Success factors in crowdfunding : the role of the entrepreneur

(1)

40

Thesis Work

Success Factors in Crowdfunding:

The role of the entrepreneur

Program: MSc in Business Administration – International Management Name: Oliver Farkas / Student No.: 11373822

Supervisor: dr. Mashiho Mihalache Second reader: dr. Vittoria Scalera Date: January 26, 2018

(2)

2

Statement of originality

This document is written by Student Olivér Farkas 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

Due to the increasing number of crowdfunding platforms that provide promising fundraising opportunities for capital seekers, more and more entrepreneurs start to turn to this form of venture financing when starting a business. The rising market of crowdfunding has also attracted academic attention and numerous studies have been published that examined the success factors behind crowdfunding. However, only limited studies have focused on factors related to the project launchers. The objective of this study is to analyze whether certain founder characteristics (i.e: experience, community involvement) affects the success of crowdfunding projects. This study analyzes 162 projects from Kickstarter to find evidence whether project launcher features moderate relationships related to successful funding. The results of the binary logistic regression analysis indicate no support for such moderations, while strengthening evidence on the negative relationship between funding goal and success.

Keywords: crowdfunding, success factors, project launcher characteristics, project

(6)

6

1. INTRODUCTION

It is widely known that new ventures and small businesses require several resources in the early phases of their life cycle, and financing is one of the most important part among them. However, many new ventures have troubles in raising this capital and therefore face a so-called “funding gap”. (Agrawal, Catalini, & Goldfarb, 2015; Macht & Weatherston, 2014; E. Mollick,

2014)

In recent years, a comparatively new form of fundraising technique called “crowdfunding” emerged which allows new ventures and small companies to directly contact the wide-ranging public for financial contribution in order to overcome the above mentioned gap and get their new and innovative ideas off the ground. (Macht & Weatherston, 2014; E. Mollick, 2014; Ryan, Wingerden, & Pieter, 2011)

Apart from being inspired by crowdsourcing, crowdfunding represents a unique type of fundraising, primarily supported by the increasing number of internet platforms that helps resource seekers reach a large number of backers and appeal funding for their ideas. (Belleflamme, Lambert, & Schwienbacher 2013; Bayus 2013; Mollick 2014).

The market of crowdfunding has been growing rapidly in recent years. From a 2.7 billion US$ volume in 2012 it increased to a 34.4 billion US$ just by 2015 and forecasts still show similar increase in the upcoming years.(“Statista,” 2017) Evidence for the importance of crowdfunding can also be found in practice too, for example if we look at the board-gaming industry. From a disappearing and declining market, board-gaming industry reached its highest level due to the continuously increasing successful projects in crowdfunding platforms. The emerging trend in the this industry shows a role model for declining industries as focusing on this way of

(7)

7

financing could be a driving force for them too. Through crowdfunding, even larger enterprises can benefit from several factors, such as direct consumer connection as well as less risky internationalization as they can use crowdfunding to gain popularity in host countries and even use it to measure demand and popularity related to their products or services. (E. Mollick, 2014; “Statista,” 2017)

The fast growth in this model of funding generated a large interest of study, however the research on this phenomenon is still in its early stages. In order to successfully exploit the opportunities of the crowdfunding phenomenon, it is really important to be aware of the drivers of crowdfunding success.

A large part of existing literature focuses on such success factors, in order to have a better understanding of the dynamics behind crowdfunding. Scholars analyzed several characteristics that tends to relate to successful campaigns. A pioneer article from the subject is from Mollick (2014) in which he identifies several project characteristics that enforces project success. These features include “quality signals” (Mollick, 2014, p.7), that can generally inform backers about the project’s quality (i.e: description, website, video), furthermore he identified important characteristics that the project launcher should carefully consider during the creation of his or her campaign (i.e: funding goal, duration). Following up on these results, more and more scholars have started to dive into the topic. (Cordova, Dolci, & Gianfrate, 2015; Koch & Siering, 2015; E. Mollick, 2014)

Nevertheless, available literature only limitedly takes into consideration whether the success of crowdfunding projects depends on the entrepreneur (project launcher) him or herself. This paper suggests that it is an important area to investigate, since entrepreneurs can have different experiences (both in time and form), different social backgrounds (friends and families), and

(8)

8

different contribution to the crowdfunding community. These differences open up the question, whether the project launchers can bring any success factors and if they can, what are these and how do they influence project success.

This paper makes an attempt to examine the success factors that entrepreneurs can bring for the success of their fundraising and empirically test whether those factors contribute to project’s success.

Therefore, my research question is:

Does project launcher characteristics affect relationships between project characteristics and crowdfunding success?

The thesis is structured as follows; first the paper will provide an overview of the most important concepts and available academic literature on success factors and crowdfunding. Based on the theories, this study frames 9 hypotheses. Next section includes the presentation of the sample frequencies, data collection, as well as the description of the measures of the main variables used in the analysis. Afterwards, next section provides the analysis, including the correlation matrix and the results of the hierarchical binary logistic regression analysis. Finally the results are discussed and the study is concluded.

(9)

9

2. THEORETICAL FRAMEWORK

In order to shed light on the underlying factors that influence the success of crowdfunding projects, this paper introduces several concepts. This section provides more details regarding the available funding methods on which entrepreneurs can rely to fill funding gaps they are facing, the mechanisms behind crowdfunding, that model’s importance, and the different types of success factors within crowdfunding projects. Moreover, it also introduces the role of the entrepreneurs in funding success.

2.1. Theoretical background

The most common types of entrepreneurial financing

To understand the mechanisms of crowdfunding, one must first be familiar the available forms of funding. Attracting investors to finance projects and startups is highly difficult. An inability to obtain the necessary funds is one of the chief reasons that new businesses fail. Below, the main investment options available for entrepreneurs are explained in more detail.

Family and friends

The entrepreneur’s inner circle—namely, his or her friends—represents one of the most

important type of funders. Seeking funding from this close group is a seemingly ideal way to raise the capital a new business needs to begin operating. Family and friends can also be a long-term resource, as these individuals are less motivated by strict investment concerns and more driven by support and loyalty to the entrepreneur. The advantage of this type of financing is that it does not require a highly detailed business plan, but great motivation and trustworthiness.

(10)

10

However, the main disadvantage of this type of funding is the fact that the founder combines business with personal relationships, which might lead to unfortunate situations arising from the fact that the savings of relatives and friends are at risk. The founder can try to overcome this factor by signing a contract that states the repayment terms; however, there is still a risk of the business going bankrupt and of the founder being unable to comply with the agreed-upon terms. This type of financing is mostly used in the first round of business funding (Čalopa, Horvat, & Lalić, 2014; Giurca Vasilescu, 2009; startups.co, 2014).

Bank loan

The second funding option consists of applying for a bank loan. Bank lending is the most common source of external financing for many small or medium-sized enterprises (SMEs) and entrepreneurs. Such individuals are often heavily dependent on traditional debt to establish their startup and to meet their cash flow and investment needs. While bank lending is commonly used by SMEs, traditional bank financing poses challenges for SMEs, and this is especially true for newer, innovative, and rapidly growing companies with a higher risk-return profile. The elevated risk might lead to limited capital or necessitate higher interest costs (OECD, 2015; Wiens & Bell-Masterson, 2015).

In order to be eligible for a loan, founders should first present a well detailed business plan for their project. This document should clarify the startup’s obligations related to its provided product or service, and it should also contain a thorough plan of how the investments will be spent. The drawback of relying on a bank loan lies in the fact that the regulations and requirements are quite strict. Moreover, the firm needs to repay the loan, along with interest (Rocket Lawyer, 2017; startups.co, 2014).

(11)

11

Angel investors

Angel investors (business angels) can be generally described as wealthy individuals who use their own wealth to invest in new ventures to help entrepreneurs to realize their business ideas. Angel investors’ investments are typically made in the form of loans or equity in expectation

of earning a high profit after their exit. Apart from contributing monetarily, angel investors are also active in terms of providing valuable experience, skills, expertise, and business networks to the young company. Čalopa et al. (2014) described this investment approach as “smart funding” that facilitates business activity, entrepreneurship, and the creation of new value.

Angel investors are becoming a more and more important investor group, one that can bridge the funding gap between financing from friends and family and formal venture capital financing. They are also a suitable substitute for bank lending; banks are often reluctant to invest in high-risk startups (Čalopa et al., 2014; Giurca Vasilescu, 2009; Mason, Botelho, & Harrison, 2016; OECD, 2015; Rocket Lawyer, 2017; startups.co, 2014).

Venture capitalists

Venture capitalists (VCs) are basically professional companies whose main activity is to find and invest in high-potential early stage companies. Their activity is similar to that of business angels, but an individual’s own resources are not used. In most instances, funding consists of pooled capital gathered from numerous parties. A VC firm raises that money by giving investors the opportunity to be included in a fund that they later use to buy shares in the target companies. These funds offer startups significant financing; however, to attract VCs, entrepreneurs need to have a solid and clear, well explained business plan. An often-mentioned downside of this type of financing is that in return for their investments, VCs usually take a higher portion of the company shares than other types of funding. This means that the

(12)

12

entrepreneur must relinquish some degree of ownership, which might limit that founder’s decision-making autonomy (Law, 2015; OECD, 2015; Rocket Lawyer, 2017; startups.co, 2014).

The concept of crowdfunding

Since crowdfunding is a new and developing model in entrepreneurial finance, the term does not have an “official” definition, and discussion remains ongoing (Forbes & Schaefer, 2017; E. Mollick, 2014) .

The most repeated and reoccurring, common definition is that of Schwienbacher and Larralde (2010, p. 4). They defined crowdfunding as “an open call, essentially through the Internet, for the provision of financial resources either in form of donation or in exchange for some form of reward and/or voting rights in order to support initiatives for specific purposes” (Cordova et

al., 2015; Schwienbacher & Larralde, 2010). Briefly, crowdfunding can be described as the act of collecting small amounts of capital from a large group of people (Kuppuswamy & Bayus, 2015a).

Relying on these definitions, many authors tend to believe and have agreed on the idea that crowdfunding derives from the wider concept of crowdsourcing, which includes using the “crowd” to gain ideas, feedback, and solutions to develop firm actions (Paul Belleflamme,

Lambert, & Schwienbacher, 2014).

Kleemann et al. (2008, p. 6) argued that “crowdsourcing takes place when a profit-oriented firm outsources specific tasks essential for the making or sale of its product to the general public (the crowd) in the form of an open call over the internet, with the intention of animating

(13)

13

individuals to make a [voluntary] contribution to the firm's production process for free or for significantly less than that contribution is worth to the firm” (Kleemann, Voß, & Rieder, 2008).

Although crowdfunding originated from crowdsourcing, it is still a unique way to collect funds. The main difference between the two concepts is that while crowdsourcing refers to gathering labor resources, crowdfunding hinges on another factor: financial resources (Paul Belleflamme et al., 2014; Cordova et al., 2015; Kleemann et al., 2008; E. Mollick, 2014).

This open call for funding can be shared via the increasing number of online crowdfunding platforms (e.g., Kickstarter, Indiegogo, and Starteed). These provide an efficient channel for project founders and backers to connect directly without any monetary intermediators. Such intermediaries would only make the investment process more complex and costly (Cordova et al., 2015; E. Mollick, 2014). Crowdfunding platforms also include all kind of tools that enable financial transactions to be materialized. These include “legal groundwork, pre-selection and the ability to process investment transactions” (Cordova, Dolci, & Gianfrate, 2015, p.2).

A direct connection with entrepreneurs allows potential backers to learn more about the projects and its founders, and they can also gauge the project’s level of popularity among other backers. These capabilities permit potential funders to gather a significant amount of information before making an investment decision. After examining this model, researchers have determined that social information might influence a project’s ultimate success (Kuppuswamy & Bayus, 2015b). From the entrepreneur’s perspective, this online connection, which permits founders to share information about a project through several channels, helps them to reach a higher number of potential investors and strongly facilitates their marketing activities. Furthermore, the online platforms help entrepreneurs to launch their campaigns with

(14)

14

minimal expenses (Cordova et al., 2015; Kuppuswamy & Bayus, 2015b, 2015a; E. Mollick, 2014).

Another unique characteristic that makes this financial phenomenon distinct from other funding options is that these crowdfunding platforms help entrepreneurs not only to obtain the desired financial investment, but also to test their new ideas and measure customer reactions to potential products (Cordova et al., 2015; Harms, 2007; E. Mollick, 2014; Ordanini, Miceli, Pizzetti, & Parasuraman, 2011).

Given the above discussion, crowdfunding tends to exploit online social networks that allow target consumers to join online communities and to actively participate in sharing ideas, feedback, and recommendations regarding new ideas and products (Cordova et al., 2015).

Finally, entrepreneurs can also benefit from fame achieved following a successful and highly popular project, as previous backers or satisfied customers may start following the activities of such founders and their brands. In other words, these individuals become potential future buyers, and their interest can also attract the attention of other potential investors (e.g., VCs and angel investors; Cordova et al., 2015; Forbes & Schaefer, 2017).

Crowdfunding models

Within the larger category of crowdfunding, we can differentiate four main models based on the type of rewards accrued by contributors. These rewards can also be considered the motivations of the funders (E. Mollick, 2014; Rossi, 2014).

(1) Donation-based crowdfunding: This crowdfunding platform model is designed to support project founders without any expected monetary or non-monetary earnings. The backers’ return

(15)

15

on their investment takes the form of emotional wellbeing, rather than financial rewards. Backers are motivated by the positive feeling of supporting a charitable, inspiring, arts-based, or personal project. Initiatives in this category often focus on social, humanitarian, and environmental issues. Belleflamme et al. (2014) pointed out that one of the main advantages of this model is its low cost compared to other sources of capital. Apart from the fee charged by the platform, the project launcher does not face other outstanding expenses (Paul Belleflamme et al., 2014; Gedda, Nilsson, Såthén, & Solberg Søilen, 2016; Kuti & Madarász, 2014; E. Mollick, 2014).

(2) Debt-based crowdfunding: In this second model, contributors play a lender role where they lend money to the project founder (debtor). Funds can be seen as personal loans, and project backers can expect a return on their investment. This model has two variants, one in which the entrepreneur is obligated to pay interest and one in which he or she is not (Gedda et al., 2016; Harrison, 2013; Kuti & Madarász, 2014; E. Mollick, 2014).

(3) Equity-based crowdfunding: In case of this model, backers can be seen as investors. The equity model of crowdfunding enables backers to buy “a stake in a company or a share of the profit” (Harrison, 2013, p. 5). This investment involves more risk for the backers, who are relying a lot more on the success of the project due to their ownership status. Vulkan et al. (2016) found that this model has substantial potential in the future, as it differs from other models of crowdfunding that can bring more popularity to rely on this type of fundraising. The equity-based projects that they investigated pledged a much higher average amount of investments and were able to reach higher funding goals. The more developed pre-money valuation of the projects and the clearer goals of the project funders resulted in a positive financial return. However, in this case, the reward is in the form of a financial product, and as such, it can be subject to government regulations. Effective forms of this model have developed

(16)

16

in the UK, but the implementation progress of this model is slower in the US due to higher regulations (Gedda et al., 2016; Harrison, 2013; Kuti & Madarász, 2014; Vulkan, Åstebro, & Sierra, 2016).

Last, but not least (4) Reward-based crowdfunding: The essence of this approach is that backers receive monetary rewards for their contribution to a company or its project. These non-monetary rewards may take different forms, such as credit within the product (e.g., a movie or book) as a supporter, early access to products or services, early bird discounts on the final products, or project-themed accessories (e.g., tee-shirts, keychains, and stickers). The value and size of the different rewards are mainly determined by the amount of the contribution; a higher contribution usually means a higher reward category. The reward categories are posted and described by the project founders on the campaign page (Kuti & Madarász, 2014; Lukkarinen, Teich, Wallenius, & Wallenius, 2016; E. Mollick, 2014).

The crowdfunding phenomenon has been growing in recent years. Nevertheless, the reward-based approach is the most widely used form, as distributing shares via equity crowdfunding can be more complex. Around 44% of all platforms are centered on the reward-based model, followed by donation-based platforms (28%) and debt- and equity-based models (approximately 14% each; Paul Belleflamme et al., 2014; Kuppuswamy & Bayus, 2015b; Rossi, 2014).

This paper examines reward-based crowdfunding projects and uses data from one of the most popular reward-based platforms, namely, Kickstarter.

(17)

17

Crowdfunding success

When starting a crowdfunding campaign, project founders usually have a clear goal in mind. Unlike many other classical forms of entrepreneur financing, crowdfunding provides a unique opportunity for entrepreneurs to target a wide variety of goals extending beyond financial benefits. These new opportunities mainly stem from the fact that through crowdfunding, project founders can share their ideas with the larger public, which increases customer awareness through publicity and fame. Public awareness may attract various parties, such as new potential investors, partners, and other new connections. Therefore, crowdfunding also allows entrepreneurs to target marketing goals (Macht & Weatherston, 2014; E. Mollick, 2014; Rossi, 2014).

Additionally, if the entrepreneur’s goal is to gain more information and insights regarding demand for his or her product and service ideas, crowdfunding is a suitable instrument to test new offerings. Such assessments are highly valuable in the early development phase (Rossi, 2014). The results can inform project launchers about their products’ popularity among potential buyers. If the campaign does not attract the desired amount of attention, the project launchers can still learn from it and use the information to re-design their ideas. From that perspective, even a project that does not reach its funding goal can be considered a successful campaign (Kaartemo, 2017; Rossi, 2014).

As the above discussion has demonstrated, many factors related to the success of crowdfunding projects are worthy of research. However, this paper measures success in terms of whether the project launcher reaches the targeted goal (E. Mollick, 2014).

(18)

18

2.2. Literature review

In recent years, entrepreneurs needing to raise capital to start a business have benefited from crowdfunding platforms designed to back their creative and novel ideas. An entrepreneur in search of monetary support can use a crowdfunding platform to publish his or her idea, which can also serve as a public relations campaign. Through crowdfunding, he or she can assess the idea’s reception and obtain pre-orders from potential customers. Crowdfunding is also financially attractive as compared to a bank loan. It is a useful tool for entrepreneurs and seems effective in terms of supporting and encouraging them.

However, according to statistics from Kickstarter, a well-known American crowdfunding platform, the probability of creating a successful project is much lower than the probability of failing (“Kickstarter,” 2017). Moreover, Mollick (2014, p. 6) has also argued that “Among crowdfunded projects, failures happen by large amounts, successes by small amounts.” This means that large numbers of projects fail to reach their pledged amounts and that few projects manage to reach their goals (P. Belleflamme et al., 2013; E. Mollick, 2014).

When crowdfunding is properly used, it can serve as an effective tool for entrepreneurs. Despite the current view of crowdfunding as a risky tool for entrepreneurs, it presents significant possibilities for the future (Moutinho & Leite, 2013; Younkin & Kashkooli, 2013).

As the introduction mentioned, academic research on the crowdfunding phenomenon is still in the early stages. However, due to the growing interest in this online form of financing, more and more papers are examining this topic (Kuppuswamy & Bayus, 2015a).

Within the past few years, several studies have sought to identify the factors that impact the success of crowdfunding projects (i.e., whether or not the funding goal is reached). These

(19)

19

analyses have confirmed that there are several characteristics that can be a driving force for success within crowdfunding projects (Koch & Siering, 2015).

Crowdfunding success factors

Project characteristics

Due to the increasing number of crowdfunding success stories, a significant amount of research has started investigating the drivers behind these accomplishments. A large proportion of these studies have focused on previous campaigns and have analyzed characteristics that could potentially explain their positive outcomes.

Quality signals

One of the most frequently cited studies in the crowdfunding literature is that by Mollick (2014). This exploratory study analyzed data from Kickstarter, and while taking into account several variables, the author found that by adding videos, linking the project to a website, and providing updates and descriptions, a founder could increase the likelihood of his or her project being successful funded. In addition, the number of comments on a project page was also positively related to the level of funding. These variables are collectively referred to as “quality signals” (Mollick, 2014, p.7), meaning that by looking at these project characteristics, one can

make an educated guess regarding a campaign’s quality (Kaartemo, 2017; E. Mollick, 2014; Moritz & Block, 2014).

These findings on quality signals have been supported by several other studies, underlining their importance as a means of communicating and of providing information to backers. Apart from supporting Mollick’s (2014) results, Gascon et al. (2015), Hobbs et al. (2016), and

(20)

20

Mollick and Nanda (2015) also highlighted its outstanding impact in terms of attracting the attention of the crowd, especially within reward- and donation-based crowdfunding platforms (Gascón, Rodríguez, Monforte, López, & Masip, 2015; Hobbs, Grigore, & Molesworth, 2016; Kaartemo, 2017; Kim, Por, & Yang, 2017; Koch & Siering, 2015; E. Mollick, 2014; E. Mollick & Nanda, 2015).

Regarding communication with backers, another quality signal has also been added to the group, namely, the depth of the textual description, plays a role, as crowdfunding platforms permit project founders to post information about their products and services when launching their campaigns. The amount of information posted is positively related to the degree of help that the founder is providing to contributors. This information is typically similar to a regular product description or a user review on an online shopping platform. In this context, improved text (more detailed information) tends to be more useful for readers. In addition, Marom and Sade (2013) have also found that success is related to the language of the project description, and Xiao et al. (2014) highlighted that communication with platform members and visitors through continuous updates is also important in determining a project’s level of funding (Forbes & Schaefer, 2017; Koch & Siering, 2015; Marom & Sade, 2013; Mudambi & Schuff, 2010; Xiao, Tan, Dong, & Qi, 2014).

Funding goal

Another key takeaway of Mollick’s (2014) study regards the significant impact of the funding target and the project duration on the outcome. Mollick (2014) proved that the probability of a project being successfully funded decreased as the funding goal and campaign timeline increased (E. Mollick, 2014; E. Mollick & Kuppuswamy, 2014).

(21)

21

This factor later drew significant research interest and has been supported by other studies as well. Research has suggested that if a project founder is trying to motivate backers to contribute, he or she should keep in mind that successful projects tend to have a lower, more realistic funding goals than failed projects. This has especially proven true in studies examining reward- and donation-based platforms (Cordova et al., 2015; Gascón et al., 2015; Kaartemo, 2017; Kim et al., 2017; Koch & Siering, 2015).

However, entrepreneurs may face a degree of confusion on this point, as other studies (e.g., Lukkarinen et al., 2016) relying on equity-based crowdfunding data have provided contradictory evidence. These works have revealed that contributors are more motivated and active in campaigns with higher funding goals (Kaartemo, 2017; Lukkarinen et al., 2016).

Therefore, previous studies have advised project founders to have a deliberate goal in mind so that they can clearly communicate their beliefs and the potential of their projects. Extensively analyzing founders’ needs and examining the different options in terms of crowdfunding platforms are also critical research topics (Gedda et al., 2016; Kaartemo, 2017).

The conflicting evidence and studies on the importance of the right funding goal size highlight the potential in terms of further research and discussion. Consequently, to further assess the impact of the size of the funding goal, the author hypothesized that there is a negative relationship between a campaign’s funding goal and its success.

H1: There is a negative association between the size of the funding goal and the success of a

crowdfunding project.

(22)

22

As mentioned above, both the duration and funding goal are positively related to funding. Kuppuswamy and Bayus (2015b) investigated the dynamic pattern of backer support over the funding cycle and concluded that support follows a U-shaped pattern; projects receive most of their contributions during the first and last weeks. Lukkarinen et al. (2016) and Mollick (2014) further found that an increase in the campaign’s duration decreases the probability of the final project success. They argued that while projects need a certain period of time to obtain recognition, the overall deadline should be close enough to generate a feeling of urgency, motivating contributors to fund the project. Otherwise, backers will not feel a sense of time pressure of missing a chance, and potential contributors can also be demotivated by waiting for others’ reactions (Cordova et al., 2015; Kaartemo, 2017; Kuppuswamy & Bayus, 2015b; E. Mollick, 2014).

Considering these findings, duration is among the most important variables when creating a crowdfunding campaign. Several advisory articles for fund seekers have suggested effective campaign lengths for the different available crowdfunding platforms. Consequently, in order to further evaluate the evidence regarding the impact of project duration, the author hypothesized that there is a negative relationship between a campaign’s duration and the success of the crowdfunding project.

H2: There is a negative association between a campaign’s duration and the success of the

crowdfunding project.

Rewards

Last but not least, rewards are another factor that can drive project success. First of all, Meer (2013) argued that higher priced rewards decrease the probability of project success based on reward-based crowdfunding platforms. In addition, Mollick and Nanda (2015) argued that the

(23)

23

number of rewards is also related to a successful outcome, proposing that the more rewards are available for backers, the more likely is the crowdfunding project to perform well (Kaartemo, 2017; Meer, 2013; E. Mollick & Nanda, 2015).

Nevertheless, Hörisch (2015) has suggested that the quality of the reward options is more important for backers than the number of reward options. His claim was also supported by Hobbs et al. (2016). These differing views may be a result of the above researchers considering different models of crowdfunding (Hobbs et al., 2016; Hörisch, 2015; Kaartemo, 2017).

Finally, Forbes and Schaefer (2017) followed a unique approach. Most of the available literature on crowdfunding success factors has taken the form of quantitative analyses, but these two researchers used a qualitative method to gain further insights into factors that lead to successful crowdfunding. As regards rewards, their interviews demonstrated that backers are more likely to invest in a project if the reward is the product itself. In other words, the main motivation for funding a project is the desire to buy the product (Forbes & Schaefer, 2017).

However, Forbes and Schafer’s (2017) findings on the relationship between product as reward option and project success have not yet been widely backed by quantitative evidence. Consequently, to validate their evidence regarding the impact of rewards, this author hypothesized that there is a positive relationship between a campaign’s product as a reward option and the success of the crowdfunding project.

H3: There is a positive association between the availability of the campaign’s product as a

(24)

24

Considering these studies on project characteristics has illustrated that success factors can take a variety of forms. By focusing on these findings, project launchers can increase their chances of success while consciously crafting their campaigns to successfully reach their funding goals.

The role of the entrepreneur

Obtaining early phase venture funding is difficult. Success depends on various factors, but the main value behind a project is the idea itself. Funding parties need to see and believe that the idea in which they are investing can affect a large group of targets and that they will be likely to realize the expected return on that investment. A project’s underlying idea can target the social wellbeing of a given community or consist of an innovative solution intended to disrupt existing markets. Therefore, the expected return may vary according to the essence of the concept (Frydrych, Bock, Kinder, & Koeck, 2014).

However, after an idea attracts potential funders, those individuals must also consider other factors before making an investment decision. Examples from practice have revealed that potential investors (e.g., VCs) consider the entrepreneur important, as he or she is the person responsible for the success of the ideas (Block, Fisch, & Praag, 2016; E. R. Mollick, 2013). Schumpeter (2003) has defined an entrepreneur as a relatively unique actor willing to break through traditional structures and to change known ways of doing things. The Schumpeter-type entrepreneur is “individualistic, self-directed, has an inner drive to innovate” (Schumpeter & Backhaus, 2003).

Studies have found that when deciding whether to invest in a particular company, VCs are particularly attentive to the quality of the startup team. In particular, they rely on the heterogeneity of individual skills and founders’ backgrounds and past successes as a sign of potential future success (E. Mollick, 2012).

(25)

25

To ensure that they are well informed about the backgrounds of founders, VCs often rely on a highly personal selection process, gaining information about the above-mentioned factors through face-to-face meetings, shared activities, and the extensive use of shared network connections within the VC industry (E. R. Mollick, 2013).

Considering these points, the question arises of whether all founder characteristics are similarly important with respect to crowdfunding. During a crowdfunding project, investors (the crowd) have limited information about project founders, as there is no possibility to meet them in person.

Experience

The existing literature has suggested that entrepreneurs with prior experience founding firms have a better chance of being funded by external investors. To build a firm, a startup founder must bring together a group of people with various kinds of expertise, gain access to capital and other resources, and act as an effective manager to implement a successful business plan. To effectively carry out these tasks, a founder needs to have a wide range of skills. An entrepreneur may be born with a general set of skills, which he or she may then enhance by investing in the human capital (e.g., formal training). However, a more effective way to obtain the necessary skillset is through “learning by doing.” Participating in the process of building a company is highly important given that certain entrepreneurial skills are overlooked and are difficult to learn without real-life experience. Because of the significance of “learning by doing,” experienced startup founders are expected to have a more complete set of skills and to

therefore perform at a higher level. Past experience thus translates into increased potential for investors in a future company (Koch & Siering, 2015; E. R. Mollick, 2013; Zhang, 2011).

(26)

26

There is limited research on project launchers’ previous crowdfunding experience as a success factor. Koch and Siering (2015) analyzed this element as a variable, but they did not find any significant relationships. However, a recent study by Kim et al. (2017) identified a positive relationship between previous experience and crowdfunding success. Their analysis was based on 116,956 Kickstarter projects (Kim et al., 2017; Koch & Siering, 2015).

Following up on the study of Kim et al. (2017), this paper argues that previous crowdfunding experience permits project launchers to increase their probability of success. This thesis does not differentiate between past failures and successes, as both experiences could provide useful insights for founders regarding what was or was not effective during a previous campaign. Rather, one can assume that learning from the past allows founders to more successfully design their next campaigns to achieve their goals. Considering the above-mentioned relationships related to project characteristics points to an interesting question, namely, whether the crowdfunding experience of project launchers can help them overcome those factors that tend to decrease the probability of success or to enhance those factors that increase the likelihood of success.

Consequently, the author hypothesized that the experience of the project launchers has a positive moderating effect on the relationship between project characteristics and crowdfunding success.

A project launcher with previous experience may be able to arrive at a more realistic valuation of his or her idea. In such situations, even a higher funding goal can attract a sufficient number of potential backers. Through their prior projects, founders may also build a loyal pool of repeat backers. Such individuals follow a project founder’s future activities and can draw attention to

(27)

27

an initiative, attracting additional new backers and enabling a higher funding goal. After a highly successful campaign, fame on the part of the founder can have a similar effect.

Overall, previous experiences may provide the project launcher with a well-developed skillset and confidence that he or she can use to craft a more successful strategy when aiming to realize a higher funding goal.

H4: The project founder’s crowdfunding experience positively moderates the relationship

hypothesized in H1.

The above-mentioned advantages of experience can also help a project launcher with other adjustments during the process of developing a new campaign. As Mollick (2014) and Lukkarinen et al. (2016) pointed out, founders need to rely on relatively short campaign durations to maintain backers’ feelings of urgency. However, this paper argues that past experience weakens this negative relationship because a founder who knows how to attract attention for a longer period of time can gain more backers and help his or her campaign to reach or exceed its funding goal. Exceeding the funding goal is typically possible on most crowdfunding platforms (Lukkarinen et al., 2016; E. Mollick, 2014).

Furthermore, an experienced project launcher may be able to more efficiently set reward options and to develop more popular choices that are closely related to the underlying idea. Such a model better positions the campaign for success.

(28)

28

Educational background

“Learning by doing” might add to the founder’s skillset, but to become a successful manager

of a young business, entrepreneurs also need the right educational background. Formal training can provide important skills that support effective action. Learning from business case studies and professional teachers can help entrepreneurs overcome issues that have previously been studied in a hypothetical environment. Education also develops an entrepreneur’s critical thinking skills, preparing him or her to face future difficulties. Allison et al. (2017) hypothesized a possible positive relationship between entrepreneur education and crowdfunding performance, but the evidence did not provide adequate support (Allison, Davis, Webb, & Short, 2017).

Gender

Considering the characteristics of project founders raises the question of whether gender has an impact on project success. Literature on the influence of this variable is highly limited. Barasinska and Shäfer (2014) sought to determine whether this new form of venture financing has reduced the barriers facing female borrowers and entrepreneurs. Unfortunately, their results did not indicate whether gender impacted the likelihood of a crowdfunding project succeeding (Barasinska & Schäfer, 2014; Kaartemo, 2017; Marom, Robb, & Sade, 2016).

However, a recent study by Marom et al. (2016) investigated gender dynamics on Kickstarter. Their findings revealed that male entrepreneurs tended to set higher funding goals on average, but in terms of success rate, women were more likely to reach their funding targets. As the

(29)

29

section on funding goal has already noted, Marom et al. (2016) explained this difference as stemming from the fact women tend to set lower funding targets. Nevertheless, after they controlled for the funding goal size, women still enjoyed a higher success rate. The researchers explained the difference via characteristics other than a decrease in discrimination. For example, women are more risk averse and less confident. Hence, they may underestimate the demand for their ideas and set lower and more achievable funding goals than men. Moreover, Marom et al. claimed to have found evidence of taste-based discrimination ,within their survey of Kickstarter investors. (Marom et al., 2016). Taste-based discrimination is explained as the act of “an economic party an economic player who dislikes or prefers not to be associated with

individuals of a given race, gender, ethnicity, religion, certain defined status, or other defined personal characteristic” (Marom et al., 2016, p.7).

Commitment to a community

As discussed earlier, founders often gather investments from friends and family. In that environment, they are involved in an interconnected group in which they are personally known through shared personal activities. On crowdfunding platforms, a so-called “crowdfunding community” is present, and that group helps project launchers to more effectively gain support. By contributing to others’ projects, an entrepreneur can enter this community and gain access

to a network of potential supporters (E. Mollick, 2014).

Koch and Siering (2015) analyzed whether the number of projects that the project founder previously backed has a direct impact on project success, and they observed that this factor has a positive influence on crowdfunding success. Hence, founders who actively back other entrepreneurs are more likely to succeed with their own projects than are other founders (Koch & Siering, 2015).

(30)

30

Considering these studies, this paper argues that project launchers involved in the crowdfunding community can potentially acquire a group of loyal “friends” who will help them to overcome general difficulties during their campaigns. Communities can arise as project launchers can see their backers’ profiles. If their backers have their own campaigns, the

launcher can visit their pages. Moreover, the positive feeling of being backed can encourage a founder to support other members’ campaigns, either as a sign of gratitude or out of a genuine interest in their ideas. Members of these communities can also learn from each other by analyzing other projects and discussing possible success factors and tips. The questions arise of whether this community gives project launchers a higher probability of success and whether this relationship could help to mitigate negative factors or to strengthen the impact of positive variables.

Consequently, the author hypothesized that community involvement on the part of a project launcher has a positive moderating effect on the relationship between project characteristics and crowdfunding success.

Similar to experience, these communities can strengthen a founder’s basic backer pool in many ways. These communities not only attract other project founders as potential backers, but also allow investors who supported a founder’s past projects to contribute once again. These investors can identify those projects backed by the funded entrepreneur, which may lead to further investments in other projects.

With the help of such a network, project founders may be able to design their campaigns in a more flexible manner and to aim for even higher funding goals. The more they contribute to other entrepreneurs, the larger the network available to them. When they enjoy significant

(31)

31

support from general community connections, project founders have better chances of raising a high amount of money and of achieving strong recognition for their campaign.

H7: A project founder’s crowdfunding community participation positively moderates the

relationship hypothesized in H1.

The above-mentioned advantages associated with the crowdfunding community may also affect the campaign duration; project founders with community involvement may benefit from a longer campaign, as they may use that timeframe to efficiently exploit their connections.

In addition, the insights gained from the crowdfunding community may improve a project launcher’s campaign in a general sense, including in terms of the reward options. These

enhancements may strengthen the positive impact of product-related rewards.

As the above discussion demonstrates, the literature on the link between fund-seeker characteristics and project success is highly limited, especially as regards whether these variables moderate the relationship between project success and project characteristics.

(32)

32

Figure 1: Proposed conceptual model of the research

Project characteristics Successful funding

Project founder ‘s characteristics

(33)

33

3. DATA AND METHODS

3.1. Sample and data collection

This section represents the beginning of the empirical part of the paper. First, Kickstarter is introduced in more detail, including information about its nature and rules, the reasons behind its popularity, and its operations. Afterwards, the basic features of the sample are described, as is the data collection process. Furthermore, the variables included in the study are presented. Finally, the section briefly outlines the statistical methods used to test the expected relationships discussed in the literature review. The full survey is found in the appendix.

Kickstarter

The study’s overall population is comprised on different projects from one of the most famous

and widely used crowdfunding platforms, Kickstarter.

Kickstarter is a reward-based crowdfunding platform, which means that in return for backing a project, supporters receive rewards based the amount of funding they provide (the reward is defined by the entrepreneur/project launcher who is setting up the campaign). Therefore, they are not exclusively motivated by financial profits. This platform offers entrepreneurs a unique opportunity to collect funds without having to give away equity or ownership of their ideas. This factor is one of the key benefits responsible for the high popularity of this model for raising capital (Bi, Liu, & Usman, 2017; Frydrych et al., 2014; “Kickstarter,” 2017).

The site’s mission is “to help bring creative projects to life” (“Kickstarter,” 2017). Since Kickstarter launched in April 2009, over 14 million backers have supported projects. The site has helped entrepreneurs to raise over $3.4 billion, which makes it the largest reward-based crowdfunding platform (“Kickstarter,” 2017).

(34)

34

Thousands of projects are active at any time in a wide range of categories, such as art, music, food, games, photography, and technology. On Kickstarter, project launchers (e.g., artists, musicians, and designers) have complete control over and responsibility for their projects. After registering on the site, they can start building a project page, where they can post videos, upload pictures, give descriptions of their ideas, and manage the reward options for different contribution amounts. These rewards are usually early bird offers or copies of the envisioned product, thank-you notes, or other customized gifts related to the project. Before launching their projects, creators also need to set a target funding goal (no limits) and a duration for the campaign (launch date and deadline). The duration can range from 1 day to a maximum of 60 days. The platform offers suggestions on how to adjust those settings, but there is no best approach, as every project is different to a certain level. (“Kickstarter,” 2017).

Nevertheless, project creators need to be thoughtful when deciding on these settings, as Kickstarter distinguishes itself using a so-called “all-or-nothing-funding” method. This means that if a campaign does not reach its funding goal within the given timeframe, it will not receive any funding. The accounts of project backers are only charged if the project’s funding goal is reached (successful funding). If not, they are not charged, and the project creator is not able to access the money. This is a strict rule as compared to other crowdfunding platforms featuring an option to let campaigners keep any funds raised (“Kickstarter,” 2017).

Other rules include that Kickstarter projects cannot consist of fundraising for charity and cannot offer equity. The focus is on startups in their early phases and new products. The platform has helped over 136,608 projects to be funded successfully (“Kickstarter,” 2017).

3.2. Measures

(35)

35

analysis, numerous different types of data were required. The study included one dependent variable, four independent variables, four moderator variables, and three control variables.

Dependent variable

The dependent variable was project success, which represented the funding outcome of the launched projects. This dummy variable indicated whether each project reached the target amount of funding before its deadline (0 if the project did not manage to reach its goal; 1 if the project did manage to reach its goal). Launching a project on Kickstarter requires the user to establish a project duration (1–60 days) and a required funding amount. If the project is not fully backed before the deadline, the project launcher does not receive any funds for his or her project (Cordova et al., 2015; “Kickstarter,” 2017; E. Mollick, 2014).

This data were collected from the above-mentioned master dataset.

Independent variables

The independent variables were selected to collectively test the project characteristics. These variables have generally yielded informative results in studies examining reward-based crowdfunding models and Kickstarter.

The funding goal was the target funding amount that the project launcher established during

the campaign setup stage. All currencies were converted to U.S. dollars for all the projects. These data were extracted from the master dataset.

Duration constituted the number of days between the launch date and the project deadline.

(36)

36

Product as reward was a dummy variable. It equaled 0 if the project’s product or service was

not an option as a reward and 1 if the project’s product or service was indeed a reward option. The data were collected manually from the projects’ sites.

Moderator variables

Experience represented the number of previous projects launched by the fund seeker. The

project sites and the survey provided the necessary data

Community represented the number of projects the fund seeker had supported. These data were

collected from the project sites and the survey.

Control variables

Number of rewards indicated the number of reward options that the project launcher provided

while setting up the campaign site. Data originated from the project sites.

Country was measured as a dummy variable (0 if the project was launched outside of the US;

1 if it was launched from within the US). The master dataset contained these data.

Social media was another dummy variable. It equaled 0 if a social media site was not linked to

the project site and 1 if a social media site was linked to the project site. These data were gathered from the project sites.

Website was likewise a dummy variable (0 if an accessible website was not linked to the project

page; 1 if an accessible website was linked to the project page). These data were drawn from the project sites.

Education indicated the level of education of the project launcher. It was measured on a 1–5

(37)

37

to the project, while 5 indicated that the participant had a bachelor’s or master’s degree in a field closely related to the project in question. These data were collected from the survey.

Gender was again a dummy variable: 0 if the participant was male and 1 if the participant was

female. The survey yielded the necessary data.

3.3. Sample

To create the sample, data were collected from two sources. As a first and basic source, a master database of crowdfunding projects was extracted from the website

https://webrobots.io/kickstarter-datasets/. This website uses a scraper robot to collect data on all Kickstarter projects. Data is collected and then made available once per month. Recent projects are also included (nicerobot, 2017).

The extracted dataset revealed several project characteristics. These included features such as the funding goal, the pledged amount during the campaign, the launch date and deadline, the project’s ultimate outcome in terms of funding, the number of backers, the location of the launch, and the currency and exchange rates. The dataset also included web links to the individual project pages on Kickstarter, which made it possible to collect additional project characteristics and to directly contact the project founders.

First, the main filter was applied on project outcome (whether the project was funded successfully or unsuccessfully). This filer was appropriate because the goal was to identify project drivers that enhance project success. Hence, both successful and unsuccessful projects were needed to identify the effects of these characteristics and how they contribute to a project’s outcome.

To obtain a secondary source, the author, using the above-mentioned contacts from the project sites, sent a questionnaire to 472 randomly selected project founders. Of the 171 project

(38)

38

founders who started the questionnaire, 165 fully completed the survey (response rate 32%). Of the 165 respondents, 40.6% were female, and 49.4% were male. A majority (52.1%) had at least a bachelor’s of science degree or higher, 50.9% were between 18–29 years old, 30.3% were between 30–29 years, and a minority (18.8%) were over 40 years old. It is important to mention that the response rate among unsuccessful project launchers was much lower than the response rate for successful project launchers. Therefore, more invitations were sent to this group to achieve a more balanced sample.

In the final sample, 57% of the projects were successful, and 43% were unsuccessful. As for the size of the funding goals, the project that aimed for the lowest funding target asked for $22, while the highest funding goal was $3.5 million, Due to this large maximum value, the average funding goal was $69,639.

Out of the 165 total projects, 113 were launched in the US (68.5%), while the remaining 52 were from other countries (31.5%). This latter group included several other countries, such as Australia, Canada, Mexico, Denmark, Germany, the UK, and Switzerland. Last but not least, the average campaign duration was 33.1 days, with a maximum of 60 days and a minimum of 6 days.

As the sample is based on projects that received response, it resulted with unbalanced distribution in case of some variables. To overcome this issue, funding goal and the number of

rewards variables were needed to be transformed. To do so, a log transformation was made.

Finally, the log-transformed data follows a normal distribution. The normality test also revealed missing data; the associated cases were removed, resulting in a final sample of 162 cases (N=162).

(39)

39

4. ANALYSIS AND RESULTS

4.1. Correlation matrix

Using transformed and original variables, the author developed a table that included the means, standard deviations, and the correlation coefficients for all the variables (Table 1).

First, by examining the values for the independent and dependent variables (project characteristics and funding success), the author arrived at the following conclusions:

The variables funding goal and project success had a strong negative correlation (r= -.526, n=162, and p < 0.01), which suggests a close relationship between the size of the funding goal and project success (r was above 0.5 in absolute terms) (Pallant, 2013). Calculating the square of r = -.526 revealed that the coefficient of determination was 0.2767, which meant that the variables shared 27.67% of their variance.

Duration and project success also had a negative, albeit more moderate, correlation (r = -.419,

n=162, and p < 0.01), as the correlation value, r, was below 0.49 but above 0.3 in absolute terms (Pallant, 2013). The coefficient of determination was 0.1756, meaning that the variables shared 17.56% of their variance.

The variables product as reward and project success had a moderate positive correlation (r = .356, n=162, and p < 0.01) (Pallant, 2013). Additionally, the coefficient of determination was 0.1267, meaning that these variables shared 12.67% of their variance.

Furthermore, examining the values of the independent variables demonstrated that duration and funding goal had a medium positive correlation (r = .356, n=162, and p < 0.01) (Pallant,

(40)

40

2013). In addition, the coefficient of determination was 0.1467, meaning that the variables shared 14.67% of their variance.

However, product as reward was only slightly negatively related to both project goal (r=-.160, n=162, and p < 0.05) and duration (r = -.158, n=162, and p < 0.05) (Pallant, 2013).

Second, by analyzing the values for the moderator variables (project launcher characteristics), the author identified the following relationships:

The variables experience and funding success were slightly positively correlated (r = .268, n=162, and p < 0.01), as the correlation value, r, was below 0.29 (Pallant, 2013). The coefficient of determination was 0.0718, meaning that the variables shared 7.18% of their variance.

Community and funding success had a medium positive correlation (r = .413, n=162, and p <

0.01), as r fell between 0.3 and 0.49 (Pallant, 2013). The coefficient of determination equaled 0.1706, meaning that the variables shared 17.06% of their variance.

Table 1 further illustrates that experience (r = -.331, n=162, and p < 0.01) and community (r = -.219, n=162, and p < 0.01) had a medium and small negative effect, respectively, on the

funding goal (Pallant, 2013). Moreover, both experience (r = -.327, n=162, and p < 0.01) and community (r = -.296, n=162, and p < 0.01) had a medium negative effect on duration.

However, both experience (r =.157, n=162, and p < 0.05) and community (r =.245, n=162, and p < 0.01) had a small positive effect on the likelihood of the product itself constituting a reward.

As for the link between these two moderator variables, experience and community had a medium positive correlation (r =.338, n=162, and p < 0.01), and the square of r=.338 indicated

(41)

41

that the coefficient of determination was 0.1142, meaning that the variables shared 11.42% of their variance (Pallant, 2013).

Finally, the correlation matrix mainly suggested medium correlations among the other variables. Education (r =.377, n=162, and p < 0.01), country (r =.314, n=162, and p < 0.01),

number of rewards (r =.272, n=162, and p < 0.01), and website (r =.327, n=162, and p < 0.01)

all had a medium positive effect on project success. As regards the remaining relationships, the medium positive correlation between website and product as reward stood out with its higher correlation (r =.456, n=162, and p < 0.01). The coefficient of determination was 0.2079, meaning that the variables shared 20.79% of their variance.

(42)

40

Table 1. - Means, Standard Deviations and Pearson Correlations

Mean S.D. 1 2 3 4 5 6 7 8 9 10 11 12 1. Project success 0.579 0.495 1 2. Funding Goal 8.328 2.093 -.526** 1 3. Duration 32.761 12.377 -.419** .383** 1 4. Product as reward 0.818 0.387 .356** -.160* -.158* 1 5. Experience 2.119 5.664 .268** -.331** -.327** .157* 1 6. Community 0.465 0.500 .413** -.219** -.296** .245** .338** 1 7. Education 2.868 1.360 .377** -.183* -0.108 0.086 0.093 0.110 1 8. Gender 0.409 0.493 0.088 0.000 0.052 -0.005 -0.101 -0.083 .204* 1 9. Country 0.679 0.468 .314** _-0.151 _-0.097 _.199* _.205** _0.155 _.182* _0.078 ₁ 10. Number of rewards 1.967 0.843 .272** -0.036 -.292** .229** .384** .279** .178* .169* 0.144 1 11. Website 0.736 0.442 .327** -0.075 -.231** .456** .169* .302** 0.110 0.150 .200* .325** 1 12. Social Media 0.516 0.501 0.014 -0.138 0.020 -0.067 .185* .198* -0.039 0.063 0.035 0.049 0.047 1

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

(43)

43

4.2. Hierarchical Binary Logistic Regression Analysis

In order to test the hypotheses, a binary logistic regression analysis was made (Table 2). This is the most suitable analysis for our test because of the binary dependent variable, meaning that there are only two possible outcomes (in this case: success and failure).

In Model 0 of the hierarchical binary regression model, the constant-only provides a decent base point for the analysis as it does not contain any of the predictors yet, and all perceptions are based on the largest outcome category of the research’s dependent variable. (Peng, Lee, & Ingersoll, 2002) In that case, Model 0 represents that all cases included in the analysis are successful (Project success = 1), this model has a 56.8% classification accuracy of Project success estimation. The classification overall percentage will show us the improvement in the classification accuracy throughout the further steps of the analysis. Later, these improvements over the null-model will be further analyzed by additional statistical tests, including the significance of the predictors, the overall model evaluation and Cox & Snell R square tests. (Peng et al., 2002)

Model 1 is considering control variables only. The classification accuracy improved to 73.5% while the model summary shows that Cox & Snell R² = 0.264 and Nagelkerke R² = 0.354, demonstrating that the model explains between 26.4% and 35.4% of the variance in project success.

In Model 2, the moderator variables experience and community were added. That improved the classification accuracy to 77.2%, while the model explains between 36.3% and 48.7% of the variance in project success.

(44)

44

Model 3 further includes the independent variables. It further improved the classification accuracy to 84%, while the model explains between 53.5% and 71.7% of the variance in project success.

Finally, in Model 4 the interaction terms were added. Interaction terms were calculated as the product of the independent variables and moderator variables. (Interaction term = IDV*MDV) (Pallant, 2013) The overall percentage of classification accuracy is 90.7% meaning that the final model is highly appropriate to predict the outcome. The closer the overall percentage is to 100% the more appropriate the modes for testing the hypothesis. In the final model demonstrates Cox & Snell R² = 0.577 and Nagelkerke R² = 0.774, demonstrating that the model explains between 57.7% and 77.4% of the variance in project success.

Table 2 presents the results of the binary logistic regression. To understand the outcome, we need to consider the significance (p-value) and additionally the Ex (B) which signals the probability of a curtain outcome (determining how an increase in the variable affect the successful funding).

(45)

45

4.3. Results

Table 2 provides the output of our binary regression analysis (see Model 4). First, H1 proposes that the size of the funding goal has a negative association with the project success. The results in Model 4 show significant effect on success (p = .001), as well as Ex (B) is lower than 1, which means that H1 is supported. According to Ex (B) an increase in the size of the funding goal decreases the probability of the project success (Ex (B) = 0.415). The outcome of the correlation matrix also supports this argument (r= -.526, n=162, and p < 0.01), indicating that the higher the funding goal set by the entrepreneur, the less likely that the project will succeed. Therefore, H1 is accepted.

Table 2: Hierarchical Binary Logistic Regression

Variable Sig. Exp (B) Sig. Exp (B) Sig. Exp (B) Sig. Exp (B)

Constant 0.000 0.027 0.000 0.024 0.140 13.769 0.548 4.361 Education 0.000 1.746 0.000 1.794 0.001 2.143 0.001 2.525 Gender 0.522 0.773 0.700 1.187 0.788 1.180 0.243 2.364 Number of rewards 0.118 1.496 0.469 1.245 0.283 1.614 0.205 2.071 Social media 0.829 0.922 0.307 0.649 0.283 0.543 0.361 0.524 Country 0.009 2.867 0.052 2.331 0.039 3.256 0.058 3.714 Website 0.002 3.904 0.065 2.467 0.397 1.899 0.369 2.299 Experience 0.041 1.367 0.580 1.094 0.162 26.009 Community 0.003 3.841 0.005 5.613 0.648 6.040 H1 Funding goal 0.000 0.500 0.001 0.415 H2 Duration 0.016 0.929 0.132 0.938 H3 Product as reward 0.068 4.399 0.047 11.228 H4 Funding goal_Experience_INT 0.429 1.148 H7 Funding goal_Community_INT 0.598 1.221 H8 Duration_Community_INT 0.660 1.028 H5 Duration_Experience_INT 0.002 0.849 H6 Product as reward_Experience_INT 0.358 3.561 H9 Product as reward_Community_INT 0.285 0.082

Valid N=162 (number of cases)

Success factors in crowdfunding : the role of the entrepreneur

Thesis Work