On Stranger Tides -
An exploratory study on how digital start-ups navigate through business model adaptation in volatile
environments
Master’s Thesis 30 credits
Department of Business Studies Uppsala University
Spring Semester of 2020
Date of Submission: 2020-06-03
Jasmin Elmi
Sophie Sensenhauer
Supervisor: Henrik Dellestrand
We are one of these digital developments, (...) and we kind of drive this area in itself more than anything else. So, we don't implement
digital development, we are digital development.
- One respondent from our interviews (2020)
ACKNOWLEDGMENT
After almost five intensive months of writing this thesis, we have finally reached the finish line and we are extremely excited and proud of what we have accomplished. It has been a highly emotional, exciting, and educational journey. Throughout this semester we have faced obstacles and difficulties which have been challenging at times but it has also made us grow and find new successful ways of working as a team through an era of social distancing. It has also made us realize the value of turning to people around us and asking for advice. We would like to acknowledge everyone involved in our thesis process and express our deepest gratitude towards the ones that helped us along the way. Thank you for all your great contributions, the result of this thesis would not have been possible without you.
First and foremost we would like to show our utmost appreciation to our supervisor, Henrik Dellestrand. He has always provided us with valuable feedback that has helped us reach a higher quality in our writing. Thank you for your constant availability, support, and invaluable guidance throughout our circuitous journey. Secondly, we would like to thank all participating respondents from our case firms. Thank you, for dedicating your time and for contributing with your in-depth answers and insights through our interviews. We hope this thesis will be of value to you, too!
We would also like to thank our fellow peers for further advice and constructive feedback on our thesis through each seminar, and all our friends and family for the continued support.
____________________ ____________________
Jasmin Elmi Sophie Sensenhauer
ABSTRACT
In recent years, new technologies have led to disruption and change within most industries, resulting in the emergence of digitally-born start-ups. The purpose of this thesis is to complement the scarce theory about what the drivers of business model adaptation in start-ups consist of and how the adaptation process is managed to stay competitive. The findings not only provide theoretical contributions, but can also help managers to steer their start-ups through the growth phase.
The research is set within the realms of business model theory in the strategic management field. An exploratory study with a qualitative, inductive approach was chosen to gain insights into the business model dynamics of nine firms from the fintech and healthtech industries.
The results showed that foreign market adaptation, industry dynamics, funding, and legislation are perceived challenges, whereas legislation and exogenous shocks are opportunities that drive business model adaptation in start-ups. The business model elements of strategic decision- making, resources and capabilities, and network and partners were found to be integral to the adaptation process, as their core components not only need to be adapted to the environment but also aligned with each other. Those components should be revised in an iterative trial-and- error process driven by feedback.
Keywords: Business model, Business model adaptation, Digital start-ups, Fintech, Healthtech, Challenges, Opportunities, Strategy, Resources and capabilities, Network and partners.
Word count: 17222
TABLE OF CONTENT
1 INTRODUCTION... 1
2 BUSINESS MODEL ADAPTATION IN VOLATILE ENVIRONMENTS... 5
2.1 Business models ...5
2.2 Business model adaptation ...6
2.2.1 Strategy ...9
2.2.2 Resources and capabilities ...10
2.2.3 Network and partners ...11
2.3 Research framework ...12
3 METHODOLOGY ... 14
3.1 Research method and design ...14
3.2 Setting and selection ...15
3.2.1 Industry setting ...15
3.2.2 Sample criteria and case firms ...16
3.2.3 Data collection and selection of respondents ...18
3.2.4 Operationalization and interview design ...20
3.3 Data analysis and coding process ...22
4 EMPIRICAL FINDINGS ... 27
4.1 Challenges ...27
4.2 Opportunities ...30
4.3 Strategic decision-making ...32
4.4 Resources and capabilities...34
4.5 Network and partners ...37
4.6 Summary of cross-case empirical findings ...39
5 DISCUSSION ... 42
5.1 Challenges ...42
5.2 Opportunities ...43
5.3 Strategic decision-making ...44
5.4 Resources and capabilities...46
5.5 Network and partners ...47
6 CONCLUDING REMARKS ... 49
6.1 Practical and theoretical contribution ...49
6.2 Limitation and future research ...50
REFERENCES ... 52
APPENDICES ... 60
Appendix 1 - Email invitation to interview participants ...60
Appendix 2 - Operationalization and interview guide ...61
Appendix 3 - Follow-up questions ...65
Appendix 4 - Empirical data for challenges ...67
Appendix 5 - Empirical data for opportunities ...68
Appendix 6 - Empirical data for strategic decision-making...69
Appendix 7 - Empirical data for resources and capabilities ...70
Appendix 8 - Empirical data for network and partners ...71
Overview of figures and tables
Figure 1: Strategic components of the business model, Wirtz et al. (2016) ... 9Figure 2: Theoretical model ... 13
Figure 3: Data structure of concepts, themes, and aggregate dimensions in line with Gioia et al. (2012). ... 25
Table 1. Summary of case firm sample criteria ... 17
Table 2. Case firm start-ups from both the financial and healthcare industry ... 17
Table 3. Respondent background and interview information ... 19
Table 4. Pattern matrix inspired by Santos & Eisenhardt (2009) and Eisenhardt & Graebner (2007) ... 40
Table 5. Pattern matrix II inspired by Santos & Eisenhardt (2009) and Eisenhardt & Graebner (2007) ... 40
1 INTRODUCTION
In the context of digital transformation, innovative and disruptive companies emerge, seeking to scale and grow quickly. In the new ‘winner-takes-all economy’ (Brown et al., 2018), these start-ups are under great pressure to adapt in order to be competitive. One strategy of said adaptation is “actively (aligning) the firm’s business model to a changing environment” (Saebi et. al., 2016, p. 3). Therefore, this thesis seeks to answer two questions: Firstly, what do challenges and opportunities driving digital start-ups to business model adaptation consist of, and secondly, how do digital start-ups manage the business model adaptation process to stay competitive?
Digital transformation has been a catalyst for driving efficiency through technology in almost every industry, with firms employing new digitalized business models to increase value creation (Verhoef et al., 2019). ‘Product-as-a-service’, data-driven business models, and platform-based ecosystems are just three of the most commonly adopted ones (Fitzgerald et al., 2013; Westerman et al., 2011). Additionally, owing to the nature of change brought about by digital transformation, new businesses have been entering industries that have formerly been protected by entry barriers, competition, and supplier power (Van Alstyne et al., 2016). As a result, digitally-born start-ups have been challenging long-established firms with their ability to efficiently fulfill emerging customer needs (Verhoef et al., 2019; Rogers, 2016). One of the leading industries in digitalization is the knowledge-intensive finance sector, which is characterized by highly-digitized assets, transactions, and labor. Another industry subject to disruptive transformation is healthcare (Forbes.com, 2019). However, compared to their counterparts in finance, healthcare companies generally struggle to employ digitalization to increase innovation and profits (Harvard Business Review, 2016). Both these industries have typically been traditional and stagnant, with a few large players dominating. However, digital transformation enabled innovative ventures to enter the industries and disrupt the industry structure with new types of business models (Li et al., 2017; Gomber et al., 2018).
Digitally-born start-ups claim increasing space in today’s business landscape. They tend to utilize digitalization to create innovation in products, services, and business models (Teece, 2010; Wirtz et al., 2016). Merriam-Webster defines a start-up as “a fledgling business
research. Common characteristics that are used to define start-ups include operating with a high risk of failure, handling great uncertainty and complexity, lack of experience, high flexibility and dynamics (Zaech & Baldegger, 2017) and resource constraints (Chan et al., 2018). In established companies, managers often face resistance to change the more mature the firm gets, while start-ups usually excel because those barriers are considerably lower (Zaech &
Baldegger, 2017).A strong ability to grow is also a factor often counted in to distinguish start- ups from any other kind of company. Start-ups do not need to be tech-oriented to be classified as such, however, it is common that either product/service or business model are derived from the latest technological innovations (Robehmed, 2013).
Digital transformation and business models are closely intertwined. According to Brown et al.
(2018), new customer needs have been contributing to the digital transformation of industries and the subsequent revolution of business models to fulfill those needs. Business models describe how a firm creates, delivers, and captures value (Teece, 2010). Balboni et al. (2014) define the business model as a significant driver of growth, and therefore it is an important topic to study within the strategic management realm. Konya-Baumbach et al. (2019) argue that business model considerations are a crucial building block for start-up growth. Just like the macro-environment constantly changes, business models should also be actively adapted to maintain their suitability with a volatile environment (Balboni et al., 2014). Digital transformation changed seemingly-fixed market rules, and according to a McKinsey survey, only 4-26% of companies succeed to create a competitive advantage in a digital context (De la Boutetière et al., 2018). This statistic also includes start-up companies, failing early on in their venture due to a lack of appropriate strategies and business models. Therefore, in the context of digital transformation, constant adaptation of the business model is crucial to stay competitive.
Digital transformation, therefore, has created volatile market conditions, allowing start-ups to enter formerly well-protected industries. Wirtz et al. (2016) and Verhoef et al. (2019) found that to handle constantly changing environments, companies should adapt their business models as business models have a positive effect on a firm’s competitiveness and serve as a source of competitive advantage. Start-ups are under especially high pressure to grow and scale quickly due to i.a. resource constraints. In order to do so, they should engage in a continuous revision of their business model. Start-ups are forerunners when it comes to adopting new business models to bring innovative products or services to the market. The first years in a
start-up life cycle are characterized by strong growth to survive and become competitive. In order to grow and become a scale-up and eventually an established organization, adaptation to the environment is key (Saebi et al., 2016). Therefore, business model adaptation is a key factor to facilitate a start-up’s growth phase.
Balboni et al. (2014) note that growth is dependent on market dynamics and the industry structure. This suggests that the vast majority of start-ups fail with their quest to grow amongst others due to inadequate adaptation of their business models or strategies. However, actively managing adaptation is critical, because in contrast to other small firms, the timeframe from founding to becoming a major player in the market is significantly shorter for start-ups.
Business model adaptation remains a challenging undertaking for start-ups in general, but to narrow the scope of the study this thesis looks at how start-ups have successfully managed business model adaptation.
From a theoretical point of view, although extensive research can be found on start-ups, business models, and drivers of business model change, these concepts have rarely been related to each other. Business model adaptation specifically has mainly been researched within incumbent and digitalizing firms, mostly conducted in the quantitative research domain (Saebi et al., 2016; Chakravarthy, 1982; Cavalcante, 2012). Constantly changing is mandatory to survive in a volatile environment, but start-ups are under higher pressure than incumbents to do so. A great contribution to organizational adaptation of technology-based new ventures has been put forward by Andries and Debackere (2006), whose insights regarding the adaptation process to product and process innovation also informed this thesis. Whereas Andries and Debackere (2006) focused solely on dynamic capabilities and resources, De Reuver et al.
(2009) researched the drivers behind business model adaptation in tech start-ups. This thesis expands on their knowledge with process-related business model research to close a gap noted by Balboni et al. (2014): “The available literature on the growth processes of new ventures operating in the high-tech and science-based sectors is still very limited” (p.2)
As a result, there is a need to further research the process of business model adaptation in the context of digital start-ups. Hence, we pose the two following research questions:
• What do challenges and opportunities driving digital start-ups to business model adaptation consist of?
• How do digital start-ups manage the business model adaptation process to stay competitive?
The purpose of this thesis is to challenge existing business model adaptation theory and contribute to a better understanding of start-ups by identifying what the drivers of business model adaptation consist of and how the business model adaptation process is managed by digital start-ups in order to stay competitive. The thesis will add to the existing strategy-, business development- and organizational research as business model adaptation in start-ups has been less researched than in incumbents. New companies serve as a worthwhile research subject for organizational theories and industry-specific research (Bamford et al., 1999).
The study will also contribute to the understanding of managers, who try to steer their new ventures through vastly changing environments. Wirtz et al. (2016) found that financially successful firms are more involved in business model activities than less financially successful firms, so there is likely to be a direct influence of business model management on profitability.
Furthermore, in terms of a country’s economic development, highly-digital and science-based industries play an important role. However, their growth process and development are poorly understood by policymakers and researchers. The literature on such firms is still somewhat limited (Balboni et al., 2014). As the subjects of this study are digital start-ups that have survived the first critical years, we aim to show that effective business model adaptation leads to competitive fintech and healthtech-based ventures. The era of digital transformation requires companies to adapt their business model, and constantly formulate new hypotheses about their value proposition to stay competitive, so this is a topic of high interest and relevance to investigate further.
2 BUSINESS MODEL ADAPTATION IN VOLATILE ENVIRONMENTS
2.1 Business models
This thesis is largely grounded upon the business model definition provided by Teece (2010).
A business model is a holistic overview of an organization's activities, evolving around a firm’s value proposition and providing understanding about how firms create, deliver and capture value from their activities (Teece, 2010; Osterwalder et al., 2010). So when referring to a business model, we mean which customer problem or challenge an organization tries to solve, how it can make money from it and how it can do so in the long term.
Business model theory emerged in the 1950s but was not established as a separate research stream until the dotcom boom in the 90s (Teece, 2010). Business models have been especially relevant for digitally-oriented firms who must often find alternative methods to create competitive advantage and new mechanisms to capture value (Konya-Baumbach et al., 2019).
Koch and Windsperger (2017) and Christensen (2001) argue that business models can be the main building block of competitive advantage and facilitate superior firm performance, especially when they are novel and difficult to imitate. Also, business models are connected to elevated competitive advantage. As a result, business model activities should be considered for venture development and -growth (Wirtz et al., 2016, Balboni et al., 2014). Also, a new business model can become an innovation. Such innovation stemming from digital transformation has resulted e.g. in the creation of platforms and business ecosystems. Those are often associated with new ventures’ exceptional success and strong ability to grow (Teece, 2010; Wirtz et al., 2016), while being more complex than traditional business models, since an ecosystem includes all actors that contribute to a firm’s value creation (Koch & Windsperger, 2017).
Since a business model addresses basic existential and strategic questions, any firm should analyze its components carefully against internal and external characteristics to create a sound basis for the subsequent business strategy. Teece (2010) emphasizes that just like a firm’s environment changes, a business model should be dynamic and subject to development (Saebi
et al., 2016). This notion will be discussed further after the description of antecedents to business model adaptation.
2.2 Business model adaptation
Theory suggests that business models should be adapted as a response to changes (Balboni et al., 2014; Demil & Lecocq, 2010). Effective business model adaptation could mitigate the risk of failure for start-ups (Konya-Baumbach et al., 2019), so it is a suggested method to build and sustain competitiveness and address change over time. The management’s goal should be to continuously sustain the fit between environment and organization. Therefore, we view the business model through a dynamic lense.
Before a start-up launches its product to the market, it may struggle with defining the initial business model, because information is scarce and only becomes available with experience (Andries & Debackere, 2006). Technology-based companies tend to face especially high levels of uncertainty due to the digital transformation era. Therefore, digital start-ups are likely to be subject to progressive adaptations during transitioning from an idea to a stable firm (Demil &
Lecocq, 2010). The first two company phases start-ups go through are research and product development and commercialization, with the latter being characterized by strong growth. In these highly uncertain and ambiguous periods, their business models go through major developments, with all of their elements being affected (Andries & Debackere, 2006).
Researchers found that those changes are driven by shifts in the firm’s environment.
Since start-ups have rarely been the subject in business model adaptation research, factors that drive business model adaptation were mostly studied in relation to incumbents. Those factors may not be applicable since firm characteristics of incumbents significantly differ from start-ups, but the current literature provides a starting point. Inductively studying those drivers in the scope of the chosen industries will help to set the context in which the business model adaptation of the case firms takes place.
Business model adaptation can be driven by both external and internal influences (Demil &
Lecocq, 2010), such as threats from the environment, and stakeholders who are part of the
firm’s value network. Fonseca and Domingues (2017), Tankhiwale (2009), and Balboni et al.
(2014) also found pressures from legislation and market dynamics as driving forces. The former is an influencing factor for instance when legislators strive to create supportive start-up environments. Supportive regulation is introduced by countries that aim to create an innovative economy through fostering entrepreneurship, supporting networks, and venture initiatives (Balboni et al., 2014). The PESTLE-analysis also includes an assessment of legal factors from the external environment as those influence organizational performance (Fonseca &
Domingues, 2017).
New information and technologies can also become opportunities driving business model adaptation. New information is considered a driver because initial business models are grounded upon incomplete information, and more information becomes available during the later start-up phases (Andries & Debackere, 2006). Next to that, De Reuver et al. (2013) and Björkdahl (2009) state that technological changes, e.g. launching a new technology, and innovation initiate adaption. Balboni et al. (2014) found that business models are enablers of transforming technological development into growth for start-ups. Therefore, it can be assumed that digital business models are especially prone to changes. However, De Reuver et al. (2009) found that legislation factors are even more influential than new technology and market dynamics. Next to new technology, internationalization has been a source of market change in the last decades and represents major opportunities for incumbents and start-ups alike. The latter is likely to seek foreign market entry a.o. because of high competition in the home market or to exploit first-mover advantage.
Though increasing sales or efficient acquisition of resources may be the result of internationalization, going abroad also adds great complexity to a firm’s strategic and operational activities (Grant, 2016). With regard to digital start-ups, it would be interesting to look at the effect of internationalization on the business model. Since start-ups are defined by resource constraints, they are likely to tap into foreign markets through transaction market entry. According to the findings of Johanson and Vahlne (2009), the psychic distance between two markets increases with geographical distance, especially in banking (Grant, 2016). So the internationalization process is likely to be stepwise, meaning unlocking closer markets first through relationships with established players. De Reuver et al. (2013) state that the decision to launch the core service of the firm precedes the internationalization process, while also
and managers should analyze how the internationalization process and greater target group influences the other parts of the business model (De Reuver et al., 2013). Next to external challenges and opportunities, internal factors such as top management’s choices and dynamics within and between core elements of the business model (Demil & Lecocq, 2010; see 2.3) drive business model adaptation. Also, fast growth as a start-up characteristic likely affects business model change (Wirtz et al., 2016; Zott et al., 2011). Due to this condition, which is related to the scarcity of resources, new ventures’ growth process is influenced by strategic factors, such as making active decisions as to which objectives and opportunities to focus their resources on.
Andries and Debackere (2006) found that, especially during the starting phase, organizations alternate between opportunity detection and resource mobilization, adjusting their business model in each cycle. Saebi et al. (2016) argue that firms are more likely to adapt their business model when facing external perceived threats than opportunities since opportunities are closer connected to keeping the status quo of a business model.
Business model adaptations are changes in the business model. Especially in the first years, business models need to be adapted to the constantly changing environment, otherwise companies are likely to fail. It may be seen as both a revision process of the basic assumptions a start-up is founded on and a continuous fine-tuning process of operations (Saebi et al., 2016;
Demil & Lecocq, 2010). The adaptation can affect all parts of a business model: value proposition, target market, value chain, financial structure, and the architecture that links those elements. In the following, basic assumptions regarding business model adaptation will be presented.
Firstly, there is no standard adaptation process or outcome that is applicable to all firms.
However, Teece (2010) noted that once a change in one area is identified, there should be adaptations in all other areas, too. The failure to adapt in a timely manner could negatively impact profitability due to the business model’s bad fit with the environment (Saebi et al., 2016). Secondly, as to the process of business model adaptation itself, there are two views. On the one hand, researchers suggest that extensive analysis precedes effective change. On the other hand, some argue that the process is defined as ‘feedback loops’ (Casadesus-Masanell &
Ricart, 2010), trial-and-error cycles (Achtenhagen et al., 2013; Sosna et al., 2010) or
‘experimentation-and-learning’ periods (Cavalcante, 2012; McGrath, 2010; Balboni et al., 2014). It can be assumed that start-ups are mainly in the experimentation period because they must still explore the core logic on which their firm should be grounded (Sosna et al., 2010).
Demil and Lecocq (2010) state that exploring other business models or novel directions is superseded by testing those findings from the exploration period, and vice versa. A more reactive presumption describes that organizations alternate between opportunity detection and resource mobilization, adjusting their business model in each cycle, especially in the initial phase (Andries & Debackere, 2006). Progressive business model adaptation is thus characterized by changes in internal or external variables and their consequences. Instead of punctuated, radical change, companies in volatile environments should actively engage in those virtual loops (Casadesus-Masanell & Ricart, 2010). Therefore, continuous adjustments through experimentation are the proposed definition of the business model adaptation process.
In their article about business model origins and developments, Wirtz et al. (2016) provide recommendations regarding the study of business model adaptation. Strategy, resources and capabilities, as well as network and partners form the strategic part of a business model (see figure 1 below). The business model adaptation process, an interplay between environmental influences and growth aspirations of new ventures, can thus be examined through changes within and across those components to see how digital start-ups manage challenges and opportunities in the market. Demil & Lecocq (2010) emphasize that it is these structural changes that are the symptoms of business model adaptation.
Figure 1: Strategic components of the business model, Wirtz et al. (2016)
2.2.1 Strategy
Firstly, it is assumed that start-up growth is mainly due to strategy. A facilitating business strategy is needed to promote business model change and stay competitive (Andersson &
Eriksson, 2018; Fonseca & Domingues, 2017). According to Balboni et al. (2014), if one new
venture grows faster than another it can ultimately be traced back to strategic factors because the new venture showed how well it can respond to changes in the environment. There is no one strategy that trumps the others in terms of responsiveness and adaptability, but a strategy is superior when it is aligned and coherent with a firm’s organizational capabilities, functional strategies, and internal and external environment (Fonseca & Domingues, 2017). Wirtz et al.
(2016) describe the organization’s core strategy as an integral part of a business model and although not much is known about business model dynamics, researchers tend to agree on the primary role strategy is playing (Teece, 2010; Achtenhagen et al., 2013). The process of business model adaptation involves the development of a strategy with which a firm can respond effectively to the environment (Chakravarthy, 1982; Cockburn et al., 2000), hence such a strategy exists in close connection to those continuous learning processes which are described under 2.3. New ventures experiment with new strategies and organizational structures, which is manifested by learning processes facilitating adaptation (Balboni et al., 2014). Another integral relationship is often drawn between strategy and resources and capabilities when studied in relation to business strategy and profitability. This is in line with the resource-based view that assumes that strategy is built on the critical resources and capabilities that lead to success and competitive advantage (Grant, 1991, Sandberg & Hofer, 1987). All in all, business strategies are strongly impacting business model development (Cockburn et al., 2000), so they should be included in the study about start-ups.
2.2.2 Resources and capabilities
The Resource model is the second main strategic component of business models and includes a firm’s core resources and capabilities (Wirtz et al., 2016). Fonseca and Domingues (2017) identify the unique combination of resources, capabilities, and activities as the components of a firm’s value proposition.
For one, resources influence the extent of strategic possibilities managers have because resource scarcity limits the scope for experimentation, which is an important part of the growth phase of the firm (Cavalcante, 2012; Andries & Debackere, 2006; Chakravarthy, 1982). Start- ups tend to operate under resource constraints (Forbes.com, retrieved 2020) while being pressured to attain unique resources through organizational learning to achieve a profit for the firm (Andries & Debackere, 2006). Wirtz et al. (2016) and Osterwalder et al. (2010) indicate that resources and strategy are frequently studied in connection to business model
development; Andries & Debackere (2006) verified a direct relationship between resources and business model adaptation. Resources can be of human, technological, financial, or network nature, and firms need to include considerations such as when to hire new employees, which skills need to be developed, and when to place new technologies in their adaptation process (De Reuver et al., 2013).
Capabilities originate from resources and are the main sources of competitive advantage and profitability. They determine the efficiency with which firms can transform resources into products or services (Grant, 1991; Collis, 1994). When environments change, the ability to develop new capabilities increases competitiveness (Barney, 2001). Coequally, since we view the business model through a dynamic lens and assume that resources need to constantly be changed and reconfigured to preserve their fit with the environment, the corresponding capabilities are needed to do so. Teece et al. (1997) call these ‘dynamic capabilities’. Successful start-ups are likely to develop and display them in ambiguous and volatile environments. In the business development field of study, researchers usually look at initial and acquired resources and capabilities to see how they have been developed over time and augmented or changed the value proposition (Demil & Lecocq, 2010). An integral part of this transformation process is choices because they determine new combinations of resources and consequently the impact of those on the firm’s product or service. Dynamic capabilities, in turn, are embedded in routines, determining the efficiency with which organizations create value through resources.
As a result, routines affect how a firm can change and adapt over time (Teece & Pisano, 1994).
Grant (1991) states that new ventures learn new routines easier than large incumbents, especially in markets undergoing technological change.
2.2.3 Network and partners
It is not sufficient to base strategic management research solely on traditional assumptions such as Porter’s five forces or the Resource-based view, because of changes in industries, products, and customer demands (Koch & Windsperger, 2017). Some researchers, therefore, propose adopting a network-centric view when assessing the interactions of a firm with its environment.
Due to industry barriers breaking away and the classic supplier-customer relationship changing, firms should be researched within the context of their business ecosystem, intersecting multiple industries. The defining feature of this network view is that value is not added along a sequential supply chain, but rather co-created with the partners and even
competitors in the network (Bowman & Ambrosini, 2000; Koch & Windsperger, 2017). The worth of the ecosystems is amplified by the companies in it and the partnerships contributing to a firm’s value proposition. This network effect arises because the network actors contribute and access their separated resources and capabilities, which is also due to the technological advancements of digital transformation (Yoo et al., 2012, Balboni et al., 2014). Within the lean start-up theory, the contribution of feedback from interaction with the external environment to start-up success is highlighted (Felin et al., 2019). Lean start-up theory describes entering the market with a basic version of a new product or service and then, step-by-step through feedback, improving the product. The process can be thought of as experimentation, as a hypothesis is created and then directly tested through iteration and interaction. Through those
‘feedback loops’ (Casadesus-Masanell & Ricart, 2010) start-ups can better handle resource constraints and go faster to market with a minimal viable product. However, not only feedback from customers suggesting unfulfilled needs or using the product should be incorporated in the process, because it could be misleading, but also listening to suppliers and other actors from the value chain. As a result, new value propositions can be tested through those so-called ‘social proofs’, which could emerge in the form of new partners that are convinced that the future product will create value for the firm. Therefore, feedback is an important concept and contributor to start-up growth (Felin et al., 2019).
2.3 Research framework
Since business models are an integral component of achieving growth and competitiveness, they are also assumed to be crucial for start-ups. The literature review showed that the strategic core elements of a business model, namely strategy, resources and capabilities, as well as network and partners play a central role in the adaptation process. Start-ups need to engage in effective business model adaptation due to their unique situation defined by resource constraints, intense competition and pressure to grow. Despite this, start-ups have seldom been subject to business model research. In this thesis, we focus on two areas: firstly, the content of challenges and opportunities leading to business model adaptation; secondly, how new ventures manage the adaptation process in order to stay competitive in the growth phase. We frame the process as a logical sequence: In continuous loops defined by experimentation, or trial-and-error, start-ups face challenges and opportunities that cause them to adapt their
business model t0. Integrating the findings from the literature review, the following model has been developed to illustrate the process of business model adaptation in start-ups. The two focus areas are designated with question marks:
Figure 2: Theoretical model
The initial version of a business model is represented with the box labeled ‘t0’. The box labeled
‘t1’ shows an adapted business model after challenges and opportunities have influenced a start-up. The arrow on the bottom symbolizes a move from t0 to t1 with a continuous business model adaptation process.
Little is known about the setting and process of start-up business model adaptation, so the separate concepts are intentionally kept broad to foster explorative research. Thereby, a greater understanding of the topic at hand can be achieved through a qualitative research method.
3 METHODOLOGY
3.1 Research method and design
The aim of this study is to gain a deeper understanding of what challenges and opportunities drive digital fintech and healthtech start-ups to adapt their business model, and how the business model adaptation process is managed within smaller and more agile organizations, in order for them to keep their competitive edge. The idea of researching drivers of business model adaptation (challenges and opportunities) and how start-ups manage the business model adaptation process to stay competitive and survive, originated from identifying a lack of previous research on business model adaptation in the context of digital start-ups. In order to obtain a greater contextual understanding and get deeper insights into the main drivers of adaptation and how these are managed, a qualitative research method was chosen (Azungah, 2018; Hart & Dewsnap, 2000; Miles & Huberman, 1994).
As the general argumentation in previous literature indicates that the management dimension of business model adaptation has been highly understudied (Sabei et al, 2016) and particularly considering that it has been unexplored in the context of digital start-ups, this study takes on an exploratory design with an inductive approach. The research questions originated from an interest in the phenomenon rather than building on a particular theory (Woiceshyn &
Daellenbach, 2017). However, Johnston (2014) emphasizes that even if the inductive approach tends to build theory from the data collected, an existing theory is still important to conduct a rigorous study. In inductive research approaches, a theory is the outcome of research and the raw data, engendering observations, themes, and concepts. Thus, this approach was mostly applicable for the study at hand since it was only after the data had been collected that the connection between empirics and theory was fully understood and different concepts were clearly identified (Azungah, 2018; Bryman & Bell, 2011).
Due to the existing research gap of the modern phenomenon of digitally-born start-ups in relation to business model adaptation and drivers of adaptation, a case study was needed in order to grasp this broad topic. Thus, this thesis adopted Yin’s (2009) definition of a case study as empirical research, investigating a modern phenomenon within real-life events where the separation between context and the event is unclear. In order to reach the case study goal of
gaining a broader understanding as well as expanding and analytically generalizing the theory, a multiple case study with 9 cases was conducted. Conducting a multiple case study using between 4 to 10 cases also augments external validity and contributes to a better foundation for analytic generalization (Yin, 2009; Eisenhardt, 1989). The meaning of ‘case’ itself differs from Yin’s general definition of a case study and since it can take on the shape of either individuals, organizational forms, or entities, it requires to be defined. Hence, the unit of analysis within each different case will be the start-up as an organizational form (Andries & Debackere, 2006;
Yin, 2009).
3.2 Setting and selection
3.2.1 Industry setting
As explained above, in recent years digital transformation has reshaped many industries. New technologies have led to disruption and changes within most markets, resulting in digitally- born start-ups to enter (Kupp et al., 2017; Van Alstyne et al., 2016). Two industries that have been highly affected by digital transformation are the industry of financial services as well as the healthcare sector. Research from 2019 conducted by the Forbes Technology Council found that almost every industry is impacted by disruptive technology, with the financial and health industry being particularly influenced. Both industries will also continue to rapidly be disrupted in the upcoming year, resulting in a higher degree of active fintech (financial technology) and healthtech (healthcare technology) start-ups (Forbes.com, 2019). Fintech has broadly been described as highly innovative financial services and its coexisting technology-enabled business models. The umbrella concept, therefore, includes any technological innovation in the scope of payments, insurance, regulations, and banking (Browne, 2017; Mention, 2019). In regards to what is defined as a healthtech start-up, in the context of this study, the definition by the World Health Organization is used. It describes healthtech as the application of technology together with knowledge and skills that enhance the quality of life and solve everyday health problems. This includes but is not limited to, digital platforms, and mobile applications about medicine, mental health, wellness, and hospital care (World Health Organization, 2020).
Given the time and scope of the study and to be able to generalize on the result, an industry
limitation was necessary. Additionally, due to the financial and health sector being highly disrupted by many innovative digital start-ups (Neuburger, 2018), the chosen industry setting is conducive to answer the research questions and makes up a good laboratory for exploring challenges and opportunities related to business model adaptation. Investigating case firms from two different industries will induce both a within-cases analysis but also cross-cases analysis to see if the results differ between industries, which enhances generalizability and minimizes errors of external validity (Miles & Huberman, 1994; Yin, 2009).
3.2.2 Sample criteria and case firms
In addition to the choice of selecting case firms from two different industries, a set of criteria was constructed to carefully select relevant case subjects applicable to our research purpose.
This was necessary in order to provide a clear replication logic which enhances external validity (Yin, 2009). Due to the design of the study being outlined as exploratory, an indiscriminate approach was taken when choosing case firms. However, the cases were not randomly selected. Rather, a set of specific sample criteria was followed which aimed towards case firms’ relevancy for fulfilling the research questions and purpose of the study whilst still targeting a broader range of organizations, a commonly used strategy known as theoretical sampling (Eisenhardt, 1989). Cases were sampled in a way that the start-ups were mature enough to have gone through an adaptation or change in their business model. Hence, the criteria important to fit the purpose were the age of the firm and the need to be in the commercialization phase, meaning the stage when products and services are past the prototyping period and becoming more standardized (Andries & Debackere, 2006).
Additionally, they had to have gone through some kind of business model change and the corporate purpose must have a clear technological focus, i.e the product, service or business model were born out of digital transformation and therefore are digital (Robehmed, 2013;
Teece, 2010; Wirtz et al., 2016). Due to the time and scope of the study, only Swedish founded start-ups with their headquarter in Sweden were chosen for the study.
Table 1. Summary of case firm sample criteria Sample Criteria
At least 4 years old
Currently or recently in the “Commercialization Phase”
Have gone through some kind of business model change A digital start-up with a technological focus
Founded with their HQ in Sweden
The initial contact was made through email, but due to rejections from many firms to take part in the study, a trial-and-error approach was initiated and a selective search for case firms was conducted through reaching out to our personal network on LinkedIn. Additionally, conducting deeper industry research helped to narrow the search down to 9 case firms that were in line with the criteria. In the table below, key figures and background information are presented in order to give a better understanding of the chosen case firms. One of the case firms (number 1) stands out a bit from the other firms with both a greater number of employees and higher revenue. This case firm was considered to be relevant to the study due to the firm seeing itself as a start-up in the growth phase, similar to the other case firms. Since business models is seen to be a sensitive topic, all case firms were assured anonymity and therefore identified by a number instead of the official name of the start-up.
Table 2. Case firm start-ups from both the financial and healthcare industry Case Firm Industry Founded Employees Revenue
2019
Product/
Service
Case Firm 1 Fintech 2013 1500 Employees 430 Million
SEK
Digital neo bank
Case Firm 2 Healthtech 2014 9 Employees 3,5 Million
SEK
Digital application for diagnosis of cancer
Case Firm 3 Fintech 2015 8 Employees 10 Million SEK
Digital lending platform
Case Firm 4 Healthtech 2012 6 Employees 1 Million
SEK
Digital platform to treat sleeping disorder
Case Firm 5 Fintech 2012 40 Employees 13,1 Million
SEK
Digital neo bank
Case Firm 6 Fintech 2015 28 Employees 51,6 Million
SEK
Digital micro-lending
Case Firm 7 Fintech 2013 35 Employees 19,7 Million
SEK
Digital payment service
Case Firm 8 Healthtech 2014 5 Employees 1 Million
SEK
Digital care platform
Case Firm 9 Healthtech 2014 6 Employees 850 000
SEK
Digital platform for cognitive behavioral therapy
Sources are interviews and secondary data provided by the respondents in the form of annual reports, press releases, pitch-decks, and website information.
3.2.3 Data collection and selection of respondents
The empirical data was collected through both primary data and secondary data with the main focus on primary data to gain a better understanding of the content of challenges and opportunities and how the business model adaptation process is managed. This was done in order to tailor the way data was being collected and how the theory was operationalized to fit the two research questions. Therefore, the data collected was relevant for the research topic and would be useful in answering the questions and purpose of the study, which, according to Hox and Boeije (2005), is the most crucial benefit of using primary data. The primary data was collected through semi-structured focused interviews with one respondent from the top management of each start-up, accounting for a total of 9 interviews. In the context of this study, top management was defined in accordance with the most common definition, described as all
inside top-level executives involved in making different strategic decisions affecting the firm.
This includes the founder, co-founder, Director, CEO, CGO, CCO, CMO, and head of any business unit (Carpenter et al., 2004). The top management was chosen as the sampling criteria of interview subjects since they tend to be the ones creating the company strategy, making decisions on the adaptation of the firm's business model as well as making general decisions affecting how the firm does business and what the future of the company will look like (Andersson et al., 2007).
Considering the design of this study, semi-structured focused interviews were considered the best suitable choice of primary data collection. This is mainly based on this source of evidence being more open-ended, less structured, and flexible, granting the respondent to freely express their thoughts whilst still allowing for some type of structure by following a pre-written interview guide with specific themes (Azungah, 2018; Yin, 2009). Semi-structured focused interviews are usually conducted during a shorter time, between 45-60 min and allow for follow-up or clarification questions in need of more information, creating an atmosphere of deeper two-way communication that provides valuable in-depth information. The advantage of these interviews, therefore, lies in the interplay between the respondent and the one interviewing as well as in the questions asked (Silverman, 2004).
Information regarding the initial video interview, date, and time, as well as background information on the interviewees’ role and years at the start-up, is presented in table 3 below.
To ensure each respondent’s anonymity and follow the European Commission’s guideline on General Data Protection Regulation, the full names of each interview subject were replaced with letters A-H (European Commission, 2018). The letters were later used in the empirics section as identification when presenting statements or quotes from a specific respondent.
Table 3. Respondent background and interview information
Respondent Start-up Role at the start-up Tenure length
Date of the interview
Length
Respondent A Case Firm 1 Head of Operations &
Strategy
2,5 years 2020-03-30 60 min
Respondent B Case Firm 2 Business Development Director
2 years 2020-03-31 40 min
Respondent C Case Firm 3 Co-founder & CEO 5 years 2020-04-03 60 min
Respondent D Case Firm 4 Co-founder & CEO 6 years 2020-04-07 50 min
Respondent E Case Firm 5 Co-founder & CEO 4 years 2020-04-14 45 min
Respondent F Case Firm 6 Co-founder & CGO 5 years 2020-04-16 45 min
Respondent G Case Firm 7 CEO 1 year 2020-04-17 60 min
Respondent H Case Firm 8 Co-founder & CCO 5,5 years 2020-04-20 60 min
Respondent I Case Firm 9 Co-founder & CEO 3 years 2020-04-20 50 min
To increase the construct validity of the study and achieve triangulation, secondary data from company pitch-decks, newspapers, and reports on the industry and topic were collected to complement the empirical data gained through interviews. Trusted secondary sources such as reports from Deloitte and McKinsey were chosen, as well as newspapers like Forbes and Dagens Industri. Using secondary data could be seen as a limitation in some cases, due to the fact that it has been collected in a different context and might therefore not reflect the exact purpose of the study (Ghauri & Grønhaug, 2010). However, we considered the use of multiple sources of evidence beneficial for validating the information given, as well as for conducting the cross-case analysis to better grasp and argue the real-life theoretical phenomena (Eisenhardt 1989; Yin, 2009).
3.2.4 Operationalization and interview design
To contact the interviewees, an email invitation was sent out to the selected respondents to ask for participation in the study (see Appendix 1). The interview invitation was clearly structured to give the potential respondent a brief understanding of the research purpose and what the study could contribute to, both on an industry level as well as for the organization (Ghauri &
Grønhaug, 2010).
In addition to the email invitation, the first draft of an interview guide with 5 general questions and 12 theme-specific questions was written up. The interview guide was designed to adopt different themes based on previous research presented in the theoretical section and in relation to the research questions. Accordingly, questions were divided into sections of Business model, Drivers of Business Model Adaptation, Processes, Resources & Capabilities, Strategy, Network & Partners, and Competitiveness. According to Ghauri and Grønhaug, (2010) certain steps need to be taken to make sure the interview guide was accurate before starting the interviews.
Firstly, the initial draft was checked by a third party and adjustments were made by adding one more question, “Could you describe how the business model looked like at t0?” in order to gain a better understanding of the history of the business model and how it has changed over the course of the company's lifespan. We also prepared one clarification question and notes of examples to each question in order to be able to rephrase them during the interview if the respondents did not understand. Secondly, the modified first draft was tested with a pilot interview with a respondent who had worked both within a fintech and healthtech start-up. The respondent and the firm were selected with the same sample criteria as the interviewees in order to achieve as accurate results as possible. The pilot interview helped us test if the questions were understandable from the respondents’ perspective as well as try out how much time was needed to answer each question (Ghauri & Grønhaug, 2010). After the test-interview, we merged the questions “What are your key resources and how have these changed the last two years?” and “What are your key capabilities and how have these changed the last two years?”
into one question since the pilot respondent referred to both resources and capabilities in each question. We also changed the phrasing on the question about the strategy to be more open- ended and focused on how their strategy now differs from when they first started instead of asking directly what within the strategy they have changed and why, which was the initial formulation. The rephrasing was necessary since we realized that asking very specific questions about the company strategy was sensitive and made the respondent uncomfortable.
After the third party check and pilot interview, the complete interview guide (see Appendix 2), was structured to start with general questions in order to make the respondent feel comfortable and familiar with the interview layout, to thereafter continue with theme-specific questions in order to provide deeper insights on business model adaptation and the drivers affecting the business model to change (Bryman & Bell, 2011; Ghauri & Grønhaug, 2010). Each question
was carefully thought out in accordance with the theory and compared to the research problem several times to test the consistency and make sure it would help answer the purpose and research questions of the study (Ghauri & Grønhaug, 2010). The guide was sent to the respondents a few days prior to the interview for them to read up on the purpose of the study and browse through the questions. Sending out the questions in advance is not always recommended as it gives the respondents the extra time to prepare and could lead to biased answers (Bryman & Bell, 2011). However, the topic of business models can be defined in many different ways, and sending out the questions in advance, therefore, allowed us to make sure that the respondents understood the purpose of the study and could give more in-depth answers.
A data protocol was also attached to the interview guide, informing the respondents on how the authors would handle the collected data.
Before starting the actual interview, the purpose of the study, the adopted definition of business models, and business model adaptation along with the structure of the interview was explained to the interview subject. Permission to record the interview was asked each respondent before the start and if approved they were informed that they had the right to stay anonymous and that it would only be used for analysis, to then be deleted after the transcription was complete to ensure confidentiality (Bryman & Bell, 2011; Ghauri & Grønhaug, 2010).
Throughout the interview, important notes were taken down as a part of the case protocol to be used as help when later processing the data, which also increases the reliability of the study (Yin, 2009). All interviews were conducted in English as this also functions as the language of communication between the authors.
A second round of interviews was also conducted with selected respondents through email, (see Appendix 3). The purpose of re-interviewing was to ask follow-up or complementary questions in regards to how the business model adaptation processes are managed and were sent to those who had not answered this in the original interview. This was done in order to secure the necessary data needed to answer the purpose of the study and increase the reliability and validity of the study (Ghauri & Grønhaug, 2010; Kirk and Miller, 1986).
3.3 Data analysis and coding process
Due to the exploratory research design and the goal of deeper insight and understanding of the phenomenon, the data analysis was conducted in line with Miles and Huberman's common
approach of analyzing qualitative data. The three-step process started with data reduction followed by data display and conclusion drawing (Miles & Huberman, 1994). Data reduction was performed by the authors separately through an individual content analysis inspired by the structure of Graneheim and Lundman (2004). As a first step in processing the collected primary data, all recorded interviews were transcribed in order to convert the data from audio into text.
In studies of exploratory design, the authors often tend to be overwhelmed with a lot of data and therefore need a structured way of processing the data, which also was the case in our study due to a total of 140 pages of transcription (Ghauri & Grønhaug, 2010). Hence, the most suitable methodology for processing and displaying the data was done in line with Gioia's method of structuring data when conducting an inductive research approach (Gioia et al., 2012).
Individual coding process and categorization of 1st and 2nd-order concepts
Processing of the same collected data was done separately through individual content analysis to increase reliability and not risk being influenced by each other's perception of the transcripted text (Ghauri & Grønhaug, 2010; Graneheim & Lundman, 2004). Hence, we started by making a copy of each transcription for both authors to go through the same material. To make sure that the individual coding easily could be merged into 1st order concepts, 2nd order themes, and aggregated dimensions in accordance with Gioia et al. (2012), we decided to follow the same structure when conducting the coding. A set of rules was decided between the authors to make sure the structure was followed. These rules indicated that both authors would first read through each interview several times and then mark everything interesting with yellow color and comment keywords on interesting observations. Thereafter, having our research questions in mind, quotes from all respondents that followed the same theme or pattern, and could help us answer the question were grouped together. Each group of quotes were then marked with one keyword representing a potential 2nd order theme and summarized with a few sentences in close relation to the transcripted answers representing the potential 1st order concept.
Merge of individual coding: agreement on concepts, themes, and aggregated dimensions After both authors had conducted their individual content analysis, the second round of coding was done by comparing each other's coding to see what concepts and themes we had found and to come to an agreement on the final structure. Each of us read through the other one’s content
interpretation of the data transcription. The second round of coding was done with help of coloring and both authors used the same strategy. To exemplify, the grouped quotes (1st order concepts) and the keywords of the groups (2nd order themes), that both authors had identified as being important to answer the two research questions, was marked with green color to automatically be added into a coding scheme. To make sure that we did not get any duplication of quotes, all concepts and themes that both agreed on were directly entered into a separate document. Red color marked one author disagreeing with the other one’s 1st order concept and 2nd order theme and orange color marked any finding identified by one of the authors but not by the other one. All red markings where one of us disagreed with the other were discussed back and forth until an agreement was reached on which 1st-order concepts and 2nd-order themes to display. All orange markings where findings made by only one of the authors were read and extra time by the other, and if agreed on to be helpful, it was also entered into the document.
After the coding was done, the second step recognized by Miles and Huberman's (1994) process of analyzing data was to display the chosen merged concepts, themes and aggregated dimensions from the separate document into a coding scheme in accordance to Gioia et al (2012), see figure 3. As the third step in the process, conclusions, and verification was drawn from the data to highlight how the concepts, themes, and aggregated dimensions are connected and impact each other. In order to get a full view of all coded quotes related to the 1st order concepts, see Appendix 4-8.
Figure 3: Data structure of concepts, themes, and aggregate dimensions in line with Gioia et al. (2012).
Cross-case analysis
In multiple case-studies the common approach is to also conduct a cross-case analysis after having looked into the within-case analysis (Eisenhardt, 1989). The cross-case findings from start-ups within the healthcare industry were set in relation to and compared with findings from start-ups within the financial industry. The cross-case industry analysis was inspired by Santos and Eisenhardt’s (2009) approach of rating reappearing mechanisms and Eisenhardt and Graebner’s (2007) way of matching recognized patterns of relationships in regards to the theoretical framework. The ratings are based on frequency and reaction. When the interviewee mentioned something without the interviewers indicating it beforehand, when they reacted strongly to something or the amount they talked about one theme were indicators for frequency and reaction. Strong identification with the 2nd order theme and the aggregated dimensions was characterized with ‘A’, a medium-strong identification was labeled ‘a’ and if there was zero identification to the theme or concept it was represented by ‘0’. The number 0 not only indicates that something was not mentioned but also when an interviewee negated a correlation between a theme and their business model adaptation. The outcome of the cross-case analysis is visualized in a pattern matching matrix at the end of the empirical section.
4 EMPIRICAL FINDINGS
4.1 Challenges
Both external and internal drivers of business model adaptations were perceived by the respondents as challenges with most of the challenges highlighted to arise from the external environment. A common trait seen from most respondents' answers was that digital start-ups from the same industry emphasized similar or even the same challenges. Across all empirical data, four second-order themes, foreign market adaptation, industry dynamics, funding, and legislation were presented as challenges and thereby identified as strongly influencing business model adaptation.
Foreign market adaptation
Adaptation to new foreign markets when internationalizing their business was seen as a challenge for multiple respondents within both fintech and healthtech start-ups. Entering new markets required adaptation by completely changing the whole business model to fit the local health care system or adaptation on parts of the business model in terms of needing to rebuild the product to fit the local regulatory system. This is particularly strengthened by the following respondents:
“Since we are in digital health, our challenge is to adapt the business model and revenue model to the markets we enter, so, unfortunately, we can't have the same model in all countries. [...] We use the UK and Norway to test out other types of models because we realized that the Swedish model is quite
unique compared to most other countries”. - Respondent D
“First of all, when we went into the UK, we had to adapt to a new currency into our system. We had to equip a new payment scheme. Then obviously for the US, that's an even larger change. So, we are also live in the U.S. market and there, we had to completely rebuild our products to adapt it to the
local market and to make it function there in the first place”. - Respondent A
Two of the case firms had also chosen internationalization into a distant foreign market in order to minimize competition and exploit the first-mover advantage. However, for one of them, the
