The impact of business model innovation on the performance of Romanian e-commerce startups

(1)

The Impact of Business Model Innovation on the

Performance of Romanian E-commerce Startups

Master’s Thesis

Student: Aurelia Teslaru/ Student № 10825568/

Amsterdam Business School, University of Amsterdam MSc. in Business Administration – Strategy Track Supervisor: Dr. Stephan von Delft

Date of submission: 28th June, 2015 Version: Final

(2)

2

Statement of originality

This document is written by Student Aurelia Teslaru who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Abstract

Trial-and-error and experimentation, or more broadly speaking, learning have been recognized as determinants of business model innovation. At its turn, business model innovation is believed to influence the performance of companies. Therefore, this paper will study the mediating role of business model innovation in the relationship between learning and performance. This study takes an entrepreneurial perspective and aims to determine if the relationships described above applies on Romanian e-commerce startups. Moreover, the moderating roles of entrepreneur’s experience and environmental dynamism are analyzed in relation to learning, performance and business model innovation. The results, however, showed no significance for the mediating effect of business model innovation. On the other hand, entrepreneur’s experience and

environmental dynamism were found significant in moderating the relationship between learning and performance. The results provide useful insights for both theory and practice, as well as directions for future research.

(5)

5

The Impact of Business Model Innovation on the Performance

of Romanian E-commerce Startups

1. Introduction

Both established firms and startups deliver their products or services into the market through business models. According to Chesbrough & Rosenbloom (2002), the value of that product or service remains latent until a suitable business model is deployed. The authors recognize the importance of finding the right fit between the business model and company’s goals in order to achieve success.

Business model innovation plays an important role in the success of companies, given the fact that nowadays the products’ life cycles are shortening (Chesbrough, 2007; Margretta, 2002; Zott & Amit, 2010). In addition, in the existing literature, business model innovation has been associated with experimentation and trial-and-error learning, or more broadly speaking, with learning (Sosna, Trevinyo-Rodríguez, & Velamuri, 2010; McGrath, 2010; Chesbrough, 2010; Chesbrough, 2007). Therefore, learning (experimentation, trial-and-error) leads to business model innovation, which in turn increases the performance of companies.

The Internet provides companies with the opportunity to sell products and services around the world, 24 hours a day while in the same time it drastically reduces costs and increases convenience both for customers and firms (Reifer, 2002; Evanschitzky, Iyer, Hesse, & Ahlert, 2004). Moreover, in the last years, internet became the main market gate for startups (Chang, 2004). According to Zott, Amit, & Massa (2011), ‘doing business electronically’ is defined as e-commerce. Therefore, a study about how the two learning mechanisms influence the success of

(6)

6

e-commerce startups is needed in order to improve firms’ performance and provide entrepreneurs with useful insights about the underlying mechanisms of business model innovation.

Given the importance of designing an appropriate business model, little research has been conducted in the field of entrepreneurship (Zott & Amit, 2010; George & Bock, 2011). It is known that the success and growth potential of startups is linked with the development of business models (George & Bock, 2011). In this study, the concepts of experimentation and trial-and-error learning will be analyzed in relation with business model innovation in e-commerce startups in order to determine how performance is influenced. Learning will be analyzed through the lens of the following three dimensions: strategic knowledge dissemination, strategic knowledge interpretation and strategic knowledge implementation (Sirén, Kohtamäki, & Kuckertz, 2012).

This paper contributes to both theory and practice. On the one hand, given the fact that the literature on business models in general lacks consistency and the research on business models in the entrepreneurial field is fragmented (George & Bock, 2011; Zott & Amit, 2010; Zott, Amit, & Massa, 2011), this paper aims to bridge the gap between business model innovation and entrepreneurship by examining the influence of two processes: experimentation and trial-and-error learning on business model innovation and performance in e-commerce startups. Experimentation and trial-and-error can be referred to as processes because they both employ activities which rely on resources (time, money, employees) in order to get to the desired result (change, higher performance).

On the other hand, the implications for practice are also noticeable. By analyzing the relationship between the two processes and the performance of startups in e-commerce, entrepreneurs can

(7)

7

benefit from useful insights on how to design their business model in order to create and appropriate more value. Value creation and appropriation are central elements of business models (Chesbrough, 2007; Teece, 2010). Therefore this study delivers insights, useful for increasing the chances of success of startups in e-commerce by exploring how emerging companies can create and appropriate more value.

This paper is structured as follows: first, a review of existing literature on the subject in matter is developed; second, the theoretical framework is outlined; third, the research design is described; subsequently, the empirical results are outlined followed by a discussion section; finally, the conclusion and contributions of this paper are presented in relation with theory and practice.

2. Literature review

2.1 Introduction

The following chapter discusses the main insights from the literature on business models, business model innovation and the link between business models and entrepreneurship. I start by elaborating the evolution of the field of business models and by highlighting the key concepts. Subsequently, I describe the importance of business model innovation on firm performance and success and the role of experimentation and trial-and-error learning on achieving innovation and thus, competitive advantage. Then, the Romanian entrepreneurial ecosystem will be described. Finally, I further explore the link between business model innovation and the entrepreneurial field which will lead to the identification of the research gap.

2.2 The emergence of business models and key concepts

The study of business models is a relatively new field. Although business models have been used for a long time, most of the research in this topic was conducted after the year 2000, confirming

(8)

8

that the concept of business model gained momentum together with the internet boom (George & Bock, 2011; Margretta, 2002; Zott, Amit, & Massa, 2011). Given the fact that research in the field started more prominently a few years ago, there is little agreement in what concerns the definition and the conceptualization of the term. Moreover, the concept of business model is often studied without clearly defining it, which leads to even bigger confusion. Frequently, authors use the term in a manner that fits their research purpose without adopting a universal definition, which delays the progress in this field (Zott, Amit, & Massa, 2011).

However, there are a few concepts which appear to form a common ground on how a business model can be defined. In most of the research conducted, business models are described in terms of value creation and appropriation. More specifically, how firms deliver value to customers and how they manage to make customers pay for the value delivered, converting it in profits (Teece, 2010; Amit & Zott, 2001; Chesbrough, 2007). Two key concepts which appear to form common ground in business model definition are the value proposition and the revenue stream. The value proposition is focused on customers, on how to satisfy their needs or more broadly, on value creation while the revenue stream refers to how the company can benefit from the value created for customers or more broadly, value appropriation (Johnson, Christensen, & Kagermann, 2008; Osterwalder & Pigneu, 2009). Therefore, developing a good business model is an essential step toward success and performance, both in new ventures and in established companies (Margretta, 2002; Zott, Amit, & Massa, 2011).

2.3 The importance of business model innovation

Teece (2010, p.20) describes business models as being “management’s hypothesis about what customers want, how they want it and what they will pay, and how an enterprise can organize to best meet customer needs, and get paid well for doing so’’. Business model innovation goes a

(9)

9

step further and takes a more dynamic approach in the sense that it acknowledges the need of constantly improving the business model in the face of competition (Sosna, Trevinyo-Rodríguez, & Velamuri, 2010).

Business model innovation can enhance competitive advantage and help a company differentiate from competitors (Markides & Charitou, 2004).One of the first steps in designing a new business model is keeping in mind competitors (Casadesus-Masanell & Ricart, 2011). Companies can improve their business models by weakening those of competitors and thus, reducing their importance in the market or by strengthening their own business model and create competitive advantage relative to competitors. However, a business model can provide a strong competitive advantage only as long as it is difficult to replicate (Margretta, 2002; Teece, 2010).

By constantly improving their business models, companies can keep competitors away and differentiate from them. However, firms must pay attention to a few aspects when engaging in business model innovation. Lindgardt, Reeves, Stalk, & Deimler (2009) identified a few potential problems when it comes to business model innovation: too many attempts to innovate in the same time can lead to a lack of coordination and also, the company might not have enough resources to support them; managers might hold on to their idea even if it is not a promising one; too many ideas without trying to scale them can lead to false expectations regarding the validity of the ideas.

Researchers agree on the fact that the key of creating competitive advantage through business models rests in innovation, or more specifically in learning from experimentation and trial-and-error (Chesbrough, 2010; McGrath, 2010; Sosna, Trevinyo-Rodríguez, & Velamuri, 2010; Thomke, 2003; Andries, Petra, Debackere, & Looy, 2013). Learning is seen as a renewal

(10)

10

mechanism which is used in order to achieve business model innovation (Sosna, Trevinyo-Rodríguez, & Velamuri, 2010; Chesbrough, 2010; Johnson, Christensen, & Kagermann, 2008). Also, learning can be used when managers or entrepreneurs intuit that a change in the business model is needed but they cannot articulate it. In other words, the desired and necessary business model is not apparent (Teece, 2010). Managers and entrepreneurs which are able to learn and adapt are thus, more likely to succeed. Therefore, learning plays an important role in business model innovation.

The learning mechanism, together with the two processes described above, can be better understood through the lens of the following dimensions: strategic knowledge dissemination, strategic knowledge interpretation and strategic knowledge implementation (Sirén, Kohtamäki, & Kuckertz, 2012). After initiating trial-and-error or experimentation, the process of strategic knowledge dissemination proves useful. This process refers to sharing the information between the startup’s employees. After sharing the insights, people start to develop their own

interpretation of the events and then, by interacting with each other they will develop a shared understanding of the subject in matter. This is referred to as strategic knowledge interpretation. The last step is the implementation of the strategic knowledge gained into subsequent activities, which might as well be further trial-and-error or experimentation initiatives (Sirén, Kohtamäki, & Kuckertz, 2012).

The terms ‘experimentation’ and ‘trial-and-error’ are used interchangeably in the recent literature

(Chesbrough, 2007; Nicholls-Nixon, Cooper, & Woo, 2000). While some authors focus only on one of them with respect to business model innovation, others use the terms alternatively. Next, I will explain the use of each term and form a common ground with regard to their meaning and importance for innovation.

(11)

11 2.3.1 Experimentation

Experimentation allows companies to try different business models, by experimenting with several possible choices before adopting a specific one (Chesbrough, 2010). Experimentation is important because it presses companies to develop and improve the quality of their products, processes and systems (Thomke, 2003). An important aspect which must be taken into consideration when experimenting with business models is, according to Chesbrough (2010), to maintain the high fidelity of the experiment. This refers to the degree to which the experimental conditions in which the experiment is conducted are representative for the market itself. More specifically, the dynamics of the market and the conditions under which it operates must be replicated in order to ensure the reliability of the experiment. Also, in order for experimentation to occur, employees have to have the required degree of authority to undertake the necessary steps.

There are several forms in which experimentation can take place: some companies might form small startups companies in order to explore alternative business models while others use joint ventures and spin-offs to experiment outside their current business model (Chesbrough, 2007). However, the advantages of experimentation are not immediately seen. The results must be interpreted and understood in order to make the best out of experimentation (Chesbrough, 2007; Chesbrough & Rosenbloom, 2002). A further step towards the interpretation of the results is achieved through strategic knowledge dissemination and sharing the information (Sirén, Kohtamäki, & Kuckertz, 2012).

To sum up, experimentation with business models is used for describing the processes associated with trying multiple alternative paths and choosing the most suitable one. In order for experimentation to take place there are several requirements which must be met: maintaining the

(12)

12

high fidelity of the experiment, employees need to have the right degree of authority to conduct the experiment and the results must be interpreted in an appropriate manner.

2.3.2 Trial-and-error

According to Sosna, Trevinyo-Rodriguez, & Velamuri (2010), in order to achieve long term success, sustained business model innovation is a key factor. They argue that the changing nature of markets can quickly determine old business models to become obsolete. Companies need to adapt their business model to the market needs in order to stay competitive. The authors propose that competitiveness can be achieved through experiential error. They refer to trial-and-error as a process through which the business model of a company is initially developed as an experiment, followed by trial-and-error learning.

Trial-and-error can be seen as a component of the business model life cycle and development. At the establishment of a company, after some features of the business model are defined, trial-and error is used to refine it and develop the direction the firm should follow (Morris, Schindehutte, & Allen, 2005).

Zahra, Sapienza, Davidsson (2006) also place the concept of trial-and-error in an entrepreneurial frame. However, in their view, trial-and-error is only triggered when firms must cope with unplanned circumstances and when their survival is threatened. Also, they argue that trial-and-error is used to inform future decisions and actions only at an early stage of development of firms.

Even though some authors use these terms interchangeably (Chesbrough, 2007; Nicholls-Nixon, Cooper, & Woo, 2000), they have different characteristics. The goal of trial-and-error learning is to inform future decisions through undertaking actions either planned or unplanned while the

(13)

13

goal of experimentation is to assess the relationship between cause and effect through the deliberate use of different conditions. However, both learning mechanisms are part of the innovation process in a company and they both contribute to a greater performance of firms (Chesbrough, 2010; Chesbrough & Rosenbloom, 2002; Sosna, Trevinyo-Rodríguez, & Velamuri, 2010; Zahra, Sapienza, & Davidsson, 2006). Also, they both require strategic knowledge dissemination, interpretation and implementation in order to implement the needed changes.

2.4 Business model innovation and e-commerce startups

In the field of entrepreneurship, the lack of a consistent concept regarding business models has led to inconsistent research questions and findings (George & Bock, 2011; Zott & Amit, 2010). Entrepreneurship is described as the mean by which firms and employees explore and exploit opportunities (March, 1991). One of the most important goals of entrepreneurs is wealth creation (Ireland, Hitt, Camp, & Sexton, 2001). E-commerce provides the means of generating wealth (Evanschitzky, Iyer, Hesse, & Ahlert, 2004). Therefore, more and more startup companies choose to start their business on the internet (Chang, 2004).

There are several advantages of using the internet, both for consumers and startup companies. From the point of view of the firm, the startup has easier access to customers and lower costs than opening a brick and mortar store (Stewart & Zhao, 2000). From the point of view of customers, they can more easily evaluate the alternatives before buying a specific product, they can take a more informed decision, the access to information is easier and it saves time (Stewart & Zhao, 2000; Evanschitzky, Iyer, Hesse, & Ahlert, 2004; Bhatnagar, Misra, & Rao, 2000)

Wealth creation can be accelerated through the two learning processes: experimentation and trial-and-error which lead to innovation. An advantage of small companies, especially the

(14)

14

companies where the CEO is also the owner, is that experimentation can occur more easily (Chesbrough, 2010). This happens because the owner or CEO has the necessary authority to conduct the experimentation process. The business model of young companies or startups will often be influenced by the previous experience of the founder (Sosna, Trevinyo-Rodríguez, & Velamuri, 2010).

For entrepreneurs, business models play a very important role. Some of them even apply for patents in order to protect their business model (Morris, Schindehutte, & Allen, 2005). In some cases, entrepreneurs start with a well-defined business model but in most cases, they have partially formed models which lead to a need for further developing them (Morris, Schindehutte, & Allen, 2005). Further development of business models can be achieved through implementing the two learning processes described above.

The goal of entrepreneurs is to discover and exploit opportunities in order to create new products and services (Venkataraman, 1997). Entrepreneurial opportunities arise when there are competitive imperfections (Andries, Petra, Debackere, & Looy, 2013). However, the performance of the entrepreneurial venture depends on the exploitation of the right opportunities and also on innovation. Even though there is plenty literature on business models in developed and mature organizations, little is known about how startup companies from e-commerce cope with business model innovation and how this affects their performance.

Moreover, it is known that experimentation and trial-and error lead to innovation and therefore to competitive advantage but companies still find it hard to benefit from it. Research shows that two reasons why this might happen are the following: the lack of definition of the concept and the

(15)

15

fact that companies do not fully understand how their business model works (Johnson, Christensen, & Kagermann, 2008).

2.5 The importance of entrepreneurial ecosystems

In order to grow and develop, new ventures or startups need to have the right resources. Finding the right resources can be hard but this is where the entrepreneurial ecosystems come in handy (Mason & Brown, 2014). According to the authors, new ventures provide the right combination of networks, employees, capital services and governmental support which leads to a better development environment for the young companies. Even though the term ecosystem is usually used at a smaller scale, referring to a particular area from a country, further I will analyze the Romanian entrepreneurial ecosystem at the level of the whole country. This is particularly important for this study because it provides a deeper understanding of the dynamics and resources that Romania has to offer to its entrepreneurs and how this can influence the development of its startups.

Neck et al. (2004) elaborate on the components of entrepreneurial ecosystems and describe them. According to the authors, the components are as follows: informal network (entrepreneur’s friends and family as well as the relations with other companies) and the formal network (universities, government, capital sources, support services, corporations and talent).

The informal network is a valuable component because the relations with other companies can provide new technological insights and also possible future partnerships. Even though some studies show that networking is most of the time useless, in an entrepreneurial environment being up to date all the time is an important step towards success (McKeown, 2015). Friends and family also have a key role as they can provide the entrepreneur with money and support.

(16)

16

The formal network is extrinsic to the entrepreneur meaning that he or she cannot have a direct effect on it. The actions taken by the government are important mostly because of the regulations imposed. Entrepreneurs will always prefer areas where the regulations are mild and in their advantage. The capital sources are also an important influencer in deciding where to base a startup. Most of the new ventures need investors or other sources of money in order to grow. The existence of corporations or big companies is important because it draws skilled workers and talent which might eventually leave the company in order to work for other emerging companies in the area or even start their own businesses.

The universities add value through developing new talents and workers. Also, they can develop new technologies and make significant breakthroughs which can help a particular industry in the entrepreneurial ecosystem. Feld, (2012) considers as the most important contribution of universities the ideas that emerge from students. Consultants and legal support also have an important role because they can provide the startup with insights about their service or product and its viability in the current market conditions.

Drawing from the above components, the entrepreneurial ecosystem is strongly influenced by the geographic area where it is located. Mason & Brown, (2014) include several dimensions in describing the entrepreneurial ecosystem such as: the resource providers within the ecosystem, entrepreneurial connectors and the entrepreneurial environment. Although the resource providers and the entrepreneurial environment are elements discussed by Neck et al. (2004), Mason & Brown, (2014) bring a new, more dynamic perspective through the entrepreneurial connectors. They refer to connectors as to formal or informal organizations (entrepreneurship clubs, groups, associations) as well as to individuals. These connectors have the role to foster the connections between the components of the entrepreneurial ecosystem. One of the most important

(17)

17

components of an entrepreneurial ecosystem is the interacting nature of the actors (Spilling, 1996).

2.5.1 The Romanian entrepreneurial ecosystem

Entrepreneurial activity can contribute to the development and transformation of former communist societies such as Romania (Mueller & Goic, 2002). Once the Soviet empire has collapsed, Romania has faced an increase in the entrepreneurial activity (Pistrui, Welsch, & Roberts, 1997; Gundry & Ben‐Yoseph, 1998).

Based on the elements described in the previous section, the entrepreneurial dynamics in Romania will be analyzed. I will start by elaborating on the informal networks available in Romania. Then, the formal network will be presented together with the connectors.

The informal network

In what concerns the informal network, this is composed from the entrepreneur’s friends and

family and also the connection with other companies (Neck, Meyer, Cohen, & Corbett, 2004).. Behind the Romanian entrepreneurial activity, there is the desire and the strive for family security, accomplishment and freedom (Pistrui, Welsch, & Roberts, 1997). Therefore, family occupies a primary role in the life of Romanian people. Also, another important aspect is the fact that Romanian entrepreneurs are very picky when it comes to choosing their partners because of the overall high level of mistrust in this country (Pistrui, Welsch, & Roberts, 1997; Sandu, 1996; Zamfir, 1994). This leads the entrepreneurs to seek for help in the immediate family. Regarding the relationships with other companies, emerging businesses in Romania are particularly interested in developing connection with businesses from the Western Europe, as they seem to achieve higher performance (Pistrui, Welsch, & Roberts, 1997).

(18)

18 The formal network

Universities and talent

People in Romania have a relatively high level of education. The majority of them have a university degree, with an average of 15 years of education (Pistrui, Welsch, & Roberts, 1997; Lafuente & Driga, 2007). According to Topuniversities.com, there are four universities in Romania classified in top 800 worldwide: University of Bucharest, Alexandru Ioan Cuza University, Babes-Bolyai University and West University of Timisoara. Also, according to the same website, Bucharest and Cluj-Napoca represent the biggest student cities and offer the best education quality and the widest choice of subjects from Romania. The costs of studying in Romania are anything but low: between 2000€ and 5000€ per year.

Government

Before the fall of the communism in 1989, Romania had one of the biggest state-dominated economies with over 95% of the total businesses being nationalized. Therefore, the entrepreneurial activity took off in Romania only after 1989 when the President Ceausescu fell from power (Pistrui, Welsch, & Roberts, 1997). Only after 1990 it was legal to own a business when the Government authorized the establishment of private small businesses (Neef, 2002; Pistrui, Welsch, & Roberts, 1997). By the end of December 1990, there were about 100.000 small businesses opened (Ben-Ner & Montias, 1991).

Nowadays, the Romanian government is poorly involved in the entrepreneurial activity and support. For the registration of startup, the approval of three different agencies is necessary: the Ministry of Finance, the Ministry of Labor and Social Protection, and the Registry of Trade and Commerce, which leads to a high level of bureaucracy and wasted time (Brown, Earle, & Lup,

(19)

19

2005). However, Romania is well known for the high number of IT workers, having one of the highest numbers of technology workers per capita on the continent: close to 64.000 IT specialists, according to Forbes.com. One advantage is the fact that the IT workers are exempted from paying the tax on their income. This is a measure which encourages the development of the IT sector and, indirectly the entrepreneurial environment because, nowadays, most startups are internet based.

Capital sources

It appears that the primary source of capital for entrepreneurs remains their family followed by their friends (Pistrui, Welsch, & Roberts, 1997). Other possible sources of capital are: bank loans, incubators and investors. In what concerns the investors, there are few Romanian investors. Therefore, most of the startups which become successful, received investment from foreign parties. Examples of such startups are: LiveRail, Trilulilu, Mavenhut and UberVu.

Support services

The support services include the consultants, legal support and the firms in the supply chain (Neck, Meyer, Cohen, & Corbett, 2004). Romania hosts some of the biggest consulting firms around the globe such as: McKinsey, Accenture, KPMG, BearingPoint and Deloitte. Therefore, the consulting offering is good. In what concerns the legal support, there are many businesses which offer this type of service such as law firms. The firms in the supply chain vary from company to company.

Corporations

Corporations are an important element since they can be considered suppliers of talent. They attract the best workers from the area they are located in and those workers, using the skills

(20)

20

learned at their jobs, might be to ones who start the next successful startup. In Romania, the biggest concentration of corporations is in Bucharest, followed by Cluj-Napoca, the same cities with the biggest concentration of students. Therefore, it can be concluded that universities play an important role in creating high skilled workers in the areas described above.

Connectors

The entrepreneurial connectors consist in entrepreneurship clubs, groups or associations (Mason & Brown, 2014). An example of an entrepreneurship club in Romania is Young Leaders Club, which aims to bring its contribution to the Romanian business community. There are also other associations such as the Alternative University and Incubator 107 which bring a plus to the entrepreneurial community through organizing trainings and conferences with an entrepreneurial focus. The Romanianstartups.com website also contributes by linking the possible investors with the entrepreneurs and promoting the registered startups. However, the most important element remains the people, who have to opportunity to connect through the resources presented above.

2.5 Literature gap and research question

Researchers agree that an important step towards creating competitive advantage rests in business model innovation, and that experimentation and trial-and-error learning play an important role in it, together with strategic knowledge dissemination, interpretation and implementation (McGrath, 2010; Markides & Charitou, 2004; Chesbrough, 2010; Thomke, 2003; Sosna, Trevinyo-Rodríguez, & Velamuri, 2010; Andries, Petra, Debackere, & Looy, 2013; Chesbrough & Rosenbloom, 2002). Business model innovation emphasizes the importance of renewal and improvement in order to stay competitive (Chesbrough, 2010; Sosna, Trevinyo-Rodríguez, & Velamuri, 2010; Johnson, Christensen, & Kagermann, 2008).

(21)

21

Researchers identified two processes which lead to business model innovation: experimentation and trial-and-error learning (Andries, Petra, Debackere, & Looy, 2013; Chesbrough, 2007; Chesbrough & Rosenbloom, 2002; Sosna, Trevinyo-Rodríguez, & Velamuri, 2010). The two learning mechanisms hold an important role in business model innovation. When managers or entrepreneurs see the need for change but they do not know exactly how the change can be achieved, they use learning mechanisms which can be translated into experimentation and trial-and-error learning, followed by the dissemination, interpretation and implementation of the knowledge (Teece, 2010; Sirén, Kohtamäki, & Kuckertz, 2012). However, some companies still find it difficult to exploit the opportunities associated with these two learning processes (Johnson, Christensen, & Kagermann, 2008).

Research on business model innovation in the entrepreneurial field lead to inconsistent findings (Zott & Amit, 2010; George & Bock, 2011).Therefore, this paper aims to bridge the gap between the entrepreneurial field and business model innovation through the exploration of the effects of experimentation and trial-and-error learning on startup companies from e-commerce, Romania.

The research question and the sub-questions of this paper can be summarized as follows:

’’How does business model innovation influence the performance of e-commerce startups from Romania?’’

 Does business model innovation positively impact the performance of e-commerce

start-ups?

 How does business model innovation mediate the relation between learning

(22)

22

The above questions hold important contributions both to theory and practice. From the theory’s point of view, this paper contributes by addressing the gap between the entrepreneurial field and business model innovation. Moreover, it also contributes to practice by providing entrepreneurs insights about how experimentation and trial-and-error learning influences the performance of their startups and how they can use these two processes in order to create competitive advantage.

3. Theoretical framework

This academic paper theorizes that performance in e-commerce startups is influenced by two types of learning: experimentation and trial-and-error. E-commerce startups can be defined as young companies which have the potential of achieving profitability and do their business electronically (Zott & Amit, 2007; Zott, Amit, & Massa, 2011). In addition, it has been argued that experimentation and trial-and-error learning lead to business model innovation, which in turn may have an impact on performance (Andries, Petra, Debackere, & Looy, 2013; Chesbrough, 2007; Chesbrough & Rosenbloom, 2002; Sosna, Trevinyo-Rodríguez, & Velamuri, 2010; Thomke, 2003). Therefore, the impact of learning on performance through business model innovation will be analyzed in e-commerce startups. This paper will also study the moderating roles of the entrepreneur’s experience and environmental dynamism on the performance of startups.

3.1 The relationship between learning and performance in e-commerce startups

This study defines learning as a firm’s initiative to engage in experimentation or trial-and-error,

followed by the processes of strategic knowledge dissemination, interpretation and implementation. Given the low level of maturity of startups, their past experience is limited and,

(23)

23

therefore, in order to achieve performance they must undergo learning processes. Research shows that experimentation and trial-and-error both have a positive impact on performance (Nicholls-Nixon, Cooper, & Woo, 2000). However, most of the studies focus on mature firms and few analyze the impact of these two processes on the performance of e-commerce startups. Therefore, this paper tries to fill in this gap.

3.1.1 Experimentation and performance

When starting a new business, entrepreneurs are forced to develop a good understanding of the competitive environment they are playing in. This helps them create a strategy for successfully competing with rivals in that particular environment (Nicholls-Nixon, Cooper, & Woo, 2000). One way of doing this is through experimental learning, which focuses on learning by doing (Zahra, Ireland, & Hitt, 2000). Thus, the survival and performance of startups depends on their ability to process inputs and make adjustments accordingly.

The goal of experimentation is to analyze the relationship between cause and effect deliberately, by using different conditions (Zahra, Sapienza, & Davidsson, 2006). Therefore, through continuously improving the quality of their products or services, startups can achieve competitive advantage and increase their performance (Thomke, 2003; McGrath, 2010).

3.1.2 Trial-and-error and performance

Zahra, Sapienza, Davidsson (2006) argue that trial-and-error is an entrepreneurial process characteristic to an early stage of development of firms. It is used to inform future decisions and actions and to achieve competitiveness. Trial-and-error also enhances learning about both exploration and exploitation, leading to change or stability in a company according to its needs (Sosna, Trevinyo-Rodríguez, & Velamuri, 2010). This results in having a better sense of the

(24)

24

direction the company must follow and therefore, in better performance (Morris, Schindehutte, & Allen, 2005).

Both experimentation and trial-and-error are learning processes. It has been argued above that they positively influence performance. However, in order for these processes to be successful, they must be followed by strategic knowledge dissemination (sharing the knowledge), strategic knowledge interpretation and strategic knowledge implementation (Sirén, Kohtamäki, & Kuckertz, 2012). Therefore, this paper argues that learning in e-commerce startups positively influences their performance.

Hypothesis 1: Learning positively affects the performance of e-commerce startups.

3.2 Business model innovation as a mediator between learning and performance.

Innovation becomes an important determinant of performance in nowadays’ companies, mostly

because rather than investing in technology or R&D, business model innovation is more efficient and less costly (Chesbrough, 2007). However, even when a new business model is necessary, the company must make sure that the current business does not prevent in some way the creation of value through the new business model (Johnson, Christensen, & Kagermann, 2008).

In what concerns entrepreneurial firms, entrepreneurs often try to change the current way of doing business in order to create disruptions in the market and thus, develop new business models (Ireland, Hitt, Camp, & Sexton, 2001). For startup firms it is very important to innovate in order to find their place in the market and position themselves. Competitive advantage can be achieved through business model innovation, which subsequently leads to greater performance (Margretta, 2002; Casadesus-Masanell & Ricart, 2011).

(25)

25

Research shows that both experimentation and trial-and-error learning lead to business model innovation (Andries, Petra, Debackere, & Looy, 2013; Chesbrough, 2007; Chesbrough & Rosenbloom, 2002; Lindgardt, Reeves, Stalk, & Deimler, 2009; Margretta, 2002; Sosna, Trevinyo-Rodríguez, & Velamuri, 2010). These two concepts fit best in an entrepreneurial context due to the lack of experience and history of young companies (Zahra, Sapienza, & Davidsson, 2006). They constitute two learning mechanisms which allow young companies to try different alternatives and cope with the uncertainty of the future (Morris, Schindehutte, & Allen, 2005; Zahra, Sapienza, & Davidsson, 2006).

Therefore, business model innovation mediates the relationship between learning (experimentation, trial-and-error) and performance in the following way: learning determines business model innovation which, in turn, leads to greater performance companies.

Hypothesis 2: Business model innovation mediates the relationship between learning and performance.

3.3 The moderating roles of entrepreneur’s experience and environmental conditions

3.3.1 Entrepreneur’s experience

The experience of an entrepreneur can be defined as the number of years he or she accumulated after finishing studies (Robinson & Sexton, 1994). However, nowadays, people start working at earlier ages and even during their studies. Other authors defined experience as the number of previous new ventures in which the entrepreneur was involved and the management role he/she played in them (Stuart & Abetti, 1990). No matter how experience was quantified, the results of the studies were the same: previous experience is positively related to performance.

(26)

26

For the purpose of this paper, the most relevant in assessing the entrepreneur’s experience is

whether the entrepreneur started other startups in the past and whether he worked as a manager in an organization. Experience can influence the learning processes, the degree of business model innovation and also performance. Previous startup experience is important because it enables the entrepreneur to accumulate knowledge about the steps necessary to develop a company and also it helps him in establishing a powerful network. Previous experience as a manager impacts the skills the entrepreneurs has, more specifically his capability of managing people and challenging situations.

Previous experience can also influence how the entrepreneur decides to improve his/her business model: either through experimentation, trial-and-error or both. If he experienced similar situation in previous new ventures, he might choose the same path again. Also, performance hinges on the success of experimentation and trial-and-error and the definition and implementation of a suitable business model which presents competitive advantage (McGrath, 2010). Therefore, a rich entrepreneurial experience is expected to positively influence the performance of startups.

Hypothesis 3: Entrepreneur’s experience positively moderates the relationship between learning and business model innovation.

Hypothesis 4: Entrepreneur’s experience positively moderates the relationship between business model innovation and performance

Hypothesis 5: Entrepreneur’s experience positively moderates the relationship between learning and performance.

(27)

27 3.3.2 Environmental conditions

The environmental conditions are important factors between both relationships: learning-business model innovation and learning-business model innovation-performance. Dess & Beard (1984) identify three environmental dimensions which can affect performance: munificence, dynamism and complexity. This paper will focus on the second dimension, dynamism. Given the fact that this paper focuses on the e-commerce business means that the munificence dimensions, which refers to extent to which the environment is able to sustain growth, is not applicable because internet provides a lot of space for growth. Also, this paper focuses on more than one industry in e-commerce. As a result, in what concerns the complexity of the environment, it is always high and the industry is heterogeneous. The only environmental dimension which is bound to vary in the context of this paper, is the dynamism.

Dynamism refers to whether the environment is stable or unstable. In a stable environment, the two learning processes are expected to lead to greater business model innovation and therefore, to higher performance.

Hypothesis 6: Environmental dynamism positively moderates the relationship between learning and business model innovation.

Hypothesis 7: Environmental dynamism positively moderates the relationship between business model innovation and performance.

Hypothesis 8: Environmental dynamism positively moderates the relationship between learning and performance.

(28)

28

Figure 1: Conceptual model

4. Methodology

In the following chapter I will discuss the research design of this paper. Firstly, I will describe the type of data collection method used: the survey and explain the sample. Then, I will argue the chosen data analysis approach. Finally, I will give a detailed description of the variables used and elaborate on the measurements used.

LEARNING

 Experimentation  Trial-and-error 

BMI

PERFORMANCE

Entrepreneur’s experience Environmental dynamism Control variables:  Firm age  Firm size  Number of founders  Founder’s age

(29)

29 4.1 Research design

The survey questionnaire is used for the measurement of the following variables: learning, business model innovation, performance, entrepreneur’s experience, environmental dynamism as

well as for the measurement of the control variables: firm age, firm size, number of founders, founder’s age and founder’s education. This approach was chosen because it allows the

researcher to gather standardized information (same survey is used for each of the respondents) across a predetermined sample (Saunders & Lewis, 2012). Furthermore, through the survey, the data can be more easily analyzed due to its standardized structure.

The survey has been sent to the respondents via e-mail or individually, via Facebook. This method was chosen because it is more efficient in terms of time and flexibility. Also, given the fact that the respondents are located in Romania, online methods represent the best alternative. However, a possible limitation is the fact that the respondents must have a personal computer in order to take part in this survey.

4.2 Sample

In order to test the hypotheses, I analyzed young firms which conduct their business online and derive revenue from the internet, based in Romania. The data for this study was collected in 2015 from an online database containing a list of Romanian startups (www.romanianstartups.com). The list provided by this website is not a complete list of all Romanian startups. It only displays the startups which were subscribed to appear on this site by their founders. Only e-commerce startups were selected from this database. The website provides information such as the e-mail of the founders or their Facebook pages. This information was used to send the surveys.

(30)

30

Given the fact that there is no sample frame available, I use non-probability sampling. For the purpose of this paper, I will use purposive sampling and snowball sampling (Saunders & Lewis, 2012). Purposive sampling is used to select the sample members based on several criteria: type of startup (only e-commerce startups will be analyzed) and company age (only young companies were selected). Snowball sampling was chosen due to the strong relation between entrepreneurs and the communication between them. With the help of this type of sampling, after identifying the first set of respondents, I will ask them to recommend me other entrepreneurs for the purpose of my analysis.

Another important aspect is that, for the purpose of this study, only the founders and co-founders of the startups analyzed will be considered as respondents. This is due to the fact that this paper aims to investigate the role of entrepreneur’s experience in the learning-business model innovation-performance relationship.

The data from the website mentioned above contained a list of 133 launched startups which means a total number of possible respondents of 133. A total of 78 people responded to the survey, resulting in a response rate of 58%. From the total of 78 respondents, 8 surveys were excluded due to their incompleteness leading to a final number of 70 surveys analyzed.

In order to test for nonresponse bias, the first wave of responses was compared with the second wave. The t-test analysis showed no significant differences at a level of .05 which leads to the conclusion that there are no concerns regarding the nonresponse bias.

(31)

31 Table 1: Sample statistics

Item Percentage

Age  Less than 21

 Between 21 and 30  Between 31 and 40  Between 41 and 50  Above 51 1.4% 52.8% 37.5% 8.3% 0% Level of education  Below Bachelor’s level

 Bachelor’s level

 Master’s level

 Doctorate level or above 8.6% 54.3% 35.7% 1.4% Hierarchical position  Founder

 Co-founder  Employee 33.3% 66.7% 0% Number of founders  1  2  3  4  5  More than 5 18% 40.3% 33.3% 8.4% 0% 0%

Firm age  Less than one year

 1 year  2 years  3 years  4 years  5 years  6 years

 More than 6 years

6.6% 17.7% 26% 22.2% 16.6% 5.7% 5.2% 0%

Firm size  1-10 employees

 11-50 employees  51-250 employees  251-1000 employees  10001-5000 employees  More than 5000 Employees 79.2% 20.8% 0% 0% 0% 0% 0%

(32)

32

Some limitations must be acknowledged regarding the sample. First, the sample was derived from a single website which did not contain the whole population of startups from Romania which might have resulted in reduced generalizability. However, no other source with the contacts of the entrepreneurs was available to the researcher. Another possible limitation is the fact that the electronic version of the questionnaire was composed in English which is not the mother tongue of the respondents. This might lead to misinterpreting some of the questions and inconsistent answering. However, English language is thought in Romanian schools from the second grade until finishing the university and therefore this should not be a barrier in obtaining accurate responses. Also, the survey was sent via internet and not everyone has internet access. However, this study analyzes e-commerce startups and the fact that some people do not have internet access should not represent a problem.

4.3 Key informants

Given the fact that this paper analyzes more than one variable, a multivariate method needs to be applied. Therefore, I will use the second generation multivariate method: PLS-SEM and the program SmartPLS v.2. Considering the fact that there are more structural equation modeling techniques (PLS-SEM, CB-SEM), I will explain the reasons behind choosing the PLS approach.

First of all, regarding the topic of this paper, as argued above, business model innovation will be analyzed as a mediator between learning and performance. The PLS-SEM technique is more exploratory which fits with the scope of this paper. Another reason of choosing this approach is the PLS-SEM is suitable for small samples (Hair Jr., Hult, Ringle, & Sarstedt, 2013). However, this can bring additional problems such as higher sampling error in the case when the sample and the population are heterogeneous in composition (Hair, Sarstedt, Pieper, & Ringle, 2012). The

(33)

33

sample in this paper is homogenous and it consists from young companies which conduct their business online.

Also, another important reason is the fact that PLS has higher statistical power than CB-SEM (PLS is more likely to show that a relationship is significant when it is so in the population) (Hair Jr., Hult, Ringle, & Sarstedt, 2013). One important benefit when using PLS is the possibility to process different types of variables such as: nominal, ordinal, interval and ratio variables which is of great help within this paper (Reinartz, Haenlein, & Henseler, 2009; Haenlein & Kaplan, 2004). In addition, PLS-SEM does not make any assumptions about the data distribution, which means that is suitable for both normal and non-normal distributions (Hair Jr., Hult, Ringle, & Sarstedt, 2013).

4.4 Measures and control variables Independent variable - Learning

This study adopts measurements already used by prior studies. It has been argued before in this paper that I define learning through the lens of experimentation and trial-and-error, followed by the three processes of strategic knowledge dissemination, interpretation and implementation. This is particularly relevant for the type of companies studied in this paper, namely startup companies or young companies, because they lack previous experience.

In order to measure this variable, I have used the measures developed by Sirén, Kohtamäki, and Kuckertz (2012). The authors use three dimensions for measuring learning: strategic knowledge dissemination, strategic knowledge interpretation and strategic knowledge implementation. The reason behind choosing these dimensions is their presence in previous studies about strategic learning (Crossan & Berdrow, 2003; Huber, 1991).

(34)

34

The three dimensions are evaluated on a five point Likert scales which provides a range of responses to questions or statements as follows: 1=strongly disagree, 5=strongly agree (Jamieson, 2004). As a consequence, the respondent’s perceptions are measured and not objective facts.

In analyzing the data, learning is defined as a second order reflective-formative construct. The relationships between the three dimensions: strategic knowledge dissemination, strategic knowledge interpretation, strategic knowledge implementation and learning are formative because each of the three dimensions measure another aspect of learning and they are not mutually exclusive (Hair Jr., Hult, Ringle, & Sarstedt, 2013). The relationships between the items and the three dimensions are reflective, as they measure the same thing and they are mutually exclusive.

It is important to distinguish between reflective and formative measures as each of them requires a different type of analysis. While reflective measurement models are analyzed based on their internal consistency reliability and validity, for the formative measurement models it is very important to ensure a high level of content validity before collecting the data and estimate the PLS path model. After these steps, the formative measures can be analyzed for: convergent validity, significance, relevance and collinearity among indicators (Hair Jr., Hult, Ringle, & Sarstedt, 2013).

Table 2: Independent variable: Learning (Source: Sirén, Kohtamäki, &Kuckertz, 2012)

No Item(s)

SKD_1 SKD_2

Strategic knowledge dissemination:

 Within our firm, sharing strategic information is the norm.

 Within our firm, strategically important information is easily accessible to those who need it most.

(35)

35 SKD_3

SKD_4

SKD_5

 Representatives from different departments meet regularly to discuss new strategically important issues.

 Within our firm, strategically important information is actively shared between different departments.

 When one department obtains strategically important information, it is circulated to other departments. SKI_1 SKI_2 SKI_3 SKI_4 SKI_5

Strategic knowledge interpretation:

 When faced with new strategically important information, our managers usually agree on how the information will impact our firm.

 In meetings, we seek to understand everyone’s point of view concerning new strategic information.

 Groups are prepared to rethink decisions when presented with new strategic information.

 When confronting new strategic information, we are not afraid to critically reflect on the shared assumptions we have about our organization.

 We often collectively question our own biases about the way we interpret new strategic knowledge.

SKII_1

SKII_2

SKII_3

SKII_4

Strategic knowledge implementation:

 Strategic knowledge gained by working groups is used to improve products, services and processes.

 The decisions we make according to any new strategic knowledge are reflected in changes to our organizational systems and procedures.

 Strategic knowledge gained by individuals has an effect on the organization’s strategy.

 Recommendations by groups concerning the use of strategic knowledge are adopted by the organization.

Mediating variable - Business model innovation

In order to measure business model innovation, I have used the four items developed by Huang, Lai, Lin, & Chen, (2013). In order to provide consistency in the survey, I have adapted their seven point Likert scale to a five point Likert scale. Research shows that both five and seven

(36)

36

point Likert scales produce the same results and they can be easily adapted one to the other (Dawes, 2008).

Business model innovation is defined as a reflective construct, following Huang et al, (2013).

Table 3: Mediating variable: Business model innovation (Source: Huang, Lai, Lin, & Chen, 2013) No Item(s) BMI_1 BMI_2 BMI_3 BMI_4

 Our company can help customers redesign their value propositions.

 Our company can redesign the company profit formula.

 Our company can develop a new business development process without negative effects for core business.

 Our company can confirm their key resources and processes to provide products and services to customers.

Dependent variable – Performance

For the measurement of performance, I have used the items developed by Sirén, Kohtamäki, &Kuckertz (2012). The authors focus on firms’ profit performance and use a five point Likert scale in order to measure the following dimensions: cash flow, return on shareholder’s equity,

gross profit margin, net profit operations, profit to sales ratio, return on investment. Respondents were firstly asked to rate how important each dimension is for them and then to rate their satisfaction regarding their firm performance against each of the dimensions. Finally, the importance and satisfactions scores were multiplied in order to determine the weighted average performance score (Sirén, Kohtamäki, & Kuckertz, 2012). Performance is a reflective construct, following the analysis of the authors.

(37)

37

Table 4: Dependent variable: Profit Performance (Source: Sirén, Kohtamäki, & Kuckertz 2012) No Item(s) PER_1 PER_2 PER_3 PER_4 PER_5 PER_6 PERF_1 PERF_2 PERF_3 PERF_4 PERF_5 PERF_6

How satisfied are you with your firm’s performance against each of the following financial performance criteria?

 Cash flow

 Return on shareholder’s equity

 Gross profit margin

 Net profit from operations

 Profit to sales ratio

 Return on investment

How important is the measure in terms of your firm performance?

 Cash flow

 Return on shareholder’s equity

 Gross profit margin

 Net profit from operations

 Profit to sales ratio

 Return on investment

Moderating variables - Entrepreneur’s experience& Environmental dynamism

Entrepreneur’s experience is measured using the items developed by Wiklund & Shepherd,

(2003): startup experience, management experience (more than one year) and management experience of working in rapidly growing organizations (annual sales growth of at least 20%). For each of the items, if the responded had previous experience, it was coded with one. Consequently, if the responded did not have previous experience, it was coded with zero.

In what concerns the startup experience, 61.4% from the respondents started at least another startup before while 58.6% had management experience of at least one year. The smallest

(38)

38

percentage was registered by the ones who had management experience in a rapidly growing organization, accounting for 41.4%.

In order to measure environmental dynamism, I used the items developed by Sirén, Kohtamäki, &Kuckertz (2012). The survey asks the respondents to evaluate on a five point Likert scale the business environment according to several statements concerning: product demand, customer preferences and industry stability.

While environmnetal dynamism is a reflective measure, entrepreneur’s experience is a formative

one since it measures totally different aspects regarding the past experience of the entrepreneur.

Table 5: Moderating variables Moderating variable No Item(s) Sources Entrepreneur’s experience ENE_1 ENE_2 ENE_3

In which of the following areas do you have previous experience?

 I have started another startup before.

 I have worked as a manager in another organization or business for more than one year.

 I have worked as a manager in a rapidly growing organization (annual sales growth of at least 20%).

Wiklund & Shepherd, (2003) Environmnetal dynamism END_1 END_2 END_3

 Product demand is hard to forecast.

 Customer requirements and preferences are hard to forecast

 My industry is very unstable with huge change resulting from major economic, technological, social or politic forces. Sirén, Kohtamäki, &Kuckertz (2012) Control variables

This paper controls for the following personal characteristics and background: age, level of education, hierarchical position in the startup, firm age, firm size (number of employees) and

(39)

39

number of founders. According to a study by Lafuente and Driga (2007) on entrepreneurial activities in Romania, the average age of people who have entrepreneurial activities is 34.4 years. One reason behind this average might be the fact that once people have more experience they tend to form networks and enhance their resources for future projects. Therefore, there seems to be a connection between the age and the performance of the startup. For the purpose of this study, age has been measured using an interval question with five possible answers.

In what concerns the levels of education, studies show that there is a link between education and the chance of becoming an entrepreneur as well as being successful when becoming self-employed (Robinson & Sexton, 1994). This particular study revealed that self-self-employed people have on average 14.5 years of education while salaried workers have on average on year less of education. In this study, the level of education is measured through a direct question with four possible answers (see Table 5).

The hierarchical position in the company (being a founder or co-founder) as well as the number of founders is important because it might influence the performance and growth of the startup. Studies in this area contradict one another. Some researchers found a positive effect between the number of founders and growth (Cooper & Bruno, 1977; Eisenhardt & Schoonhoven, 1996) while others found no relationship between the two (Brüderl, Preisendörfer, & Ziegler, 1992; Almus, Nerlinger, & Steil, 1990).

Given the fact that this study focuses on young companies, the firm age is defined as the time elapsed since the first e-commerce activity of a particular startup. For measuring the firm size, the number of employees was used as a measure. Studies show that firm size can be related to

(40)

40

the development of the company in aspects related to profitability, productivity, survival and innovation (Beck, Demirgüç‐Kunt, & Maksimovic, 2005; Rogers, 2004).

5. Analysis

The following section elaborates on the results of this research. Firstly, the measurement model will be analyzed and then the relationships between variables will be tested. For the measurement model analysis, the reliability and validity of the construct measures will be analyzed. Also, a distinction must be made between reflective and formative measurement models since the two approaches base their fundaments on different concepts. Therefore, for the reflective measurement models the following analysis will be provided: internal consistency, indicator reliability, convergent validity and discriminant validity. In the case of formative measurement models the following analysis will be conducted: convergent validity, collinearity among indicators and significance and relevance of outer weights.

5.1 Evaluation of the measurement models

5.1.1 Reflective measurement models

There are four reflective measurement models in this analysis: business model innovation, performance, environmental dynamism and the dimensions corresponding to learning. In order to test for internal consistency, the Cronbach’s alpha will be used. There were several items which

did not exceed the threshold of 0.7 and, therefore, those item were dropped since these are reflective measurement models and it means that the item are interchangeable (Hair Jr., Hult, Ringle, & Sarstedt, 2013). The items dropped are: SKI_1, SKI_3, SKII_1, BMI_1, PERF_1, PERF_2, PERF_6 and END_3. The final results of the items are presented in Table 6.

(41)

41 Table 6: Outer loadings

Variable Item Outer loadings

Learning SKD_1 SKD_3 SKD_4 SKD_5 0.715 0.842 0.889 0.838 SKI_2 SKI_4 SKI_5 0.724 0.729 0.827 SKII_2 SKII_3 SKII_4 0.829 0.837 0.794

Business model innovation BMI_2

BMI_3 BMI_4 0.767 0.714 0.756 Performance PERF_3 PERF_4 PERF_5 0.814 0.913 0.922

Environmental dynamism END_1

END_2

0.942 0.907

As a further step, the composite reliability values will be evaluated.. Composite reliability values above 0.7 are considered to be acceptable (Hair Jr., Hult, Ringle, & Sarstedt, 2013; Nunally & Bernstein, 1994).The composite reliability values of the constructs analyzed are all above 0.7, which demonstrates a high level of internal consistency reliability, as shown in Table 7.

In order to measure if the items of a construct converge and share a high portion of variance, convergent validity is used. This is particularly important because the indicators of a reflective construct basically use different approaches to measure the same construct. Therefore, the items should share a high portion of variance (Hair Jr., Hult, Ringle, & Sarstedt, 2013). The AVE