The innovation process of crowdfunded start-ups Master Thesis MSC BA Small Business and Entrepreneurship MSC BA Strategic Innovation Management University of Groningen Faculty of Economics and Business Martijn Gaasbeek

(1)

1 _{Metaallaan 89, 9743BP Groningen, m.y.gaasbeek@student.rug.nl, s2176793} 2 _{Appendices and references not included}

The innovation process of crowdfunded start-ups

Master Thesis

MSC BA Small Business and Entrepreneurship MSC BA Strategic Innovation Management

University of Groningen

Faculty of Economics and Business

Martijn Gaasbeek1

June, 2017

Supervisor 1: dr. O. Belousova

Supervisor 2: dr. F. Noseleit

Word Count2_{: 13 166}

Keywords: Innovation process, crowdfunding, start-ups, financing method, multiple case

(2)

2

Abstract

The aim of this research is to answer the research question on how the innovation process of start-ups is influenced by the use of crowdfunding. From the literature about the innovation process models and the influence of the financing method on the innovation process, certain dimensions are derived. To answer the research question, multiple case studies are performed with semi-structured interviews. Seven cases from Canada and the Netherlands, which used pre-ordering crowdfunding, are included in this research. The analysis showed that the innovation process can be influenced in the following dimensions: the start and order, the openness, the degree of parallel activities, the speed, the access to additional resources, and the risks. The results of this study are several propositions which could be used in an academic and managerial context.

Acknowledgements:

(3)

3

1. INTRODUCTION

An important factor for innovation is the ability to obtain early-stage funding. For many entrepreneurs, this is a main obstacle to innovate (Cosh, Cumming, & Hughes; 2009). In the past years, crowdfunding has become an important way of raising capital for entrepreneurs that are seeking external financing. Empirical evidence suggests an impressive grow in the amount of external financing that is raised by crowdfunding (Belleflamme, Lambert, & Schwienbacher, 2014).

In the past literature field of external funding, research has been conducted on the influence of different financing activities and the influence on new ventures. Especially research on the influence of venture capitalists (Sapienza, 1992; Gomez-Meija et al., 1990; Alemany & Marti, 2005) and business angels (Sørheim, 2005; Politis, 2008). Little research has been done on the impact of crowdfunding on new ventures. Therefore, the purpose of this study is to examine how innovations are influenced by the consequences of using crowdfunding to obtain external financing. Innovation is not limited to the introduction of new inventions in a new business, but can also occur by large enterprises which introduce new products/processes/business models (Vaona & Pianta, 2008). However, for this research innovation will be investigated in an entrepreneurial context. This is mainly because of the interesting mix of entrepreneurship and innovation that can be found on these crowdfunding platforms.

Mollick (2014) stated that even basic academic knowledge about the dynamics of crowdfunding is lacking. Crowdfunding has a high potential to be a disruptive change in the financing world (Mollick, 2014). Therefore, research in this area can contribute to the existing literature by opening the black box of how the innovation process is influenced by crowdfunding.

Research on the influence of crowdfunding on the innovation process can be interesting for managers or entrepreneurs to understand the influence of crowdfunding on their innovation process and learn from previous cases. If the impact is better understood within the context of start-ups, managers can act upon the consequences and exploit the potential benefits of crowdfunding and act upon the potential threats.

1.1 Research question

(6)

6

RQ: What are the dimensions of the crowdfunded innovation process and what is the influence of crowdfunding on the innovation process?

1.2 Introduction to Crowdfunding

Schwienbacher & Larralde (2010) stated that crowdfunding has its roots in concepts like micro finance (Mollick, 2014) and crowdsourcing. Crowdsourcing can be described as: "the act of taking a job traditionally performed by a designated agent (usually an employee) and outsourcing it to an undefined, generally large group of people in the form of an open call" (Piller, Reichwald, &Ihl, 2007 p.87). Kleemann, Voß, & Rieder (2008) were one of the first academics to write about this ‘working consumers’.

One of the main starters of crowdfunding, instead of crowdsourcing, was Sellaband.com, which made a platform for fans to invest in their favourite bands (Schwienbacher & Larralde, 2010). This concept has expanded to other industries: journalism, beer, software, and fashion, in the years after Sellaband.com (Belleflamme, Lambert, & Schwienbacher, 2014). One explanation for the growth of this concept can be the high rejection rates of banks and venture capitalists for small, innovative ventures (Cosh et al., 2009). Another reason is the technological innovation of the internet, which made it possible to have a higher integration with the consumers (Brabham, 2008). Since then several platforms have emerged, for example the platform that is used for this study, Kickstarter (Belleflamme et al, 2010). Furthermore, crowdfunding removes the geographically limitations that usually are present in traditional finance methods (Mollick, 2014). The possibility to externally validate the project is also an important factor in crowdfunding, as ideas are ‘tested’ in the market (Mollick, 2014). Another interesting finding is that entrepreneurs had several reasons to use crowdfunding, not only raising money, but they also used it to increase the legitimacy of the innovation (Schwienbacher & Larralde, 2010). Crowdfunding can also be used as a kind of marketing form to promote the innovative product (Schwienbacher & Larralde, 2010). Lastly, another important reason for the big growth of crowdfunding is the legislation of equity crowdfunding in the United States (Espositi, 2012).

(7)

7

As already mentioned in the definition of Lambert and Schwienbacher (2010), there are different variations of crowdfunding: donations, rewards, and voting rights. A fourth variation can be added, which is providing small loans to start-ups in exchange for interest (Scholz, 2015).

Donation based crowdfunding is often found in projects which have social/environmental goals, for example charity organizations. This kind of crowdfunding is based on the crowd providing money without expecting tangible benefits or rewards. The reward-based crowdfunding can be divided in donations and pre-order of products. For donation-based crowdfunding, the investor gets a small reward compared to the investment. For pre-ordering crowdfunding, the investor gets the possibility to already buy the product in exchange for funding. The third variation is the voting rights, which means that the start-up is selling shares in exchange for funding. Lastly, the micro loans are for charity and non-charity organizations and often include an interest payment to the investors (Scholz, 2015).

Each variation has its own potential benefit for the fundraise and the investor (Lambert and Schwienbacher, 2010). For this research, I will focus on the reward-based crowdfunding, specifically on the pre-order of products, since this is the most prevalent approach and most innovative projects seek funding via this variation (Scholz, 2015). The study of Lambert and Schwienbacher (2010) showed that 80% of the fundraising initiatives offer a kind of reward for the ones who invested. The research that has been done on this variation of crowdfunding already proved the effectiveness of reward-based crowdfunding. According to research of Mollick and Kuppuswamy (2014), 90% of the successful projects became an ongoing venture and a third of them reported annual revenues of more than $100 000.

(8)

8

2. THEORETICAL FRAMEWORK

To gain a better understanding of the innovation process of crowdfunding, it is first important to look at the dynamics of the innovation process. I will further discuss the different generations and models of the innovation process within firms and proceed to the innovation process of open innovation, as it is expected to be most closely related to the innovation process of crowdfunding. Since the theory of open innovation in an entrepreneurial setting is lacking, I will base my discussion on the open innovation process of small to medium-sized enterprises literature. Furthermore, I will conclude by introducing the influence of the financing method on the innovation and the expected influence of crowdfunding on the innovation process.

2.1 Process of innovation

According to Garcia & Calantone (2002, p112), innovation can be defined as: “an iterative process initiated by the perception of a new market and/or new service opportunity for a technology-based invention which leads to the development, production, and marketing tasks striving for the commercial success of the invention.” Important aspects of this definition are that innovation is an iterative process, so it includes the introduction of new innovation as well as the introduction of improved innovations, and that innovation always includes the introduction to the market of the invention.

Schumpeter (1913) was one of the first to perform academic research to ‘innovation’. He made a proposed distinction between five types of innovation: new products that are introduced, new processes that are introduced, new markets that are created, new sources that are developed for raw materials or other inputs, and the creation of new market structures. There are different classifications of innovations. First, classifications based on the nature of what is innovated. This can be products, processes, or business models (Crossan & Aoaydin, 2010). Second, classifications based on how it is innovated. It can be radical (new technologies that results in new markets) or incremental (products that are improved by new technologies) (Garcia & Calantone, 2002). Lastly, classifications based on the consequence of the innovation. This could be disruptive or sustaining (Boons & Lüdeke-Freund, 2013).

(9)

9 2.1.1 Six generations of innovation

Rothwell (1992) created five different generations of innovation, Preez and Louw (2008) added one generation. The first generation is the technology push model and appeared in the literature in the 1960s. It was created because of the technological development that created new products for the market (Hildrum, 2007). It is a linear process which starts with a new discovery in the science and ends with the sale of a new product.

The second generation is the market pull model, which is also a linear model. The difference is that this process starts with a market need, instead of a new scientific discovery (Jaruzelski, Dehoff, & Bordia; 2006).

In the 1970s, the interactive model was created which is the third generation. This model is a combination of the market pull and technology push models. The main aspect in this model is that there is a certain level of feedback from the market. The scientific discovery as well as the market needs is the impulse for the innovation process (Jaruzelski et al., 2006).

The fourth generation is called the integrated business processes model. It is characterized by a parallel use of research, customers and suppliers. This is the first model where instead of sequenced steps in the process a parallel process is used (Hildrum, 2007).

The last generation of Rothwell is the system integration and networking model. It is an addition to the previous generation by including the use of IT methods in the innovation process. It also includes the horizontal and vertical linkages between the innovator, customer, supplier and other stakeholders via strategic alliances (Jaruzelski et al., 2006).

The sixth generation is created by Preez and Louw (2008). This generation is called the open innovation model. Instead of being only focused on the inside idea generation, this model also includes the external ideas into the process model of innovation. The innovator can use external paths to the market and externally created ideas into the development of an innovation. I will discuss this generation in more detail later in this paper.

(10)

10

Open innovation (6th generation of innovation processes) includes leading edge customers (Rothwell, 1992) and therefore I expect that this process model is the closest related to crowdfunding. I will analyse this process model in more detail in section 2.1.3.

2.1.2 Five innovation process models

Apart from the generations that were developed over the years, Salerno, de Vasconcelos Gomes, da Silva, Bagno, and Freitas (2015) made an overview of the different processes that are used for innovation. This can be categorized in five different innovation process models. I will discuss the different innovations processes in order of appearance.

Traditional process

The traditional process is a combination of the first and second generation of the innovation process discussed earlier. The sales occur after the production of the innovation and this process is typically used by large companies which have a very structured process for their innovation. This process starts with idea generation, which includes encouraging idea generation and registration of the ideas that are developed. After that there is a screening and valuation, which is followed up by the development of the innovation and ends with the diffusion, marketing and sales of the newly produced innovation (Salerno et al., 2015).

Process started by a call

By this innovation process, the process starts with a call from an organization. The call is a public or private request with certain functional requirements of the end-product. At the start of the innovation process, the call is evaluated and the feasibility is assessed. When the decision is made to compete for the call, resources are often consumed to get a better chance of winning the call. When a company wins the call, the development and production process is started. The last stage is the delivery of the product to the client (Salerno et al., 2015).

Process with parallel activities

(11)

11 Process with a stoppage

In this innovation process, there is a stoppage in the process, because of the market, the technology or both. A stoppage which is related to the market is when a firm realised that the current market is not big enough to scale up to production, therefore there is an active stoppage. During this stoppage, the organization is investing in increasing the size of the market by marketing or by searching for another lead user. When there is a stoppage that is related to the technology, the firm faces a technological bottleneck in the production of the product. During this stoppage, the firm seeks to improve the technology or search for other technologies that could counter the problem. Sometimes this bottleneck arises from a dependency on a third party or technology. In both these stoppages, the development phase in interrupted or the development is downsized to an acceptable amount of risk (Salerno et al., 2015).

Anticipating sales

The last innovation process, that is described in the research of Salerno et al. (2015), is the anticipating sales. It could be a call from the client to a specific firm or a general call seeking producers. The sale of the product is before the development and the production of the product. In this way, the client is financing the development and production of the process and the product is tailored-maid to the needs and expectations of the client. The client starts the process of the innovation instead of ending it and there is no real idea generating phase of the innovation process (Salerno et al., 2015).

2.1.3 The process of open innovation

As already discussed open innovation could be a good starting point for an innovation process model of crowdfunding, since it will probably have the most overlap with crowdfunded innovation processes. This is primarily because of the involvement of stakeholders in the innovation process and the possibility to validate the innovation externally instead of solely internal.

(12)

12

One of the major examples of open innovation is open source software, which is defined as an open development process of software by independent programmers (Gassman & Enkel, 2004). Several studies already stressed out the importance of early user integration in the development process. In the 21st century, costumers are not merely customers anymore, but are playing a more active role (Prahalad & Ramaswamy, 2000). According to Grönlund, Sjödin, and Frishammer (2010), companies are moving away from the closed innovation process where development and marketing of the new products is within the firm. The same research also suggests that companies need to stop thinking internally and move away from the standard innovation processes like the stage-gate model.

Open innovation can be distinguished into three different processes: the outside-in, the inside-out and the coupled process, which is a combination of both (Gassman & Enkel, 2004). For this research, the focus will be on the outside-in process. The main reason is that the outside-in process is the most related to crowdfunding, since it is focusing on collaboration with costumers and integrating external knowledge. The outside-in process is mostly used in low-tech industries and is used to be dominated by small to medium sized companies. They acted as a knowledge broker in the markets. Modular products and a high knowledge intensity also have a positive influence on the use of the outside-in process (Gassman & Enkel, 2004). For open innovation, firms can focus on science based or market based relationships. For this research, the focus will be on market-based relationship, since the main focus is the relationship with the customers (Du, Leten, & Vanhaverbeke, 2014).

(13)

13

The main motives for SMEs to engage in open innovation were related to the market-related targets. Open innovation was used to serve customers more effectively and another motive was to open up new markets for the firm. Knowledge acquisition is also mentioned in their results as one of the motives for SMEs. It also eliminates the likelihood of product failure and it is argued to be the single most important information necessary for new product development (Du et al., 2014).

Another important aspect of the open innovation process model is the evaluation of ideas outside the firm. The product or service can be presented to the users and the firm receives feedback on the needs and desires of the customers. In that way, the firm understands the requirements of the user better and is able to adapt to these requirements. Therefore, instead of stage-gate models, companies move towards a probe and learn process. In this process, early versions of the product are introduced and tested in the markets (Gassmann, Enkel, & Chesbrough, 2010).

Open innovation for SMEs also has some challenges. According to Brunswicker and Vanhaverbeke (2014), it requires certain management skills to integrate the inflow of information with the existing project, to apply knowledge from external and internal sources, and, lastly, to give direction to the innovation actions. Other challenges could be the Not-invented-here (NIH) syndrome, the increase in complexity to manage the innovation. Lastly, possible opportunism could make the relation between the firm and their stakeholders more difficult (Van de Vrande et al, 2009).

(14)

14

score low on feasibility. Furthermore, a formal management team with a diversified managerial capacity is often needed for successful open innovation management. (Wynarczyk, 2013) For high-tech innovations, the openness of the innovation process is likely to increase in the latter stages. The idea generation and the idea development phase are often closed, and the commercialization phase is more open in the process. For low-tech industries, this is not the case (Dries, Pascucci, Török, & Tóth, 2014).

2.1.4 Dimensions of the innovation process

To summarize the literature field of innovation process, there are certain important dimensions that are shown in each model. These dimensions are:

- The degree of parallel activities/stoppage in the innovation process - If the steps are sequential or tasks are overlapping

- The degree of openness – Placement of stakeholders within the process - The start of the process – If the process starts with technology or with demand

These dimensions are important in creating a process model for innovations that are funded by crowdfunding. However, the current process models of innovation are expected to fail in explaining the influence of crowdfunding on the innovation process model, since the involvement of the crowd cannot be found in these process models.

(15)

15

2.2 Financing methods

In the previous chapter, I discussed the innovation process. However, the way the innovation is financed also has certain influences on the innovation process (Collewaert & Sapienza, 2014). Therefore, it is important to analyse the influence of other financing methods to isolate the impact of crowdfunding on the innovation process. I will start this section with an analysis of the financing methods that could be substituted by crowdfunding.

There is a common belief that new ventures rely heavily on equity funding, private equity by the entrepreneur or equity from the venture capitalists. However, Robb and Robinson (2012) found in their study that debt financing is the most used way to finance a new venture. There are three forms of debt financing: personal debt of the owner (usually backed by other assets of the owner, for example their home), business bank loans, and business credit lines.

In this section, the focus will be on debt and equity financing, since crowdfunding will be an alternative for these kinds of investments instead of private financing methods. The main methods for external financing are venture capital and business angels. These forms of financing provide more than just financing. They also provide assistance on business design, access to their network, and supply of personnel (Freeman & Engel, 2007). There are different roles for business angels and venture capitalists. Examples of these roles are sounding boards, mentoring the entrepreneur, assistance in identifying and acquiring valuable resources, external gatekeepers in the innovation process (Collewaert & Sapienza, 2014).

2.2.1 Influence of financing method on the innovation process

There are certain consequences that derive from the form of financing. This can be categorized in different influences: the access to additional resources, the speed of the innovation process, the level or formalization, and the risk that is associated with the innovation process. However, it is not possible to completely isolate the influences and these might be interrelated. I will discuss these influences below for business angels and venture capitalists.

Access to additional resources

(16)

16

the innovation process of the new venture. Another important aspect is the screening and signalling function of venture capitalists (Florida & Kenney, 1987). By collaborating with a well-respected venture capitalist, the firms could obtain a certain positive image.

Speed of the innovation process

As already mentioned in the additional resources part, the venture capitalists or business angels could provide resources that can speed us the process. However, there is also a delay in the innovation process to obtain these funds. Manso (2011) argued that the experimentation phase is also dependent on the way of financing that is used. In more formal finance methods, venture capitalists and business angels, the experimentation phase is usually longer than with more informal methods, personal financing, debt financing. Manso (2011) stated that the speed of the innovation process is declined by external investors.

Venture capitalists need more screening ex ante, because they are representing other investors and business angels only invest their own money. Therefore, the innovation process before the actual financing is expected to be longer in the case of venture capitalists than in the case of business angels (Van Osnabrugge, 2000). Venture capitalists usually invest in a later stage in the process and they take longer to invest. Business angels take on average around fifteen weeks to invest, compared to the average of venture capitalist of twenty weeks (Van Osnabrugge, 2000).

The level of formalization

Business angels often introduce more formalized processes in the new venture, including regular reports, operational progress points and strict accounting procedures. They introduce formal monitoring in the new venture, to be able to assess the performance of the innovation. Therefore, it is expecting to increase the formalization of the innovation process into certain measurable steps. Furthermore, they need to specify roles and responsibilities in the contract (Fili & Grünberg, 2016). The same counts for venture capitalists, since they require formalized business plans to justify their investment to the investors of the venture capital (Van Osnabrugge, 2000).

(17)

17 The risk of the innovation process

The last dimension is the risk of the innovation process. This category can be a result from the other influences. However, there are some specific factors for different types of finance methods that I would like to discuss. As described by Manso (2011) the failure rates are higher for personal financed projects than for projects that are financed by venture capitalists or business angels. Another aspect is the perceived risk from stakeholders. When a well-respected venture capitalist invests in the entrepreneur, the entrepreneur is perceived to be more promising and less risky (Florida & Kenney, 1987).

However, the risk-tolerance of the financing method has a positive relationship with the innovation output of the firm (Tian and Wang, 2014). Since venture capitalists and business angels have a lower risk-tolerance than personal financed projects, the innovation output is also likely to be lower. Although it is expected that the influence on the process is less in the case of business angels than venture capitalists (Fili & Grünberg, 2016).

The results of the literature review are summarized in table 1.

Venture Capitalist Business Angel Other

Access to additional

resources

Network, external parties

Own experience of the business angels

None

Speed of the process Slowed down the

innovation process

Slowed down the innovation process, but less than VC

Faster implementation and innovation process

Level of formalization Highly formalized Highly formalized Less formalization

Risk Less risk but also

less risk-tolerance

Less risk but also less risk-tolerance

High failure rate, but more risk-tolerance Table 1Influence of financing method on innovation

(18)

18

2.3 Influence of crowdfunding on the innovation process

From the literature review, certain dimensions of the innovation process and influences from the financing method are derived. The speed of the process is both a dimension and an influence. For the dimension, I will solely discuss whether there is a stop in the innovation process. The influence of crowdfunding on the speed of the innovation process will be about whether the speed will be increased or decreased during all the phases. I will discuss these dimensions and influences below.

2.3.1 Dimensions of crowdfunded innovation process Start of the process

Although the early phases of the innovation process cannot be influenced by crowdfunding because these phases happened before the crowdfunding project started, there are possibly some effects that can be noticed by the screening of crowdfunding and the rejection of other ways of financing. As described by Bruton, Khavul, Siegel, and Wright (2014), high risk entrepreneurial opportunities are less likely to be financed by traditional forms of financing. Another important argumentation in their research was that a lot investors refuse to invest in an option when it is too low tech. As a result, these projects may fulfil their financing need by using crowdfunding. Also, the easiness to access crowdfunding could influence the decision of the inventor the launch the product themselves, instead of selling it to another firm. Everyone has access to the crowdfunding platforms and it is much easier than finding a business angel to support your product. This is especially the case when the potential of the innovation cannot be successfully signalled to potential investors. The snowball effect is used for this phenomenon in the literature (Scholz, 2015). This implies that the crowdfunding innovation process is likely to start by a low-tech high-risk invention with entrepreneurial opportunities.

The degree of parallel activities/stoppage in the innovation process

(19)

19

In some of the different models for innovation process, a certain stop for the market or stop for the technology can be found. In the case of crowdfunding, a certain stop for funding can be expected. As already stressed out earlier, the marketing and signalling activities during the funding round are important for the success of the crowdfunding project. This could take all the available time of the entrepreneur, leaving no time to actual develop the innovation. Therefore, there is expected to be a certain stop in the actual development of the innovation.

The degree of openness

As already mentioned before, one of the main reasons for entrepreneurs to engage in crowdfunding is to interact with the customer in a market-based relationship. Therefore, there is expected to be an open innovation model and regular interactions with costumers. As found in the open innovation literature, this is expected to increase the performance of the innovation. Another result that is found in the open innovation literature is that these market-based relationships should be formally managed. Lastly, to succeed in the open innovation process, a formal and high diverse management team is necessary.

Another important aspect of the degree of openness is the placement of the customers within the innovation process. The interactions with the customer are early in the innovation process, before the actual production of the product. This is relatively new in the innovation process model, since other models only include the customer at the very beginning or at the very end. The interaction with the customer is also expected to be used in the whole process, as entrepreneurs try to move the customers to their own website and use the network as a sounding board for later-stage improvements. (Scholz, 2015)

2.3.2 Influences of crowdfunding on the innovation process The speed of the process

The innovation process can be shortened by gaining valuable knowledge from the customers and the use of crowdfunding could create a certain hype around the innovation which increases the diffusion.

(20)

20

used as a source of knowledge to solve potential hick-ups in the innovation process. (Lehner, Grabmann, & Ennsgraber, 2015) As described by Scholz (2015), in some cases the introduction of the innovation on crowdfunding platforms could create a moment momentum to attract a very large diverse group of early adopters.

The funding process of venture capitalists and business angels take on average respectively 20 weeks and 15 weeks (Van Osnabrugge, 2000). This is reduced by the use of crowdfunding since the average campaign is expected to take a shorter time period (Scholz, 2015). Therefore, an increase in the speed of the process is expected, solely based on the influence of the financing method.

As discussed in the previous chapter, venture capitalists and business angels don’t provide solely the financing for the firm but also external knowledge and experience. This might be lacking in the case of funding through crowdfunding. Therefore, there is expected that the degree openness would compensate for this missing knowledge. The entrepreneur could use the ‘crowd’ for information and knowledge about certain subjects. As described by Lehner et al. (2015), the crowd could also be uses as an omnipotent source-potential.

Level of formalization

The level of formalization is expected to be lower in the case of crowdfunding than for firms backed by venture capitalists and business angels. One of the reasons is that the pressure on the entrepreneur to leave the firm is lacking and there is no need for external reporting to an investor other than the customer.

Risk

The risk is expected to be lower in the case of crowdfunding, since the idea can relatively cheap be tested in the market by using pre-ordering as a crowdfunding method. When the target goal is not met, the firms often don’t have the obligation to start producing (Scholz, 2015). Another aspect is the easiness of signalling to other potential investors. Firms can already show the potential of the product by the number of products they already sold and by that more easily attract extra external funding from other sources.

(21)

21

3. METHODOLOGY

Within this section, the design of the study will be described and the steps that are taken in the research will be given. Furthermore, the controllability, reliability, and validity will be analysed based on academic literature. Lastly, the analyses of the data will be discussed.

3.1 Research method

As shown by the literature review, the literature field is still very scattered and there is not a lot of research done to the influence of crowdfunding on innovation. This paper is aimed at exploration of the process of crowdfunded start-ups. Therefore, this research has the purpose of theory development and is based on qualitative data, obtained by a case study on multiple cases. The method of Yin (2003) is relevant for researches that try to answer ‘what’, ‘how’, and ’why’ questions, where the researcher cannot control the environment, and the researched is focused on a real-life context. Since all three factors apply to this study, the method of Yin (2003) will be used. By using multiple cases, interesting patterns and confirm or disconfirm between the cases can be analysed (Yin, 2003).

3.2 Data collection

This research will use primary data collected from several different cases. The main source of data will be in-depth interviews with innovative entrepreneurs who used crowdfunding to obtain the necessary financial resources. Through these in-depth interviews, I hope to gain an insight in the influence of crowdfunding on the innovation process. Eisenhardt (1989) stated that for multiple case studies a number between the four and ten cases should be used. Therefore, the aim of this research was to be within these boundaries. In the end, I succeeded to gain seven different cases.

(22)

22

The period in which I interviewed the entrepreneurs was six weeks and the interviews were performed face-to-face or via skype. The interviews took on average 45 minutes. The cases that were used in this research are based on Canadian and Dutch Start-ups. The interview questions can be found in Appendix A. Three of the cases were successful in reaching their funding goal and four were unsuccessful in reaching their funding goal. However, three of these unsuccessful cases are still in business and trying to find other ways to the market. An overview of the cases can be found in table 2.

Size of entrepreneurial team Target for crowdfunding Money raised through crowdfunding Pre-investment Use of protection mechanisms Number of employees currently working for the firm

Slide 2 €40 000 €201 560 Angel

investment

Patents 2

Solar Cam 2 €417 290 €14 306 None Patents 0

The Next Selfiestick 3 €50 000 €3 069 None None 0 Point Controller 2 €250 000 €845 None Design patents 0 Matter and Form 3d scanner

3 €53 948 €319 185 None Design patent

afterwards

13

Bevel 3 €178 508 €270 700 None None 13

Ledbits 2 €20 000 €3 796 Grants Design patent 0

Table 2 overview of cases

3.3 Controllability, reliability and validity

Van Aken et al. (2012) defined several potential biases which can harm the reliability of the research: researcher, respondent, and the situation. First, I will discuss the bias and the potential risk to the research, then I will explain how I mitigated this risk to increase the reliability of this research.

(23)

23

number of researchers cannot be increased, therefore the latter is used in this research (Van Aken et al., 2012). This is also done by recording semi-structured interviews and re-hear it a second time, so the researcher can spot elements that were overlooked in the first analysis. This is also a way for the researcher to be more an observer during the interview. For the data collection and analysis, a standardized protocol to further reduce the potential researcher bias will be used.

There is a respondent bias when the respondent is influenced in a cognitive bias context in such a way that the response is not accurate of the completely the truth. The decrease the respondent bias, I also made use of documents of the cases that were available through the crowdfunding platform.

A situation bias occurs when the respondent is influenced by the currents situation and therefore the answers are not completely truth or influenced by this situation. To decrease the situation bias, the results of the within case analysis are send back to the entrepreneurs to check whether it is correct or not.

Another important aspect of research is the validity of the research. The validity of a research can be separated into three different categories: construct, internal, and external (Van Aken et al., 2012).

Construct validity is the degree to which the test actually measures what it should measure. (Cronbach & Meehl, 1956). To increase the construct validity, several sources of evidence will be used (triangulation) and the semi-structured interview questions will be assessed by the supervisors of this master thesis (Yin, 2003).

Internal validity is to which extend there is a cause effect relationship in the results of the study. (Van Aken et al., 2012) Internal validity will be increased by thematic analysing the results (Braun & Clarke, 2006). These process is described in table 3.

Step Description

Step 1 Transcribing the recorded interviews Step 2 Coding scheme development

Step 3 Within-case analysis of the cases Step 4 Cross-case analysis of the cases Step 5 Generalize the results

(24)

24

Lastly external validity refers to the degree by which the results holds true in other similar situations. This will be increased by making use of different innovations and different industries. By applying multiple industries, the results of this research have a higher chance of being applicable to different situations.

3.4 Data analyses

(25)

25

4. RESULTS

In this chapter, I will analyse the results of the study of the seven cases that were analysed in this thesis. First, I will start with the within case analysis with a description of the firms and analyses based on the categories that were found in the literature study. I will do this for one successful case and one unsuccessful case. The other within case analysis can be found in appendix B. Then I will show the results of the cross-case analysis that are also categorized based on the dimensions of the literature study.

4.1 Within case analysis

4.1.1 Slide

Innovation in Motion is a firm started in 2016 by Thijs Olthof and Kaj Beetstra by launching the product, Slide, on Kickstarter. Slide is the world’s first retrofit smart curtain system and is focused on the smart home market. Slide was a big success on the platform and reached around the 504% of their initial goal. Currently they are in the engineering phase and preparing to go to mass production in a few months.

Innovation process

The entrepreneurial team is a combination of a product developer and a business developer, who started working as a cooperating team in May 2016. The product developer, Kaj Beetstra, was already testing and developing the innovation. Between May 2016 and September 2016, they created a proof of concept and started to create a prototype with a 3d-printer. They decided to use crowdfunding because of the easiness of the road to the market and the possibility to validate the concept in an early stage. Another reason is to maintain the control of the firm. The founders also considered the business to business market, but due to the availability of crowdfunding and the difficulty of the need for early investment, they chose to start with the business to consumer market. From November 2016, they started preparing the crowdfunding.

(26)

26

This was a full-time job and therefore they paused the research and development of the innovation itself. This step is described as the marketing of the product. They developed, among other things, a movie, a website, and animations and they started to expose their idea to the outside world as a pre-campaign promotion. The same holds true during the crowdfunding campaign, when there was a huge need to manage the comments and exposure on several media platforms. Another aspect was the marketing and advertisement during the campaign, which needed to be monitored and managed.

“So, Kaj and I were from the first of November working 80 hours a week to prepare the crowdfunding-campaign. During the campaign, we were fulltime managing the project.” (T. Olthof, 2017)

After the campaign until now (May 2017), they are working on research and developing to make the product ready to produce. During the campaign, there were several requests for product features which needed to be included after the campaign and they proceed from a prototype to an actual ready-to-produce product.

Dimensions of crowdfunded innovation process

Start of the process

As already described, the innovation is a product of one of the entrepreneurs who co-founded the firm with a business developer. They stated in the interview that without crowdfunding, the chances that the invention made it to the market would be lower.

Degree of parallel activities

What can be seen in the innovation process of slide, is a certain stop for marketing or stop for funding. The research and development is separated into development before the crowdfunding and development after the crowdfunding. The creation of the prototype happened before the crowdfunding, afterwards the complete focus was on the marketing of the product. When the campaign ended, they went back to the research and development phase to adapt the product to the needs of the consumer and to make it ready for mass production. However, there are some parallel activities in the innovation process, during the crowdfunding campaign. They discovered some customer-requested features which needed to be researched during the campaign to adequately answer request from customers.

Degree of openness

(27)

27

validate the product and to gain insight in the needs of the customer. They performed several follow-up surveys to get a market overview and get useful consumer data. In this way, they didn’t just open up the innovation process during the crowdfunding, but they also stayed open after the campaign. They created a community of costumers to get more data. According to Olthof, it is important to not overload your customers with customer involvement programs. There is a group that is more than willing to help, but there is also a group that doesn’t want to be involved. Communication, however, is very important in every crowdfunding campaign, according to them. They included their customers during the development phase after the crowdfunding campaign to keep them up to date about the delays in the delivery and potential hiccups during the process. They are seen as co-creators in this process.

“You should not only see them as consumers but also as co-creators.” (T. Olthof, 2017)

Influence of crowdfunding on the innovation process

Speed of the process

Although the stoppage earlier described in the dimensions of crowdfunded innovation process, there was an increase in the overall speed of the innovation process. The entrepreneurs chose crowdfunding, because they wanted to speed up the process of the innovation instead of slowly growing the business in certain steps. The main reason for the increase in speed by the crowdfunding campaign is the instant upscaling of the business in terms of funds raised. They could attract enough funding to go to mass production in a few months, which created the possibilities to directly focus on mass production, instead of slowly growing the business. Because they could get enough funding in an early stage of the innovation process, they were able to attract external expertise into the innovation process to further develop the product. Furthermore, they created an instant group of people that could be used for their feedback and input in the innovation. With this community, they could create a better fit to the market earlier in the process and they could create a further developed first product. Another aspect is that they could easily align their product with the market via crowdfunding and they could create an overview of the market. This made it possible for them to quickly adapt the product to the needs of their customer base and to test which aspects of the product are most important for these customers.

(28)

28 Access to additional resources

The lack of additional resources from venture capitalists or business angels is compensated in several ways. The first one is the community of the founders. They searched for mentors and tutors around, who could guide them through the process of innovating the product and especially launching it into the market. Another way is the exposure to strategic partners by crowdfunding. They got offers from several potential partners, who were reached through crowdfunding. They created connections with several partners via this exposure, which were in different phases of the innovation process very useful. Lastly, instead of using a partner for market research, they used the crowd to validate their idea and get feedback on certain features and the pricing of the product.

The level of formalization in the process was low, according to the founders. There was not a model or a logic behind the innovation process. They developed the product and created the marketing more based on feeling and by advice of a mentor. However, they created an overview of deliverables for the crowdfunding campaign. After the crowdfunding campaign, they created more formal checkpoints and created a more process structure of the innovation.

Risk

(29)

29 4.1.2 Solar Cam

Winfinite8 is a cooperation of a Canadian-based and a Hong Kong-based firm. They launched their new product, the Solar Cam, on Kickstarter at the end of March 2017. They already had some experience in the security cameras market and they initiated this idea based on feedback of customers on their initial product. The Solar Cam is a truly wireless security camera which is powered by the sun. Unfortunately, they didn’t reach their goal on Kickstarter. However, they are confident that there is another way to the market and they are seeking this way at the time of this research.

Innovation process

As I already mentioned, the idea came from the feedback they got on their initial products on platforms like Amazon. They started with the ideation 2 years ago, and in the beginning it was primarily calculating the possibilities of solar powered security cameras. The most important factor in this research was finding a way to put the camera in stand-by with a quick wake-up time to start recording. They hold a patent for this technology. Another year and a half they spend on research and development of the hardware and the mobile application of the security camera. By the time they started to prepare for the crowdfunding campaign, the engineering sample was ready. In December, they started the marketing part of the innovation process. This was partially recording a movie and creating some other Kickstarter elements. According to Wright (2016), it was better if they prepared a year for the Kickstarter campaign instead of the three months they actually did.

(30)

30

Dimensions of crowdfunded innovation process

Start of the process

The possibility for this invention was realised based on previous experience in the market. The entrepreneurs came up with the idea themselves and are trying to bring it to market by themselves.

In this case, the parallel activities were not within the innovation process itself, but the parallel activities were with other business lines of the entrepreneurs. The entrepreneurs occurred high profits in other business lines, which increased the parallel activities with the innovation process of the Solar Cam. There was a stoppage in the process of the innovation for the Kickstarter campaign. They prepared for four months in which they created all the necessary material for the campaign.

Degree of openness

They were hoping to create a community with their backers to directly include their feedback into the hardware and specifically the app. Unfortunately, they didn’t come that far since the crowdfunding campaign failed. Another important factor in the openness of the innovation process of the Solar Cam is that they wanted to try a smaller production with Kickstarter backers who are more engaged than the average customers. The plan was to deliver to them first and get their feedback and make sure that there are no network challenges throughout different geographic regions and possibly some oversights. They tried to build almost, what they called, a social enterprise with a very high degree of customer involvement.

Influence of crowdfunding on the innovation process

Speed of the process

The feedback they got during the crowdfunding campaign speeded up the innovation process. They built a customer base from the backers who backed their product, so that they could, even though they were unsuccessful on Kickstarter, get help from the customer who backed them. That was very helpful in increasing the overall speed of the innovation process.

(31)

31 Access to additional resources

The crowdfunding campaign was unsuccessful; therefore, the entrepreneurs must find another way to the market. These other options are brought through the exposure of the firm on the platform. According to Wright (2017): “Some people want to support, some people want to support and the product, and some others want to invest”. So, we can conclude from this case, that the lack of resources provided by other investors is compensated by the support they got from the exposure of their idea on the crowdfunding platform.

The level of formalization in the innovation process was low. There were some written documents, but that was more a generation of ideas. The plans changed fast and they had to adapt to the real world. Examples are the changes in the social media platforms, which are booming at the moment. Due to the prior experience in the product line, the entrepreneurs were confident about the success of this innovation and therefore didn’t use formal validation points in the innovation process.

Risk

(32)

32

4.2 Cross-case analysis

The cross-case analysis compares the results of each case with each other to create an overview per dimension. The dimensions will be discussed in the same order that is used before. A complete overview of the cross-case analysis and codes can be found in table 4.

Dimensions/Influence Results

The start of the innovation process 7x Own invention

Degree of parallel activities 3x Marketing parallel to R&D 4x Stop for marketing

Degree of openness 6x Market validation 6x Feedback

3x Community building 3x Exposure

1x Finetuning of proposition 1x Identifying segments Speed of the process 4x Increase in speed

3x Didn’t affect the speed 5x Feedback increased the speed

1x Deadline forced us to make decisions Access to additional resources 4x Exposure to partners

2x Easier to get venture capital 1x Pre-crowdfunding mentor Formalization 5x Written business plan

4x Milestones

3x Experience instead of research 2x Formal market research

2x Major changes in business plan

Risk 5x High risk

(33)

33 4.2.1 General

What is noticed in all the cases is that the marketing and financing phase is between the rounds of research and development. There is minor R&D before the funding stage, which is primarily focused on creating a prototype or proof of concept, and major R&D after the funding stage. 4.2.2 Dimensions of crowdfunded innovation process

The start of the innovation process

The innovation process is directly influenced by using crowdfunding as a way of financing the innovation. All the projects were started by the inventor of the innovation self. In many cases the inventor stated that the innovation was made possible because of the existence of crowdfunding. Apart from the 3d-scanner and the Bevel, the innovations can be categorized as low-tech high-risk innovations.

The degree of parallel activities was low in the cases of this research. The expectation of research and development, marketing, and funding activities parallel to each other was found in three of the seven cases. In the other cases, there was a stop of research and development to perform the necessary marketing for the crowdfunding campaign and to perform the crowdfunding campaign. In these cases, the activities were performed sequential. However, each case without this stop was not successful in getting enough funding to reach their goal. Degree of openness

The innovation process can be opened up by crowdfunding by several ways. The one that by far is most mentioned during the interviews it the external market validation of the idea. Instead of assessing the idea yourself or via some protocol within the firm, the crowdfunding platform is used to validate the potential of the idea. Furthermore, customers are used for their direct input for desired features of the product. Crowdfunding is also a way to get data on the desires of the costumers and which features they want in the product. Another way that is mentioned in the interviews is creating a community for directly asking opinions or checking acceptability of a certain price of an option. The entrepreneurs create this community of the backers that backed the product of via potential customers that didn’t back the project but got interested and filled in their e-mail. Finally, the exposure that the companies derive from using a crowdfunding platform is also often mentioned as a positive consequence of using crowdfunding.

(34)

34

This contact is mostly formally managed via the crowdfunding website, data platforms like Backerkit or via e-mail. These communication with costumers is early in the innovation process instead of at the beginning or the end.

4.2.3 Influence of crowdfunding on the innovation process Speed of the process

In five of the seven cases, the speed of the innovation process is increased using crowdfunding. The main reason is the customer feedback that is obtained in the early phases of the innovation process. In the other two cases, there was not a noticeable influence on the speed of the innovation process.

Apart from the information and knowledge that was derived from the crowd, crowdfunding also brought in four of the seven cases possibilities for partnerships. By the exposure on the platforms, the entrepreneurs were contacted by potential partners for funding of other necessary resources. In two of the seven cases, crowdfunding made it easier to access venture capital to gain additional resources. One of the seven companies actually searched for strategic partnership before the crowdfunding campaign, to compensate for the lack of mentoring of other financing methods. The ‘wisdom of the crowd’ is only used to gather information about the preferences and needs of the customers and is not used to compensate for external partnering.

Formalization

The level of formalization was moderate/low in every case. In five out of the seven cases, there were some written business plans. However, two of them mentioned that they had to change the plans often because of the changing environment. In four cases, there were certain milestones or checkpoints to assess to potential of the innovation. In three cases, the entrepreneurs where certain of their idea because of the experience they had, therefore formal market research and assessment wasn’t necessary.

Risk

(35)

35

(36)

36

5. DISCUSSION

In this section, I will interpret the results of the research and explain any new findings or understandings of this subject. The literature of the theoretical framework will be compared to the results of the case analysis and propositions for future research will be proposed.

5.1 Innovation process

From the results of the cases, a new innovation model can be generated. This new model of the process of crowdfunded innovation can be seen as an extra generation to the model of Rothwell (1992). The new innovation model is displayed in figure 1.

Figure 1Innovation process of crowdfunded start-ups

The major difference with previous generations of innovation models is the place of marketing and funding, which is between prototyping and research and development. That is also the point where customers are included in the process, where in previous models the customers were only included at the very beginning or the very end of the innovation process (Rothwell, 1992). The innovation process of crowdfunding can be described as an open innovation according to the definition of Chesbrough “the use of purposive inflows and outflows of knowledge to accelerate internal innovation, and expand the markets for external use of innovation, respectively.” (Chesbrough et al, 2006 p. 1) Since the major difference of the innovation process is the placement of marketing and funding, the following proposition could be proposed:

P1. Marketing of the product is placed between prototyping and research and development in the innovation process of crowdfunded start-ups.

5.2 Dimensions of crowdfunded innovation process

5.2.1 Start of the process

In all the cases, the entrepreneur was also the inventor of the innovation. Therefore, the expectation is that crowdfunding is a way for an inventor to be an entrepreneur. This is also backed by statements of innovators who said that without crowdfunding the chance that they would be able to launch the innovation will be lower. This is also expected based on the literature about the snowball-effect (Scholz, 2015), where the potential of the innovation cannot be properly shown to potential other investors.

Idea

(37)

37

“So would the innovation be there without the possibility to crowdfund? I tent to answer that question with a no, because I would have guessed that it would fail in the search for capital” (T. Olthof, 2017)

P2. Innovations that are funded with crowdfunded can be expected to be launched by the innovator themselves.

5.2.2 Parallel activities

Based on the theoretical framework, there was expected to be a high degree of parallel activities, especially focused on the R&D, marketing, and funding. However, the entrepreneurs who used a high degree of parallel activities were not successful in raising enough funding. The successful projects all used a certain stop for funding. One possible explanation for this result could be the intensity and professionalisation of the crowdfunding platform. These platforms are getting more and more professional and this increases the time and effort that is needed to be active on these platforms. This stop could be found in other innovation process models in the literature of Salerno et al (2015) for other activities. In this research, stops for the market and technology are described. This stop for funding can be compared to these stoppages in the innovation process. However, this is an active stop where there are other activities performed instead of a passive stop where the innovator has to wait for external factors. Based on the success rate of the innovation processes with a stoppage for funding, the following proposition could be made: P3. Innovation processes with an active stoppage for marketing/funding are more successful than innovation processes where these activities are performed in parallel.

5.2.3 Openness

(38)

38

P4. Crowdfunding opens the innovation process, through external market validation and direct customer feedback.

Within the scope of this research, I cannot make a statement about the possible connection between the degree of openness and the success of the firm, since not enough data is available on the degree of openness of the innovations.

5. 3 Influence of crowdfunding on the innovation process

5.3.1 Speed of the process

As shown in the results section, crowdfunding has the potential to increase the speed of the innovation process. Even though there might be a stoppage in the innovation process for funding, the overall increase in speed can compensate for the stoppage in the process. The overall speed is mainly influenced by the customer feedback that is accessed through the use of crowdfunding platforms. What is also mentioned in the literature is the average time needed for external funding, which could be neglected in the case of crowdfunding. The average crowdfunding timeslot is shorter than the time needed for venture capitalists and business angels.

P5. Crowdfunding increases the speed of the innovation process. 5.3.2 Access to additional resources

An interesting finding from these cases is that crowdfunding is also used to get exposure to external partners. In this way, the lack of additional resources from traditional ways of financing could be compensated. This is an addition to the openness of the innovation process, where the crowd is used to gather information. The information gathering from customer is limited to customer preferences and needs and therefore isn’t a compensation for the additional resources of venture capitalists and business angels. One possible explanation is that the costumers are merely seen as customers without specific knowledge about what is needed for the firm. This might also be just the case for product pre-ordering crowdfunding, where the investors are primarily seen as customers instead of investors. For equity-crowdfunding this might be different.

5.3.3 Formalization

(39)

39

experienced in the field they were operating in and this created less need for formalized market research and business planning.

P6. The level of formalization of crowdfunding is lower than in the case of traditionally funded start-ups.

5.3.4 Risk

What I already mentioned is that the innovations that use crowdfunding are often high-risk innovations. In most cases, the entrepreneur tried to decrease this risk by using crowdfunding. The risks are mainly reduced by the early market validation of the innovation. They were able to test the product in the market in an early stage, so the amounts invested are relatively low. This is not only a way to get funding, but it has also a large signalling function to other potential investors, in which the market-fit is already showed.

(40)

40

6. ACADEMIC REFLECTION

In this chapter, I will reflect from an academic perspective on the findings of this study. First, I will discuss the theoretical implications of this research, then the managerial implications. Lastly, the limitations of this research and possible directions for future research will be addressed.

6.1 Theoretical implications

Several theoretical implications can be derived from this research. The first one is that is that the findings suggests that crowdfunding has an influence on the innovation process of start-ups. The main factors that are influenced are the openness of the process, the level of formalization, the stop in the innovation process and finally, the order of the activities within the innovation process. The openness of the innovation and the pre-market validation of the concept are expected to have the biggest influence on the innovation process of these start-ups. A certain stop in the innovation process for marketing and funding is found in this study, which could be an interesting base for further research. Furthermore, within this research I shed light upon the potential implications of crowdfunding for open innovation models and indirectly showed the potential of crowdfunding for start-ups. This could help scholars better understand the innovations that are brought to the market by crowdfunding and interesting future research paths, which I will discuss later in this section.

6.2 Managerial implications

(41)

41

campaign doesn’t necessarily mean that there is no path to the market for that specific innovation. All the innovations of this research, except for one, are still in business and are trying to find other ways to the market.

6.3 Limitations and directions for future research

The most important limitation of this research is the scope. Since this study is limited to seven cases of six different firms, it is hard to check the external validity and make the results generalizable to other cases. Therefore, quantitative studies to the influence I described might be needed. Moreover, the research is conducted by one researcher since it is a Master Thesis, so I conducted all the interviews and the analysis which could bias the results. However, the coding was checked by a master student Communication to increase the internal validity of the research. Another limitation might be the international factor that is neglected in this study. There is no distinction made between the Canadian and the Dutch start-ups and there are no other developed or less-developed countries used in this study.

The innovation process of crowdfunded start-ups Master Thesis MSC BA Small Business and Entrepreneurship MSC BA Strategic Innovation Management University of Groningen Faculty of Economics and Business Martijn Gaasbeek

Abstract

Acknowledgements:

Table of Contents

1. INTRODUCTION

1.1 Research question

1.2 Introduction to Crowdfunding

2. THEORETICAL FRAMEWORK

2.1 Process of innovation

2.2 Financing methods

2.3 Influence of crowdfunding on the innovation process

3. METHODOLOGY

3.1 Research method

3.2 Data collection

3.3 Controllability, reliability and validity

3.4 Data analyses

4. RESULTS

4.1 Within case analysis

4.2 Cross-case analysis

5. DISCUSSION

5.1 Innovation process

5.2 Dimensions of crowdfunded innovation process

5. 3 Influence of crowdfunding on the innovation process

6. ACADEMIC REFLECTION

6.1 Theoretical implications

6.2 Managerial implications

6.3 Limitations and directions for future research