental and radical innovation, but also the ambidexterity to simultaneously advance both technological and business model innovation’. Kaulio, Thor´en and Rohrbeck (2017) further the concept by investigating the interplay between business model innovation, technological innovation, exploration and exploitation using a framework illustrated in figure 2.

(21)

Figure 2: Double Ambidexterity Framework. Adapted from Kaulio, Thor´en and Rohrbeck (2017).

The framework is used to map organisational responses to disruptions over time, highlighting the vari- ation in how a firm may react to different types of turbulence within their industry. Responses are categorised using two principal dimensions, technology and business model innovation, which are divided into exploitative and exploratory actions, respectively. To further nuance the categorisation, exploratory responses are further divided into incremental and radical responses. However, it is important to note that radical innovation is not necessarily preferable over incremental. Sorescu (2017) notes that many successful business model innovations are not radical and disruptive in nature, and that incremental business model innovation also provides potential for competitive advantages. The criteria of categorisation are detailed in Table 2.

Technology Business Model

Exploitation

Closely related to existing technology or minor adaption of current technology

Minor adjustment or fine tuning of one or several of the business model’s elements.

Exploration (incremental)

Substantial change in technology, significant improvement of existing product, process, or service.

Significant improvement or upgrade of existing product, process, or service.

Exploration (radical)

Substantial change in technology, unprecedented performance features of product, process, or service, or drastic changes that enable new application domains.

Unprecedented performance features of product, process, or service, or drastic changes that enable new application domains.

Table 2: Categorisation of Organisational Responses. Adapted from Kaulio, Thor´en and Rohrbeck (2017).

Kaulio, Thor´en and Rohrbeck (2017) highlight that the categorisation should be based on the organisation’s action, and not the outcome of said action. The authors also illustrate the importance of a

(22)

longitudinal approach when investigating business model innovation, arguing that diffusion of innovation within the firm needs to be accounted for. The longitudinal approach allows for the identification of three distinct response patterns that occur either due to a market disruption, a technology disruption or a combination of the two. Kaulio, Thor´en and Rohrbeck (2017) show that market disruptions pro- voke exploitative responses, while technological disruptions instead induce exploratory responses. When facing a combination of technological and market disruption, exploratory responses occurs mainly on the business model axis, while the technological axis consists of mainly exploitative responses. Kaulio, Thor´en and Rohrbeck (2017) note that the focal firm becomes more willing to partner with other firms and thereby open up their innovation process when facing a combined technological and market disruption, even if the focal firm previously has had a tradition of in-house value creation. Furthermore, their work demonstrates the need for further research on the interplay between technological innovation and business model innovation.

2.4 Data Driven Business Models

Acquiring, analysing and applying various types of data is seen as increasingly vital for businesses to not only stay competitive, but to survive in the long-term (Brownlow et al. 2015; Hartmann et al. 2016;

LaValle et al. 2011). This highlights the possibility to enrich existing business models with data utilisation and moving towards making data the central value offering of the firm (Hunke et al. 2017).

Brynjolfsson, Hitt and Kim (2011) show a positive correlation between data driven practices and firm performance, indicating that incorporating data utilisation into the firm’s current practices can improve output and productivity. LaValle et al. (2011) show that firms who identify data utilisation as their main source of differentiation are twice as likely to be top performers within their industry. Thus, there is currently a strong focus on how firms can incorporate data into their businesses, with analysis of large data sets becoming a focal point in this endeavour (Gandomi and Haider 2015). Hunke et al. (2017) highlight that firms can stay ahead of competitors by using business models exploiting large volumes of data. Additionally, giving data a more central role within an organisation may improve the value creation of the firm, with data analytics playing an important role for firms aiming to stay ahead of their competitors (Morabito 2015).

One of the main obstacles to incorporating data into a firm’s business is that firms are unsure of how to utilise data for value-adding purposes (Hartmann et al. 2016; LaValle et al. 2011). Therefore, new models focused on capturing the value of data are becoming increasingly important. These new business models, frequently called ’data driven business models’, constitute a new field of research which still lacks widely accepted definitions (Sch¨uritz and Satzger 2016). Hartmann et al. (2016, p. 1385) choose to define a data driven business model as ’a business model that relies on data as a key resource’. Morabito (2015, p.65) uses a similar definition, characterising data driven business models as business models that ’rely on big data to achieve a key value proposition’.

Sch¨uritz and Satzger (2016) take a different stance, arguing that there is no such thing as a data driven business model. Instead, incorporating data into the firm’s activities leads to a range of alternative business model transformations, depending on what element of the business model that is affected. The business model is condensed into three elements; value creation, value proposition and value capturing.

Value creation is described as the arrangement of activities, processes and resources needed to create and deliver the value proposition to the firm’s customers. By enriching existing products or services, streamlining current operations to reduce costs, or offering a new product or service, the firm’s value

(23)

creation is expanded (Sch¨uritz and Satzger 2016). Value proposition is the actual value offered to the firm’s customers or stakeholders. Value capturing is how the firm turns the value proposition into mon- etary value for the firm itself. It can be performed by identifying novel revenue streams or reaching new customer segments. This creates five potential patterns of data infusion, presented in Table 3.

Pattern Value Creation Value Proposition Value Capturing

I: Data-Infused Value

Creation D

II: Data-Infused Value

Capturing D

III: Data-Infused Value

Proposition via Value Creation D D

IV: Data-Infused Value

Proposition via Value Capturing D D

V: New Data-Infused Business

Model (DiBM) D D D

Table 3: Patterns of Data-Infused Business Models. Adapted from Sch¨uritz and Satzger (2016).

Sch¨uritz and Satzger (2016) also claim that new technology, such as technology aimed at novel data utilisation, can drive business model innovation and help the firm find new sources of value. However, Sch¨uritz and Satzger (2016) stress that the firm should not focus on how to turn the existing business model into a data driven one, but instead focus on novel data utilisation and let this process guide the firm to more innovative practices.

Data utilisation and analytics are now impacting a broad range of industries (Hunke et al. 2017). This development means that firms across various industries are becoming increasingly aware of the importance of data. Examples of successful implementation of novel data utilisation can be found in industries such as traditional manufacturing, raw material processing and retail (Sch¨uritz and Satzger 2016). Within the satellite sector, the utilisation and analysis of satellite data is one of the focal points of the industry’s development (Soille, Loekken and Albani 2019).

2.5 Data from Satellites

Traditionally, the satellite industry has had high entry barriers and has therefore been dominated by large space agencies. However, in recent years the cost of building and launching satellites has dropped considerably, which has lead to a growing commercial interest within the field (Simonis 2019). One trend within the satellite industry is the increased deployment of small satellites (Sandau 2010). Using a constellation of several small satellites instead of a solitary large satellite has several advantages, such as more frequent mission opportunities and faster adaptation to technological developments (Sandau 2010).

The commercialisation of space has lead to a wide array of new entrants on the market, both in the form of start-up firms and established, capital-heavy IT firms looking to expand their business into space (Denis et al. 2017). As costs for building and launching satellites decreases, so do the entry barriers into the industry. As a result, the satellite industry is currently undergoing a disruptive transformation with

(24)

an uncertain outcome (Denis et al. 2017).

The rising number of satellites in orbit means a greater amount of data being generated, which poses a challenge since these data volumes need to be analysed in a quick and efficient manner (Milcinski et al.

2019). Denis et al. (2017) describes how the antenna systems on the ground used to receive satellite data, the so-called ground station infrastructure, will become of increased importance as the amount of generated data grows. Such large data sets are often referred to as ’big data’, a concept that has gained prominence in several industries in recent years (Gandomi and Haider 2015).

2.5.1 Big Data as a Concept

The term ’big data’ is a relatively new concept that is not well defined by academics (Gandomi and Haider 2015). In a global survey of IT and business professionals, Schroeck et al. (2012) found that there was considerable confusion as to what big data actually is. Laney (2001) suggests an approach consisting of three V’s; Volume, Velocity and Variety. This approach has been widely used and expanded upon to convey the concept of big data (McAfee et al. 2012; Gandomi and Haider 2015; Hartmann et al. 2016). A fourth V, veracity, is often added to the three others (Schroeck et al. 2012). According to Sorescu (2017), the three V’s can be found within business models that are achieving competitive advantages.

The first V, volume, refers to the amount of data transferred and stored (Gandomi and Haider 2015;

Schroeck et al. 2012). Big data, as the name indicates, involves handling large amounts of data. The amount of generated data is growing at an increased speed and is estimated to reach over 44 trillion gigabytes in 2020 (Sch¨uritz and Satzger 2016). Velocity refers to the speed of data creation, transfer and analysis (Gandomi and Haider 2015; Schroeck et al. 2012). The popularity of smartphones and similar devices has increased the pressure on processing data quickly, since large amounts of the large amount of data generated by mobile devices presents an opportunity for retailers and marketers (Gandomi and Haider 2015). Variety refers to the heterogeneity of the data that is generated and analysed (Gandomi and Haider 2015; Schroeck et al. 2012). Companies must be able to process and analyse both structured and unstructured data, from a variety of sources and in a wide range of formats, in order to utilise it. Veracity refers to the varying reliability of data (Gandomi and Haider 2015; Schroeck et al. 2012).

Unreliable data is now commonplace due to the large amounts generated. One dimension of big data is therefore the ability to handle uncertainty that is built into the data. Data fusion of multiple sources of unreliable data is one way to reduce uncertainty and facilitate more accurate data analysis (Schroeck et al. 2012).

The four V’s are highly applicable to satellite data, since it consists of large volumes, is required at a high speed, with varying reliability (Baumann et al. 2016). Even though all satellite data originates from satellites, it is still heterogeneous due to a lack of standards within the industry, with custom solutions for various satellites being commonplace. Furthermore, big data from satellites has the added difficulty of requiring further conversion into several, distinct layers of information in order to represent the information to its full extent (Tiede et al. 2019). Many satellite owners and satellite data customers lack the expertise and resources for processing and analysing large data sets originating from satellites, which constitutes an obstacle for widespread utilisation of satellite data (Siqueira et al. 2019).

2.5.2 Satellite Data Processing

As highlighted by several researchers, artificial intelligence and machine learning applications are starting to play key roles within the field of Earth observation (Moumtzidou et al. 2019; Sumbul, Demir and Markl 2019; Datcu et al. 2019). Machine learning is becoming a necessity due to the volume, availability and

(25)

quality of satellite data (Milcinski et al. 2019). While there are many machine learning options available for general imagery, not many support the complexity of EO data (Milcinski et al. 2019). Additionally, there is an increased demand for data that has been processed and is ready to use for the end customer (Siqueira et al. 2019). Since transfer of large amounts of data is resource-consuming, one important aspect of efficient data handling is to perform the processing close to the source of the data (T. Huang 2019; Neteler et al. 2019).

There are several levels of processing for satellite data. NASA’s Earth Observation System (EOS) has developed a classification system for processing that has become widely used (Y. Huang et al. 2018;

Piwowar 2001). This classification system is detailed in Table 4.

Level Description

0 Reconstructed, unprocessed instrument data. Full resolution, no information lost. All communications artifacts (synchronisation frames, communications headers, et cetera) are removed.

1A

Reconstructed, unprocessed instrument data at full resolution, time-referenced, and annotated with ancillary information, including radiometric and geometric calibration coefficients and georeferencing parameters. Level 0 data are fully recoverable from level 1A data (no loss).

1B Level 1A data that have been processed to sensor units (radar backscatter cross section, brightness temperature, optical, et cetera). Level 0 data are not recoverable from Level 1B data.

2 Derived geophysical variables, such as sea ice concentration, ocean wave height, et cetera. At the same resolution and location as Level 1A source data.

3 Variables mapped on uniform spatial grid scales, usually with some completeness and consistency (missing points interpolated, complete regions mosaicked together from multiple orbits, et cetera).

4 Model output or results from analyses of lower level data (i.e., variables that were not measured by the instruments but instead are derived from these measurements).

Table 4: Levels of Data Processing. Adapted from Parkinson, Ward and King (2006).

(26)

3 Conceptual Framework

In this chapter, a conceptual framework that presents the relevant frameworks for the thesis is outlined.

Arguments for why the presented frameworks are relevant for this thesis are also given.

3.1 Establishing the Business Model Elements

In order to build the thesis on a generalisable framework with a strong theoretical foundation, the business model is defined as consisting of the six elements outlined in the previous chapter, found in Table 1. These six dimensions were chosen due to the lack of a commonly agreed upon definition, thus leading us to choose the most agreed upon elements in order to attain a reliable definition. Furthermore, a definition of a framework for a business model was needed to fulfil the purpose and answer the first sub-question. Hence, this categorisation enabled the mapping of the firm’s current business model using a clearly structured framework, facilitating the identification of potential new business models, in accordance with Chesbrough (2010). We chose to define the meaning of our business model elements in line with Osterwalder and Pigneur’s (2010) elements, since their work is designed primarily for practical applications rather than academic purposes. It is therefore easy to apply and use their general classifications that are suitable for a wide range of firms.

Following are the six elements briefly described in accordance with Osterwalder and Pigneur (2010):

1. Value Proposition

The value delivered to a customer, helping to solve a problem or fulfil a need. This is achieved by products or services. Value can be qualitative or quantitative, meaning it can be created for instance in the form of customer experience or price. A value proposition could satisfy an entirely new customer need, which many times is the case regarding technology related products. Accessibility is one example of a value proposition, meaning that customers who previously lacked access to a product or service are offered availability. A firm can also create value by ’getting the job done’, meaning that the firm takes care of the customers needs without hassle, as exemplified by Rolls-Royce and how they provide manufacturing and full service of jet engines, which allows their customers to focus on other aspects of their business.

2. Customers

The different groups of people or organisations the firm intends to reach and serve by it’s value proposition. Customers are grouped into different segments sharing common needs, behaviours, relationship requirements or willingness to pay for the same aspect of the offer. There could be one to several customer segments defined within a business model, which need’s are important to understand to have a successful business model.

3. Revenue Model

The revenue generated from each customer segment, by delivering value. There can be several revenue streams from the same customer segment, however the revenue streams themselves are either one time occurring as a result from one-time customer payment, or reoccurring as ongoing payments for value delivery or post-purchase support.

4. Cost Structure

(27)

The most important costs incurred to operate a business model and its elements. A business model can be more or less cost-driven, which intent is to minimise costs wherever possible. In opposite, a value-driven business model is less concerned with costs and more focused on value creation.

5. Key Activities

The most important activities or actions a firm must take to keep the business model operational and successful. Thus, the most important activities to be able to create the value proposition, generate revenue and keep customers.

6. Key Resources

The most important assets required to make a business model work. Key resources can be in the form financial, human, intellectual or physical, and the resources can be owned, leased or accessed by partnerships.

It is important to note that this type of business model framework represents a static state, while business model innovation is a dynamic process (Sosna, Trevinyo-Rodriguez and Velamuri 2010). However, the framework was in spite of this chosen because it provided a useful way of comparing the business model in two different points in time. In order to incorporate the dynamic aspects of the business model, a longitudinal component was added, similar to what is suggested by Hedman and Kalling (2003). This is further strengthened by Kaulio, Thor´en and Rohrbeck (2017), who stress the importance of longitudinal approaches as well as that of the contextual setting. This made us expand our business model framework.

Further reasoning for this is that by placing the model in a longitudinal setting, the business model transformation can be more accurately described, strengthening the validity of our work. The organisational ambidexterity theory was leveraged to capture processes occurring during the longitudinal component, as the firm shifts from its current to its future business model.

In summary, our proposed business model framework is connected to a dynamic aspect by being intertwined with organisational ambidexterity, and the business model elements are summed up in figure 3.

Figure 3: Summary of the Business Model Elements.

(28)

3.1.1 Innovation of a Business Model

Building on the previously described framework, aspects of innovation also had to be considered due to the thesis’ purpose of innovating a business model. Hence, the following further defines our business model framework by adding innovation aspects, once again intertwined with organisational ambidexterity.

The aggregated literature on business model innovation does not agree on a mutual understanding or definition of what constitutes innovation. To enable a structured analysis, a definition of business model innovation in accordance with Amit and Zott (2012) was selected, since their definition is widely cited which provides credibility. Business model innovation was thus conceptualised as changes made to one or more of the business model’s core elements. Furthermore, the double ambidexterity framework presented by Kaulio, Thor´en and Rohrbeck (2017) was chosen and take into consideration when analysing business model innovation. This choice was made in order to gain a deeper understanding and identify links between business model innovation and organisational ambidexterity. By using the framework of Kaulio, Thor´en and Rohrbeck (2017) another dimension was added to the analysis, enabling the evaluation of innovation as being exploitative or exploratory. Exploratory innovation was further categorised as either incremental or radical, which provided the analysis with a greater degree of nuance.

3.2 Dynamics of Data Utilisation and Ambidexterity

In the aggregated literature on data driven business models, the implicit understanding is that data driven or data-infused business models rely on utilisation of data to create value. This distinction has a significant impact for our thesis, since the investigated firm has a value offering that is centred on delivering data.

However, the firm does not actually utilise data to create value for its customers. Therefore, delivery of data can be likened to delivery of a physical good or service, in the sense that simply delivering data does not require a data driven business model. Such a distinction enabled us to postulate that the firm’s current business model is not data driven, thus enabling the application of a framework for data driven business model innovation.

Furthermore, in accordance with Schüritz and Satzger (2016), novel data utilisation is equated to technological innovation in the sense that both act as drivers for business model innovation. Connecting novel data utilisation with technological innovation made the framework on double ambidexterity presented by Kaulio, Thorén and Rohrbeck (2017) a suitable selection for this thesis, since it explores the interplay between technological innovation and business model innovation. The framework developed by Kaulio, Thorén and Rohrbeck (2017) was also selected since it is used to map organisational responses. When the focal firm’s earlier responses were mapped, evaluation of how novel data utilisation compares to previous organisational responses was made possible. The double ambidexterity framework was also used to map novel data utilisation as an organisational response (see figure 4), further strengthening its suitability.

(29)

Figure 4: Framework Illustrating Organisational Response and its Corresponding Business Model.

Since the result of novel data utilisation within the business model was investigated, the patterns of data- infused business model innovation developed by Schüritz and Satzger (2016) were used. These patterns were identified using a wide range of firms, and was therefore deemed suitable for the investigation of the focal firm of this study. The elements described by Schüritz and Satzger (2016) were translated into the six elements of our previously outlined business model framework (see figure 5). This translation was necessary to assess what degree of business model transformation was achieved by novel data utilisation, by investigating the change to each of the business model’s underlying elements and categorising the resulting business model using the patterns described by Schüritz and Satzger (2016). Due to the novelty of the research field, this framework was the first of its kind and was therefore a natural choice.

Figure 5: Translation of the Business Model Elements Outlined by Sch¨uritz and Satzger (2016).

(30)

4 Method

This chapter presents the scientific method used for the research process conducted in this thesis. The method guided our research process and was used for gathering necessary empirical data to fulfil the purpose of the thesis. The choice of method is described in detail and critically argued for.

4.1 Research Design

The study consists of two parts, the first being an exploratory pre-study which generated empirical data that help outline the direction of our research (Blomkvist and Hallin 2014). During this phase the researchers acted as insiders, initiated unstructured conversations with different employees, in particular our supervisors. Access was also gained to organisation-specific documents, which facilitated a better understanding of the organisation. This contributed to our understanding of the complexity of the research area, helping us deciding on the scope of our thesis. Further, problems regarding the creation of new business models based on market shifts and utilisation of new technology emerged, and especially technology regarding data processing. In this regard, a perception of a generally negative attitude towards data utilisation within the firm was noted. This collided with the ambition of finding new business models, which helped to further decide on the focus of the study. Aforementioned guided us into the second part, the main study, in which our analysis and conclusions are generated.

The phenomena studied in this thesis are the shifts caused by rapid technological development within the satellite industry, leading to new opportunities and potential new business models. Due to its recent emergence, this area has been subjected to relatively little research. An exploratory approach was therefore considered appropriate for this thesis work (Blomkvist and Hallin 2014; Yin 1994). The studied phenomena occurs in a real-world context, thus the phenomena should not be isolated from it, which suggests a case study approach that emphasises the context in which the phenomena occurs (Eisenhardt and Graebner 2007). Yin (1994) argues that case studies are suitable for ’how’ and ’why’ research questions, also noting that case studies are useful when the focus of the research is on contemporary events and the researcher does not need to control behavioural events, which holds true for our study. In order for the research to move in a coherent direction, Yin (1994) highlights the importance of establishing the propositions of the case study. If the case study is exploratory, Yin (1994) instead suggests that the researcher should state a clear purpose of the study. Since our work is of an exploratory nature, we have avoided stating propositions that may bias our result, and instead specified the purpose of the study to help guide us forward.

The study is decided to be a single case study, focusing only on SSC and investigating the phenomena in the context of a single firm. Yin (1994) argues that a single case study is suitable when researchers are investigating an extreme example, gain unusual research access, or have an opportunity to investigate a particular phenomena under rare circumstances, something which Eisenhardt and Graebner (2007) acknowledge as well. The studied phenomena occurs within an emerging business area that has been subject to a limited amount of research. A single case study can therefore make a relevant research contribution. However, Eisenhardt and Graebner (2007) stress that in comparison to a multiple case study, a single case study cannot provide an equally strong base for theory building. They also claim that a multiple case study can generate a more vigorous theory and provide a broader exploration of the posed research questions. Yin (1994) argues that the same criticism applies to performing a single experiment, noting that single case studies, like single experiments, are generalisable to theoretical propositions. A single case study can therefore provide valuable contributions to research by providing the a means of

Capitalising on Big Data from Space

Capitalising on Big Data from Space

How Novel Data Utilisation Can Drive Business Model Innovation

MARIA BREMSTRÖM SUSANNE STIPIC

Capitalising on Big Data from Space

How Novel Data Utilisation Can Drive Business Model Innovation

Maria Bremstr¨ om Susanne Stipic

Kapitalisera p˚ a stora datam¨ angder fr˚ an rymden

Hur nya s¨ att att utnyttja data leder till innovation av aff¨ arsmodeller

Maria Bremstr¨ om Susanne Stipic

Capitalising on Big Data from Space How Novel Data Utilisation Can Drive Business

Model Innovation

Maria Bremström Susanne Stipic

2019-06-03

Matti Kaulio

Ebba Laurin

Swedish Space Corporation

Tobias Roos

incumbent business model, the firm must simultaneously explore new business opportunities to avoid extinction.

Business Model Innovation, Data-Driven Business Model Innovation, Organisational

Ambidexterity, Satellite Data, Big Data

Examensarbete TRITA-ITM-EX 2019:380

Kapitalisera på stora datamängder från rymden

Hur nya sätt att utnyttja data leder till innovation av affärsmodeller

Maria Bremström Susanne Stipic

2019-06-03

Matti Kaulio

Ebba Laurin

Swedish Space Corporation

Tobias Roos

innovationsinsatser och leder till begränsade förutsättningar för att hantera organisatorisk ambidexteritet. Slutligen finner studien att storskaliga innovationsprojekt kan förbättra förutsättningarna för organisatorisk ambidexteritet.

Affärsmodellsutveckling, innovation, datadriven affärsutveckling, organisatorisk

ambidexteritet, satellitdata, big data

Table of Contents

List of Figures

List of Tables

Acknowledgements

1 Introduction

1.1 Background

1.2 Problematisation and Scientific Contribution

1.3 Purpose

1.4 Research Questions

1.5 Delimitations

2 Theoretical Framework

2.1 Business Models

2.1.1 Business Model Frameworks

2.2 Business Model Innovation

2.3 Organisational Ambidexterity

2.4 Data Driven Business Models

2.5 Data from Satellites

2.5.1 Big Data as a Concept

2.5.2 Satellite Data Processing

3 Conceptual Framework

3.1 Establishing the Business Model Elements

3.1.1 Innovation of a Business Model

3.2 Dynamics of Data Utilisation and Ambidexterity

4 Method

4.1 Research Design