The success of the performing arts on crowdfunding : a case study of a Dutch platform

The success of the performing arts on crowdfunding:

A case study of a Dutch platform

February 16, 2015

MSc Business Studies | Entrepreneurship & Innovation track Master Thesis

Supervisor: Dr. G.T. Vinig

Author: Anne-May Meurs (10677313) Word count: 14175

TABLE OF CONTENTS

1 Introduction ... 1

1.1 Crowdfunding and the creative industries: some more successful than others? . 2 2 Literature review (Theoretical framework) ... 6

2.1 Crowdfunding 101 ... 6

2.2 The stakeholder dynamics ... 8

2.2.1 Creators ... 8

2.2.2 Donators ... 10

2.2.3 Platforms ... 12

2.3 Patterns and determinants of success, what has been researched so far? ... 14

3 So what about the performing arts? ... 19

3.1 A word on the performing arts in the ‘Age of Internet’ ... 19

3.2 The status of the performing arts in the Netherlands, and the rise of crowdfunding. ... 20

3.3 The why, how, and what of Voordekunst.nl ... 21

4 Conceptual model ... 23

5 Methodology ... 24

5.1 Research design ... 25

5.2 Data collection ... 25

6 Analysis ... 27

6.1 Success rates across categories ... 27

6.2 Descriptive statistics ... 27

6.3 Zooming in on differences between Successful and Failed projects ... 28

7 Grammatical persons ... 32 7.1 Set up ... 33 7.2 Results ... 34 8 Conclusion ... 37 9 References ... 40 !

Preface'

With ‘just’ my thesis remaining as a last step in my educational career, it suddenly dawned upon me that this was a moment to savor, for taking that last step would mean a new phase of my life would start. Not that that is necessarily a good or a bad thing. But it did get met thinking. Up until this moment, life has been divided into ‘chunks’ of several years where the program has been outlined and the requirements to go on to the next phase are clear. Up until this moment – while having enjoyed every moment of each ‘chunk’, there was always a lingering feeling of wanting to already be in the next chunk. Suddenly I wasn’t so sure that I was ready for the next chunk, this chunk of which the size was not predefined, but which I myself could shape as I see fit. What would I make of it? Apply for a job? What kind of job? In Amsterdam, or maybe even abroad? Do I even want a job? The typical generation Y dilemma – wanting to do something meaningful, something worthwhile and fulfilling, and not just any job.

While I had started reading literature and devising a research plan for my thesis, I no longer felt this rush. Maybe in retrospect this was just the procrastinator in me looking for excuses not to work on my thesis. And I must say that the motivation to work on it is just as U-shaped as the donation pattern in the crowdfunding projects I was researching. As frustrating as the process has been at times, rearranging the structure and contents back and forth; along the way it became clearer what the structure should be.

It also gave me time to think. Around me friends had started getting jobs and started working, some of them loving what they do, others taking anything they can get, because – hey we’re still in a recession aren’t we?

If there’s anything about the Entrepreneurship theory that has stuck with me, it is the effectuation theory of Sarasvathy:

“The future cannot be predicted, but entrepreneurs can control some of the factors which determine the future.”

This not only applies to entrepreneurs but to life in general as well. Life is about making choices and creating opportunities, and when along the way the opportunity you may have envisioned beforehand changes into something else that is just the way it goes. This in combination with my recent interest in quantum physics and the human consciousness, has lead me to realize that the future becomes what you want it to be, and you attract what you want when you are determined to go for it. And with this knowledge, I can confidently take that last step into a new chunk of my life where I take charge and control the direction as the future unfolds.

'

1 INTRODUCTION''

From crowdsourcing, a way to gather ideas and input from an online community, a mechanism to fund creative projects online has emerged: crowdfunding. On one of the first platforms, Sellaband, which launched in 2006 fans could donate small amounts to artists that they liked, and in this way artists could raise the amount they needed to record a professional studio album. This way, instead of just going to a few record labels and risking rejection, artists could now ask their fans to pitch in in realizing their dream, while also maintaining complete ownership over their end product. This notion of gathering relatively small amounts of money from a large group of people caught on quickly and nowadays people in many different industries are trying to reap the benefits of crowdfunding to realize their dreams. Through crowdfunding platforms, entrepreneurs (also referred to as creators, project initiators) post a short - usually with a duration of around 3 minutes - video in which they let the public know how much money they intend to raise, how it will be spent and how backers (or donators, pledgers) can help in achieving this goal. The funding period typically lasts between 30 to 90 days. Crowdfunding knows five basic models: 1) donations 2) rewards 3) preselling 4) lending or 5) equity based. The latter two are more complex and subject to stricter financial regulation, the first offers solely an intrinsic benefit, and the reward and preselling forms offer a combination of intrinsic as well as some type of reward or product in return.

Over the past few years crowdfunding continues to grow at impressive rates. In just 3 years’ time the overall global market has grown more than 5 times the value of the $1 billion market size estimated in

2010 (see Figure 1)

(Massolution, 2013). While the collection of funds through smaller sums is not an idea that is entirely new to the world,

the Internet and

standardized platforms have made crowdfunding an entirely new industry on its own. It can be positioned among the early-stage financing mechanisms. A few exceptionally popular projects have even managed to raise over a million dollars, but the mode of projects on Kickstarter raise between $1,000 - $10,0001.

In the growing field of literature on crowdfunding, a variety of factors leading to success have been studied, as well as motivations to participate, and the role of personal and online social networks. The world’s most well-known crowdfunding platform is currently US-based Kickstarter, which employs a reward-based

all-or-nothing model of funding, meaning that the owner of a project only gets paid out if the

targeted amount is reached, otherwise funds are returned to investors. Many of the peer reviewed articles cover Kickstarter, because of the vast amount of data that has been collected since its launch in 2009 covering a wide variety of categories.

1.1 Crowdfunding' and' the' creative' industries:' some' more' successful' than'

others?'

While Kickstarter as a platform holds a success rate of around 42%, the top performers in terms of success rate are Music, Dance and Theater, with success rates of 54%, 64% and 69% respectively2 (also see Figure 2). Upon further inspection of the success rates of individual categories my attention was drawn to the fact that on two different crowdfunding sites, Kickstarter and the Dutch platform Voordekunst, the top 5 most successful categories returned three of the same categories: Dance, Music and Theater (and in the case of Voordekunst also the category Musical Theater), see Figure 2. The largest platform for the arts in the Netherlands shows even higher success rates for these categories; ranging from 78 – 92%. As a platform Voordekunst holds a success rate of 74% in a variety of artistic categories3. Since its launch in 2009, it has helped 969 projects successfully reach their goal, from a total of 52,635 donators, cumulatively raising nearly €5.3M4.

So what do these categories have in common? If I had to give them an overarching label, it would be that they all belong to the category ‘performing arts’. On Kickstarter, performing art projects make up 24% of all projects on the site, on

1

About 61% of all Kickstarter projects fall into this range. Kickstarter is one of the world’s largest crowdfunding platforms. At the time of writing over $1 billion has been raised on the site.

2

Stats from Kickstarter website as per August 27th 2014.

3

The significantly higher rate of success of this Dutch platform compared to Kickstarter can in part be attributed to the European ‘way of doing business’ where failure is often seen as face loss, whereas in the US the motto is more “fail fast, and fail often”. Other European platforms tend to have higher success rates than their American counterparts.

4

Figure 2 Top 5 most successful categories on Kickstarter and Voordekunst compared

Voordekunst they account for 47% of the projects on the site. What’s further interesting is that they appear as top performers both on a large scale platform like Kickstarter as well as on the more small scale niche platform Voordekunst.

One of the measures that has been found to be significantly correlated to success is a project’s goal amount. The higher the amount sought, the less likely the project is to succeed (Mollick, 2014). This can clearly be seen when ranking the Kickstarter data on average dollars received per project (Figure 3). The top three categories receive five figure amounts, but their success rates range between 28 – 36%. As has been noted in other research, the categories Technology, Design and Games are a league apart when it comes to crowdfunding (Marom & Sade, 2013; Mollick & Kuppuswamy, n.d.) .

The success rates of the projects raising larger amounts can in part be attributed to the larger goals they have, however it is still remarkable that the performing arts are more successful than all the other categories that on average raise lower amounts (Figure 3).

Category Projects-launched Success-rate Category Projects-launched Success-rate 1Dance 2,146 69% 1 Dance 24 92% 2Theater 6,775 64% 2 Photography 74 88% 3Music 32,224 54% 3 Music 296 85% 4Comics 4,513 50% 4 MusicalBtheater 55 80% 5Art 13,788 47% 5 Theater 139 78% Platform)overall 172,115 42% Platform)overall 1,094 80% PerformingBarts 41,145 PerformingBarts 514 %)of)all)projects 24% %)of)all)projects 47% Kickstarter Voordekunst StatsBasBperBJulyB21,B2014 StatsBasBperBAugustB27,B2014

Figure 3 Kickstarter categories sorted by average number of dollars per project

Could this simply be a coincidence? Or is there something about these projects that gives them a winning edge over other categories? And if the performing arts are more successful than others, what can we discover about them that makes them more so than the rest? This leads me to the main research question to be answered in this thesis:

“What characteristics of performing arts projects make them more successful than other categories on crowdfunding?”

To answer my question I will perform a case study on the Dutch platform Voordekunst. The nature of the research will be quantitative and descriptive, taking a closer look at a self-collected database of around 1,000 projects. Few published articles have touched on the structures and mechanisms of cultural crowdfunding projects, not to mention zooming in even closer on the categories of Dance, Music and Theater. The aim is to discover how the characteristics of these categories collectively compare to othe project categories as a whole. Lessons learned from the results could provide a better understanding of the mechanisms of crowdfunding and perhaps bear insights into how creators in different categories could learn from these projects.

Category Avg+dollars+ per+project Total+projects Percent+of+ all+projects Success+ rate 1 Technology 0000000000025,900 00000000000000007,587 4% 28% 2 Games 0000000000022,021 000000000000012,392 7% 35% 3 Design 0000000000020,364 000000000000010,029 6% 36% 4 Comics 000000000000006,802 00000000000000004,641 3% 50% 5 Film0&0video 000000000000005,988 000000000000039,774 22% 40% 6 Food 000000000000005,414 00000000000000009,875 6% 32% 7 Fashion 000000000000005,204 00000000000000007,976 4% 27% 8 Theater 000000000000004,028 00000000000000006,906 4% 63% 9 Music 000000000000003,774 000000000000032,868 18% 54% 10 Dance 000000000000003,397 00000000000000002,199 1% 69% 11 Journalism 000000000000003,025 00000000000000001,544 1% 30% 12 Art 000000000000002,983 000000000000014,192 8% 46% 13 Publishing 000000000000002,861 000000000000019,829 11% 31% 14 Photography 000000000000002,750 00000000000000005,677 3% 32% 15 Crafts 000000000000001,614 00000000000000002,299 1% 31% Stats0as0per0September023,02014

The remainder of this thesis is structured as follows. Chapter 2 is a literature review, which provides the theoretical framework that serves as a basis for understanding the dynamics and mechanisms involved in crowdfunding. The literature review is constructed of three parts; beginning with the basics of crowdfunding, then a section dedicated to the stakeholders, after which follows a discussion of research dealing with success factors. In Chapter 3 the role of the Internet as well as several factors in the cultural sector that have helped lay the foundations for crowdfunding activity are discussed. Chapter 4 then lays out the conceptual model to be used in analysis. In Chapter 5 the Methodology is described and in Chapter 6 the collected data is analysed. Chapter 7 dives deeper into the data analysing whether the way that a project description is phrased has a relation to the outcome of the project, an Chapter 8 concludes on the research question and discusses areas for future research.

2 LITERATURE'REVIEW'(THEORETICAL'FRAMEWORK)''

Crowdfunding is a prevalent topic in a variety of research streams. From business and economics, human computer interaction, social psychology and marketing to politics. The research being conducted aims to better comprehend the reasons for its existence as well as how project initiators, backers and platforms themselves can learn from what works and what doesn’t. This review of the literature is structured as follows; an introduction of what exactly crowdfunding is and its different business models, followed by stakeholder motivations to participate, the role of social networks, and determinants and patterns of success that have been identified.

2.1 Crowdfunding'101'

The origins of crowdfunding can be found in crowdsourcing, which is a conjunction of the words “crowd” and “outsourcing” coined by Jeff Howe and Mark Robinson, editors of Wired Magazine. Crowdsourcing is typically used by large companies that create an open call to a large network of unknown potential collaborators to work on or co-produce a solution to a problem. There is usually no monetary reward offered for these contributions. Instead, these contributors experience intrinsic benefits of contributing to a community (Howe, 2006). This collaboration of large groups, often dispersed over various regions around the globe, is made possible through the Web 2.0, which is not just a static information page but a platform which allows for interactive environments where users can interact and share ideas and media online (O'Reilly, 2005).

The first platform that utilized an open call to the public for the solicitation of capital was SellaBand, which launched in 2006 and was founded by three Dutch partners; Johan Vosmeijer, Pim Betist, & Dagmar Heijmans. Through their platform, fans – or “believers” as they are called on the platform – could collectively contribute to independent artists, and therefore allow the band to collect the $50,000 that would be necessary to produce an album. After some initial success, helping 43 artists achieve their target, the site was forced to file for bankruptcy in February 2010, only to go back online just two days later through the acquisition of the company by German investors. The original business model was not profitable, it relied too much on the traditional production of an entire album (consisting of approximately 12 songs) and the release of the music via CD’s, which was simply not fit for allowing smaller bands to gain popularity in a cost effective way. The revision of the business model along with new funding allowed the site to continue, since the idea was appreciated by both fans and

artists (Wired, 2010). With the decline of the importance of major record labels, as well as the decreased options artist find for borrowing money or forfeiting a stake in their intellectual property, crowdfunding helps democratize the access to finance as well as reduce market failure, by immediately connecting artists with their audiences (A. Agrawal, Catalini, & Goldfarb, 2011) . Since then, an explosive growth of platforms has been observed, many facilitating the funding of projects in specific niches varying from literature to dance performances, technology to food, and even housing projects. The largest platform – and the one that has been the source of much empirical research – is U.S. based Kickstarter.

So what is crowdfunding exactly? Ordanini et al. (2011) define crowdfunding as

“a collective effort by people who network and pool their money together, usually via the Internet, in order to invest in and support efforts initiated by other people or organizations.” referring to an earlier publication of Ordanini (2009) in the Wall Street

Journal. Belleflame and colleagues (2013) expand the definition by referring to the types of returns/rewards investors might receive: “Crowdfunding involves an open call,

mostly through the Internet, for the provision of financial resources either in the form of donation or in exchange for the future product or some form of reward and/or voting rights”.

All new-to-the-world developments are subject to a period of trial and error until a dominant paradigm emerges. The same is true for crowdfunding, as we saw with the adaptation of the SellaBand business model. Currently there are five distinct business models for crowdfunding platforms: donation, reward, pre-selling (generally categorized under reward based), loan and equity based (Hemer, 2011; Lambert & Schwienbacher, 2010) . These business models each have different ‘returns’ for investors and thus each attracts different types of investors. Owing to the type of return each structure is increasingly more complex. Giudici et al. (2012) examine the place that each of these forms takes in the larger scope of entrepreneurial finance. While the latter two involve monetary returns, the first two (donation- and reward based models) are especially interesting for researchers since investors are motivated by something else than just money. Belleflame and Lambert (2014) use price discrimination theory to discuss the circumstances that dictate which form an entrepreneur would opt for, concluding that for projects where the capital requirements are relatively low, the reward based scheme is preferent, and the profit sharing models prevail for projects where the capital requirements are generally higher.

In the following section I will further explore the motivations for participating in crowdfunding, discussed from the perspective of each stakeholder; the creator, the donator, and the platform. As you will notice, the emerging field of crowdfunding takes insights from many different areas of research; finance, economics, psychology, sociology, computer sciences and philanthropy to name a few.

2.2 The'stakeholder'dynamics'

In this section I will narrow down to the three key stakeholders as aforementioned, specifically within the realm of reward based crowdfunding.

2.2.1 Creators'

One of the first exploratory studies into the reasons to participate from a creator’s perspective finds that beyond the initial goal of collecting money, there exist several other reasons for soliciting for funds via crowdfunding. In a qualitative study, Gerber, Hui & Kuo (2012) performed 11 interviews with entrepreneurs and funders on three popular crowdfunding sites; Kickstarter, RocketHub and IndieGoGo. Their preliminary findings conclude that creators (entrepreneurs) are motivated to participate in crowdfunding in order to: 1. raise funds; 2. establish relationships; 3. receive validation;

4. replicate successful experiences of others; and 5. expand awareness of work through social media. These points indicate that crowdfunding can be used to gather

resources to progress as well as its use as a marketing tool for a product that is not yet on the market.

Building on theories of entrepreneurship, which emphasize the role of teams and community building, Hui, Greenberg & Gerber (2014) qualitatively study 47 entrepreneurs partaking in crowdfunding. The outcome of their analysis is that entrepreneurs rely on the crowd to 1. get advice; 2. design their campaign; 3. get

feedback; 4. publicity and 5. shipping rewards. Motivations to participate lie in the

perceived mutual value being achieved. Similar to raising funds from venture capitalists, crowdfunding entrepreneurs also receive some sort of network expertise and guidance through the alliances that are created.

Agrawal, Catalini & Goldfarb (2013) state that for early-stage ventures there are just two primary reasons for an entrepreneur to choose crowdfunding to collect the funds he is in search of: 1. lower cost of capital; and 2. access to more information. The lower cost of capital can be attributed to three factors. First, better matches between entrepreneurs and investors can be made, since the possible investors can be from much farther away than just the near local geographic locations. Secondly,

entrepreneurs can bundle rewards that early stage investors that are not specifically interested in financial returns appreciate (early access to products, extra information, and other merchandise or tokens of appreciation). And so another barrier to convincing investors to help in the funding is lowered. And thirdly, the information provided. Crowdfunding generates much more information than traditional finance forms, via the platforms the interest from different investors as well as recommendations on how the product could be improved is given. The access to more information can even serve as a clever marketing tool, measuring the demand from a group of end users before a product is even brought to the market. Hardy (2013) uses microeconomic theory to describe different ways of discriminating a crowd in order for a creator to reap the maximum benefits of the crowd. This theory suggests that general as opposed to

individual incentives deserve more investment on the entrepreneur’s part, when the

crowd is likely to be larger, as the value of this ‘public good’ then increases as there are more backers contributing. Further exploration of public good in an empirical setting can be found in the work of Qiu (2013).

The benefits of crowdfunding have also been described as being similar to Macht & Robinson’s (2009, p. 189) framework of benefits that business angels can bring to an entrepreneur: 1. Helping to overcome funding difficulties; 2. Provision of

contacts; 3. Facilitation of further funding; and 4. Involvement. Adding to this framework

a fifth benefit: 5. Limited/no loss of control and ownership, crowdfunding could even reduce the time and effort involved in negotiating contracts, since there are hardly any complex contracts involved and investors tend to be less sophisticated, requiring less information upfront (Macht & Weatherston, 2014) .

On the contrary, Macht & Weatherston (2014) also argue that several downsides exist in crowdfunding, namely: 1. Intellectual property protection (due to the disclosure of information about future products/service); 2. The ‘all-or-nothing’ funding model, which is most commonly employed by platforms (put in place to make entrepreneurs think about setting a realistic funding target, because when this goal is not met, all the pledges are returned to the initial investors); 3. High transaction costs/costs of publicizing the opportunity; 4. Managing a potentially large number of shareholders from different cultures and languages; and 5. Investors may not have the understanding to provide the entrepreneur with emotional support in a way that an angel might. By being aware of the advantages and drawbacks, entrepreneurs can be better prepared. Hui, Gerber & Greenberg (2012) agree that crowdfunding is not ‘easy

underestimated. Through interviews with project creators on different platforms, they discover that a crowdfunding campaign requires understanding the opportunities and responsibilities, developing campaign material, testing this material and early product prototypes, marketing the campaign, executing the outlined goals, and contributing expertise back to the crowdfunding community.

Crowdfunding has conveniently arisen in a time where access to bank finance and others, such as institutional grants, have been constrained. Entrepreneurs from different industries see this as a way of replacing or supplementing traditional forms of financing. Even when the project does not succeed – as the majority of projects still tend to do (on Kickstarter 58% fail) – the experience of participating and engaging in crowdfunding can be a rewarding one. Entrepreneurs who experienced public failure were able to learn from and adapt their campaign and eventually relaunch it (Greenberg & Gerber, 2014) . Among the creators who relaunched their campaign, the same success rate is still observable as initially, but owing to the fact that they first belonged to the ‘failed’ group, a larger proportion now passes as successful.

In sum, the benefits of opting for a crowdfunding campaign are multisided; exposure, validity and knowledge to name a few; and the efforts involved in completing such a campaign should not be underestimatd.

Creators should therefor inform themselves about the efforts involved in planning and executing a campaign. Luckily a lot of information is available on crowdfunding platforms themselves, online communities, as well as real life events that discuss current topics as experts and people interested in the field get together to have topic discussions on specific needs of different kinds of campaigns.

2.2.2 Donators'

Understanding the motivations of those that donate to these projects is less apparent since the return on investment is often not measurable in monetary returns, which is the case in reward-based crowdfunding. To better understand these more intrinsic motivations I would like to point out the work of Bekkers & Wiepking (2010) on the mechanisms that drive charitable giving. An extensive literature review of over 500 articles has led to the distillation of eight mechanisms that drive charitable giving, see Figure 4 for an overview of the mechanisms and the ways that different types of actors interact. All eight of these mechanisms can be found in the workings of reward based crowdfunding.

Figure 4: Eight mechanisms that drive charitable giving (Bekkers & Wiepking, 2011, p. 928)

Several studies have explored the motivations of donators that contribute to crowdfunding projects. Agrawal et al.(2013) identified at least four different reasons why people give to donation or reward based projects: 1. Early access to new products;

2. Community participation; 3. Support for a product, service, or idea (philanthropy); and 4. Formalization of contracts. Gerber et al. (2012) find similar incentives for funders

to donate: 1. seek rewards; 2. support creators and causes; 3. engage and contribute

to a trusting and creative community. In all the literature studied, the only factor that

came forward that might deter somebody from donating to a project is the distrust of

the creators’ use of funds (Gerber & Hui, 2014) .

Mollick (2014) goes in further on the way that funders respond to the signals of quality and herding effects that can occur, implying that these are important aspects that project creators should pay attention to when designing their campaign for maximum impact.

Studying around 100 pitches of donation-based projects in online journalism, Burtch, Ghose & Wattal (2013) drawing upon theory of “social information” find evidence of substitution. This is signaled by donators that are largely being driven by

altruism when contributing to projects. With this substitution also comes a ‘crowding

out’ effect, that is to say the more that other donators are observed to contribute, the amount an individual is inclined to contribute drops. This is explained by the relative

utility (benefit) that the individual gains from donating, which declines when the share they contribute seems to be diminished by other contributions, something that is also observable in work by Kuppuswamy & Bayus (2013), Hardy (2013), Belleflame (2010) and Qiu (2013). Because there are no direct rewards offered for contributing to donation-based platforms, altruism is found to be the leading motive to contribute to a cause. In the case of reward-based crowdfunding this may also prevail to a certain extent, in conjunction with other motives such as helping a friend or the rewards offered. Hardy (2013) attributes this to the ‘warm glow’ effect that contributors experience by participating.

Depending on the motives of the creator to start a campaign, the design of the campaign can be tailored to attract the right types of funders. Financing and non-financing consumers have been identified as valuable to a project’s success (Quero, Ventura, & Santoja, 2013) . Financing consumers of the project are primarily motivated by 1. The experience of being part of something; 2. Showing commitment to a certain

project type or method of production; and 3. The advance purchase of the product.

Although the non-financing consumer does not himself donate money, he can act as an important advocate or ambassador of a project, in this sense contributing to and benefitting from recruiting financing consumers and having the project realized.

Thus, for a large part altruism plays a key role in the motivation of donators to participate. That in conjunction with being part of a community, advocating certain causes and therefor gaining access to new products that otherwise would not be realised.

2.2.3 Platforms'

Platforms are the potential matchmakers between projects and investors. Generally they incorporate a for profit business model, taking a fee of around 5% of the total amount raised in successful projects. Some platforms even charge a fee before a project is allowed to go live, ensuring that the creator has the incentive to put in a minimum amount of work and meets certain quality criteria. And from a business perspective this makes sure that they also profit from projects that fail to meet their funding target. While Kickstarter and Indiegogo allow for a range of about 13 different creative categories, other platforms are more niche focused. For example the Dutch platform TenPages.com, where authors can submit excerpts from a proposed book and get funded, US based Seed&Spark for filmmakers, or the UK’s Spacehive for smaller community projects.

Most platforms incorporate an “all-or-nothing” return rule, which means that only upon reaching the stated goal, do the project creators receive their funds, and otherwise they are returned to the respective investors. In contrast, IndieGogo for instance also gives creators the option to fund their project via “direct donation”, in which case the creator will receive all accumulated funds irrespective of whether the goal has been reached. In an experimental setting these two models have been tested against each other, but findings were inconclusive as to which model is better for a platform’s marketplace. Two suggestions from this experiment indicate that if a platform wishes to place high-risk high-reward projects, they would likely benefit more from the return rule, and if they wish to place an emphasis on projects with a high likeliness of succeeding, the direct donation model would better serve this goal (Wash & Solomon, 2014) . As a result of this inconclusiveness flexible funding is offered on increasingly more platforms, upon reaching say 80% of the targeted amount by the time the campaign has closed, the creator is given the chance to readjust his goal amount in order to meet the success target and prevent forfeiting the accumulated money.

Belleflame and Lambert (2014) identify and discuss the multisidedness of crowdfunding platforms, their strategies, and competition among the platforms. The business opportunity for these platforms arises from the three criteria stated in an article by Evans et al. (2011) on platform economics. First, distinct groups of customers are linked: the creators and the funders. Second, the perception of value is dependent of both parties participating. And in the third place, the platform acts as an intermediary to more efficiently facilitate coordination between the different groups than if they were to seek out each other on their own terms, most notably they reduce costs and increase success prospects. Also, the authors note that participants on both sides care about what the others within their group do; while positive effects exist for contributors for instance mimicking others’ pledging patterns, this is in contrast to the possible negative effects that are generated among entrepreneurs through increased competition due to more projects on a site competing for the same funds. However, as we will see in 2.3 these potential negative effects amongst entrepreneurs can be turned in to a positive thing through reciprocal effects, thereby even creating a kind of sub-community (Zvilichovsky, Inbar, & Barzilay, 2013) .

Braet and Spek (2010) studied crowdfunding for the film industry in Flanders. They conclude with two business implications for film industry platforms; the first being that crowdfunding platforms should be seen as a “marketing lever” to raise attention, after which more conventional forms of funding ought to be sought or the incorporation

of professional parties can take on the project. The second implication is that due to the high overall costs of funding a film such as a documentary project, seekers of funds should apply “phased funding” and incorporate milestones for each segment they wish to achieve.

Kuo & Gerber (2012) even take on a perspective that crowdfunding can be seen as a ‘creativity support tool’, which allows more people to participate in the creative economy.

In essence, as Ronald Kleverlaan (a Dutch expert on crowdfunding) pointed out, a crowdfunding platform is nothing more than a webshop; it is a website with a payment system5 (Kleverlaan, 2013). Platforms benefit from helping projects succeed, which in turn increases their revenues, and therefor platforms should do everything in their power to facilitate matchmaking. Platforms could be a more proactive matchmaker for instance by proposing investors for certain projects based on past donating behaviors (An, Quercia, & Crowcroft, 2014) . By better understanding the interactions that take place between different actors and the resources that are exchanged (from

resource exchange theory often applied in marketing) platforms can be optimized to

ease such matchmaking (Greenberg, Hui, & Gerber, 2013) . This can be achieved by coming up with and implementing features that bolster social interactions, strengthen identification, and create a heightened sense of community (Colombo, Franzoni, & Rossi Lamastra, 2013) .

The business case for platforms can therefor be attributed to the commission they make off of projects that are featured the platform and the participation of the creators and funders that are necessary to sustain this.

2.3 Patterns'and'determinants'of'success,'what'has'been'researched'so'far?'

Now that much has been said about who participate and why, this section discusses quantitative empirical research on crowdfunding. Patterns identified in donating behavior, campaign presentation, and other factors are discussed.

Most research on the determinants of success draws upon data from Kickstarter. Being the world’s largest and most well-known crowdfunding platform to date makes it a benchmark for studies in crowdfunding. The public availability of data facilitates generalizations and comparisons of research performed by different researchers. Since crowdfunding is just growing out of its infancy, what underlying patterns have been discovered so far?

5

Kleverlaan gave a guest lecture for the master course International Entrepreneurship at the University of Amsterdam in November 2013 as part of the master’s programme in Business Studies.

Momentum. Crowdfunding campaigns are subject to a U-shaped curve in the amount of backers the campaign receives over time. This is because the momentum at the beginning of a campaign is generally generated by friends & family and then the ‘crowd’ must somehow be reached to keep up the momentum and lead a project to success. Using social psychology theory on the bystander effect, it can be said that the toughest moment for creators is the halfway mark of their campaign, when the so called ‘bystander effect’ has contributors thinking that ‘somebody else’ will probably help instead of deciding to back the cause themselves. Near the end of a campaign however, new buzz is generated through project updates and the near finish, which excites potential backers, can eventually lead a campaign to completion. This sense of urgency is something that is also noted in online auctions (Kuppuswamy & Bayus, 2013) . This phenomenon explains why most of the projects that fail, do so by a long run. On Kickstarter, of the projects that reach more than 20% of their funding, 80% succeed (Mollick, 2014). The same can be observed on other crowdfunding platforms. Knowing this is important for project planning, since it gives project creators a guide to the target amounts they should be raising each week and the importance of setting a realistic goal.

Setting a realistic goal. The higher a project’s goal, the less likely it is to achieve this goal (Mollick, 2014). No surprise here. Having a higher goal means a creator has to work harder to trigger his personal network of family and friends, and also in attracting many others beyond them.

Especially when goals are extremely high (more than tens of thousands of dollars) applying phased funding can help in breaking up a large project into chunks that are more easily achievable and have realistic milestones incorporated, which Braet and Spek(2010) recommended for the film industry in Flanders.

Presentation. Via a title, catchy cover image, a short video and a written description of what the project entails a creator can reach out to the crowd to solicit a call to action. The way a project creator presents his or her project should entice potential donators to want to contribute to their cause.

Making use of Natural Language Processing techniques, researchers have been able to analyze entire descriptions and coded them in order to look for underlying patterns (Lin & Viswanathan, 2013; Mitra & Gilbert, 2014) . Mitra & Gilbert (2014) study the language used in Kickstarter campaigns. They build on theories of persuasion,

which play a large role in marketing and consumer behavior for instance. Reciprocity is a central principle in social psychology, “the sense of obligation to return a favor after receiving one” (as cited in Mitra & Gilbert, 2014 p.2) and is why creators give something back for a donation. Conjunctively, the use of positively phrased sentences (e.g. “pledgers will receive”) as opposed to phrases that exude desperation/lack of assurance (e.g. “we have not yet been able to”) (Mitra & Gilbert, 2014) have a strong relation to the success of a project.

Marom and Sade (2013) count the number of self-mentions of the entrepreneur in the ‘about section’. In artistic categories the number of self-mentions is considerably higher than in technological categories. As the authors comment, in technological projects it seems as if the focus is more on the horse (the business idea), and in artistic projects the focus is more on the jockey (the entrepreneur). A limitation of their findings is that a bias occurs because an entrepreneur could also refer to him/herself as “I” or “we” for instance, because in their analysis only exact matches to the entrepreneur’s name were taken into account. Re-running such an analysis with these terms included might therefor reveal different results.

The role of social networks. An important aspect of crowdfunding is that it leverages off of an individual’s personal network, which opens up to a wider avenue of online friends and acquaintances and ultimately to people who do not have any direct ties to the creators of the project. These are the people that will make crowdfunding a sustainable mechanism for project creators who wish to draw upon the crowd for continuous rounds or projects.

Although not all contacts in one’s network lend themselves for donation, some may be a valuable asset as promoter of the project or both (Qiu, 2013).

Both Mollick (2014) and Giudici, Guerini & Rossi-Lamastra (2013) find that individual social capital and the underlying project quality are of importance in attaining the targeted goal. Social capital is expressed in the number of contacts on social media such as Facebook and Twitter. Exponential increase in the amount of Facebook friends results in the doubling of the chances of succeeding. An entrepreneur with 10 Facebook friends is found to have a 9% chance of succeeding, one with 100 friends a 20% chance, and one with 1,000 is 40% likely to succeed (Mollick, 2014).

Giudici, Guerini & Rossi-Lamastra (2013) further study the effects of ‘individual’ (exclusive) social capital and ‘territorial’ (locally shared) social capital. Individuals pledging early in the project are mostly individual capital and give off some sort of

signal of quality to other potential backers. But when the locally shared network is large, this signaling effect is somewhat diminished. An adverse selection problem occurs in favorable local circumstances, where good-quality projects may not even need to ‘tap into the crowd’ of the Internet because they are easily able to raise funds locally. Thus it can be stateted that it is more important for project creators to have large individual social capital in order to succeed.

Marketing. After the campaign has been carefully prepped for launch it needs to be appropriately marketed. Online networks, updates and study of backer behavior over the course of projects reveal some insights:

Because of the U-shaped momentum of the pledges a project receives, early promotional activities are important to overcome the ‘valley of death’. A strong relation is found between a project’s final outcome and the project’s early promotional activities (Etter, Grossglauser, & Thiran, 2013) . Using the buzz generated on Twitter and tying it to the amounts pledged to a project, Etter and colleagues are able to determine if a project on Kickstarter will be successful within the first four hours after launch (with an accuracy of more than 76%). Additionally, it is important that these activities come from multiple sources (Lu, Xie, Kong, & Yu, 2014) . A comparison of a mix of different promotional outlets found that being ‘featured’ on – in this case Kickstarter – the platform’s front page had the greatest positive effect on the number of pledges (compared to media mentions, updates, and Twitter mentions) (Qiu, 2013).

Furthermore, it is advisable to make targeted posts to reach a certain group of potential backers instead of general pleas (J. S. Hui, Gerber, & Gergle, 2014) .

Updates. In order to gain more backers, project creators engage with the crowd via updates and posts on social networks. The relationship and effects that updates and backers have on one another has been documented in several works.

Not entirely surprising, project updates are found to have a positive influence on the outcome of a project; leading to more support in the form of project donations, especially towards the end of a campaign (Kuppuswamy & Bayus, 2013) . Successful projects show a general tendency of having posted more updates than failed ones. It can be argued that this is a result of succeeding, that a project owner would be more inclined to post after reaching subsequent milestone in his campaign, but it can also be seen as a cause for generating more donations. Xu et al. (2014) found that the success rate of projects without updates was 32.6%, whereas those that did have updates had a success rate of 58.7%. Furthermore they also identified different types of updates (social promotion, progress report, new content, reminder, answer to a question, new reward, and appreciation).

Reciprocity. One tactic that creators have shown to use to attract new funds is by returning the good deed and pledging in others’ projects. Analysis of the backing behaviors of project creators in approximately 70,000 Kickstarter projects reveals that those who backed others’ projects had a significantly higher likeliness to succeed in raising their own goal. In addition they tend to raise higher amounts and receive more total donations than projects whose creator did not reciprocate. It is even appears that a type of sub-community is created by these actions, where creators actively back other creators’ projects within their community(Zvilichovsky et al., 2013). This seems like a rational outcome, yet one can ponder what the net result of donation to others’ projects in order to receive more funds for the own project is. In any case, an added benefit could be the marketing power of having such a backer-creator act like an ambassador of the project.

Home bias. In a study of 34 projects on Sellaband, which together account for 73% of the $2.3M raised on the website, Agrawal, Catalini & Goldfarb (2011) explore the role of distance in financing music projects. They find that distance plays less of a role when deciding to invest in a crowdfunding project, however investors near the project initiator tend to invest relatively early and are less responsive to the decisions of other backers. It is thought that this is likely due to the personal connection they may have with the project initiator. Creating a market for the product is overcome through crowdfunding, thereby making it easier to raise financing from distant strangers.

Mollick(2014) also argues that the types of projects generated in a geographical area reflect the types of creativity that are prevalent in a certain area (and increases the chances of success in said area) naming country music in Nashville, Tennessee as an example. This is where the highest saturation of the market is, that in the end will consume the project.

Overall, for a campaign to be successful the creator needs to set a realistic goal, have a positive tone, be very dedicated, carefully plan and interact with his proposed crowd through updates for instance and create an online buzz. Even by reciprocating via donation to other projects can increase the chances of success. And project types that are prevalent in a certain geographic location are also likely to succeed in that location, except when that market becomes oversaturated. These patterns do not represent an exhaustive overview of all the research on crowdfunding, however they do provide a frame of reference into the internal workings of the crowdfunding phenomenon.

3 SO'WHAT'ABOUT'THE'PERFORMING'ARTS?''

Now that the literature on crowdfunding has outlined the stakeholders and certain mechanisms at play, and given the global question that guides this thesis: “What

characteristics of performing arts projects make them more successful than other categories on crowdfunding?” In this chapter I provide you with a basic backbone for

understanding some of the current developments in the performing arts, both on a global scale as well as a more focused view from the Netherlands and how it ties in with crowdfunding.

3.1 A'word'on'the'performing'arts'in'the'‘Age'of'Internet’'

With respect to the way that the Internet has helped overturn traditional structures in the production, distribution and consumption of performing arts (and in fact any art form) Ben Cameron illustrates beautifully in a Ted Talk given in February 2010, titled ‘The True Power of the Performing Arts’. I give you here an excerpt from the transcript of his talk that emphasizes the role the Internet has played in the cultural sector:

…“Chris Anderson, someone I trust you all know, editor-in-chief of Wired magazine and author of "The Long Tail," really was the first -- for me -- to nail a lot of this. He wrote a long time ago, you know, thanks to the invention of the Internet, web technology, mini-cams and more, the means of artistic

production have been democratized for the first time in all of human history. In the 1930s, if any of you wanted to make a movie, you had to work

for Warner Brothers or RKO because who could afford a movie set and lighting equipment and editing equipment and scoring and more? And now who in this room doesn't know a 14 year-old hard at work on her second, third, or fourth movie? (Laughter) Similarly, the means of artistic distribution have been democratized for the first time in human history. Again, in the '30s, Warner Brothers, RKO did that for you. Now, go to YouTube, Facebook; you have

worldwide distribution without leaving the privacy of your own bedroom.

This double impact is occasioning a massive redefinition of the cultural market, a time when anyone is a potential author. Frankly, what we're seeing now in this environment is a massive time, when the entire world is changing as we move from a time when audience numbers are plummeting. But the number of

church choirs, is exploding beyond our wildest imaginations. This group, others have called the "pro ams," amateur artists doing work at a professional level. You see them on YouTube, in dance competitions, film festivals and more. They are radically expanding our notions of the potential of an aesthetic vocabulary, while they are challenging and undermining the cultural

autonomy of our traditional institutions. Ultimately, we now live in a world defined not by consumption, but by participation.”… (Cameron, 2010)

Undoubtedly, crowdfunding neatly ties in to this notion of democratic “distribution” , “production”, “participation” and “consumption”.

3.2 The' status' of' the' performing' arts' in' the' Netherlands,' and' the' rise' of'

crowdfunding.'

What are some of the recent developments in relation to the performing arts and the subsequent birth of crowdfunding in the Netherlands?

The effects of the financial crisis are also to be felt in the cultural sector. Cultural budget cuts in 2008, and again in 2013 resulted in more people fighting for fewer amounts of euros in the cultural sector in the Netherlands. In 2013 the request for grants at Fonds Podiumkunsten (Fund for the Performing Arts) was almost twice as high as the number of requests that could be honored6. Starting from 2013, 70 cultural organizations will lose their subsidy for the period 2013 – 2016 (Haverkort, 2012); and in this same year 11 performing arts groups announced that they would cease to exist as a result of these circumstances (Kammer & van Lemt, 2013) . Although this may seem to produce a somber effect on the cultural landscape in the Netherlands, one could also ask what the role of the government is in the first place regarding the arts? And while many people think of crises as an unfair burden and an obstacle that must patiently be sat out until more prosperous times arise; crises can also be seen as an ocean of opportunities waiting for systemic change to emerge.

Through a different government funded instrument, Fonds voor

Cultuurparticipatie (Fund for Cultural Participation) the government seeks to embed

greater cultural participation in the Netherlands. One of the main tasks that this fund set out in its 2013 – 2016 policy is to stimulate cultural education, and thereby create a greater cultural breeding ground (Fonds voor Cultuurparticipatie, 2012). With this greater pool of people that want to creatively express themselves comes a task of

6

finding the resources necessary to do so. True to this day and age the Internet can facilitate in just that. Finding resources to learn skills, collaborate, and find investment; as has been laid out in the theoretical framework, crowdfunding allows for all of these.

The role of the government can thus be seen as one of enhancing the creative breeding ground in the Netherlands. A discernable gap was that these people now also had to be more entrepreneurial to realize their creative endeavors. Individuals and groups must try to seek out other ways of finance for their projects than of solely government and institutional grants. While crowdfunding may not wholly be able to create a self-sustaining environment for performing artists, it gives them greater independence and an alternative or supplementary source of finance – in combination with other elements of developing a project. It has even been suggested that traditional institutions can use crowdfunding as a type of radar for deserving projects to allocate reserved funds (Hemer, 2011).

One way that cultural entrepreneurship could be stimulated was by providing artists with a platform where they can connect with an audience and seek financial backing. With this vision in mind, Roy Cremers decided to bridge this gap by founding Voordekunst.nl.

3.3 The'why,'how,'and'what'of'Voordekunst.nl'

Voordekunst was launched in 2009 by Roy Cremers who, after several years of working at a few of Amsterdam’s popular museums and at Amsterdams Fonds voor de

Kunst (Amsterdam’s Fund for the Arts), came up with the idea for the platform. With his

initiative he wanted to stimulate cultural entrepreneurship and have artists and cultural organizations be less dependent of funds and government subsidies. In a newspaper article Cremers comments that by opting for such a medium it forces creatives to think about their marketing and communication, which will help them in securing an audience. He also notes that it is not simply a charitable site, the creator must actually enjoy presenting his or her project in a personal way and think about how he can connect and engage with his audience. More importantly he remarks that the donators will act as ambassadors of the project, when they talk about it in their personal networks for instance; so make sure to take the time to thank them properly (van der Veen, 2011).

The successes on Voordekunst.nl point to a promising future for artists. The website charges €140 for a project to be posted, and they perform an initial screening of the project. Only those that meet the minimal quality requirements will eventually be

posted on the platform. This way they maintain a focus on quality, which is also what government institutions look at in assessing a grant request.

The platform has raised well over €5 million passing its fourth year of existence, with over 1,000 successful projects collected from nearly 55,000 donators.

Just as in almost any business - it’s all about relations! This is especially true for larger crowdfunding campaigns. Project owners need to connect to actors that are capable of donating larger sums, they simply won’t make the cut if they’re relying solely on €25 donations from their personal network; it is in their extended network that they should actively seek those ‘bigger fish’ (PerformingArtsFundingCenter, 2014). This again coincides with the stimulation of entrepreneurship in the cultural segment.

Personal gain has little to with why people donate to projects on the site. According to Cremers it’s all about the “goodwill” factor (in Dutch: gunfactor) and their personal involvement with the project and/or its owner(s) (de Lange, 2013). In an article posted in Dutch newspaper NRC, Cremers jokingly referred to crowdfunding as “incrowdfunding” since the majority of funds were still coming from friends and family (Leijten, 2012). Tentative results by René Bekkers and Irma Borst of the Vrije Universiteit Amsterdam7 give support to the indication that many of the projects on Voordekunst receive their funding from family, friends and loose acquaintances. In order for crowdfunding to be a sustainable system for repeated funding, creators need to move beyond their intimate group of friends and acquaintances. Albeit that these close relations are a necessary first step and can act as a stepping stone to reaching the real ‘crowd’, people that will invest in a project because it matches their personal interests or for other motives would like to see a project realized. Friends, family and other extended acquaintances will gladly donate once, and maybe even two or three times to someone’s project, but if the project itself does not per se match their personal interests the largest part of them will not likely donate a fourth or a fifth time. On the other hand, by coming in contact with crowdfunding via a friend, donators might discover other projects that get them excited and want to donate. Cremers and his colleagues are therefore working hard on stimulating repeat donations and reinforcing relations among (existing) investors and creators8. One of the ways they do this is through partnerships with creative institutions across different Dutch municipalities.

7

Presented during the first Dutch academic research seminar on crowdfunding organized by the Vrije Universiteit, Amsterdam Business School and the Crowdfunding Hub on June 5th _{2014 in the Beurs}

van Berlage, Amsterdam. The theme of the seminar was “Crowdfunding in the cultural sector”.

8

From blog post Voordekunst October 24th 2014: http://voordekunst.wordpress.com/2014/10/29/wij-werken-aan-1/

4 CONCEPTUAL'MODEL'

Now that the literature on crowdfunding has been reviewed, background information on the performing arts and crowdfunding provided, and the subject of the case study has been outlined; to recap, the research question I have set out to answer:

“What characteristics of performing arts projects make them more successful than other categories?”

Just as technological projects have been compared to artistic projects (Marom & Sade, 2013) , in the analysis of the data the aim is to compare the combined category of Performing arts to the body of other categories on the platform Voordekunst. The theoretical framework discussed factors that have a relation to the outcome of a crowdfunding project. With four years of project data available on the platform, I will take a closer look to see how performing arts projects contrast to other projects on the platform based on the data that is publicly available on the platform. How do they differ, and can anything can be said about why they seem to be more successful than other project categories?

To analyse this, the conceptual model takes in to account the grouping of the body of collected project data by which category they belong to: either Performing arts or Other (Figure 6).

Figure 5: Conceptual model

“Classification is a powerful conceptual tool that is often not seen by practitioners.” (Shields & Tajalli, 2006: p.323) . By applying a label to a group of cases that seem to have some inherent properties that make them seem alike is a first step in uncovering underlying mechanisms that might produce these characteristics.

Performing*arts* (Dance,(Music,(Musical( Theater,(Theater)( Others* ((( ( Project* variables( (i.e.(Goal,( amount(raised,( social(indicators)( * Category((( (

5 METHODOLOGY'

The basis of my research is a case study of the crowdfunding platform Voordekunst, which has over 1,000 completed projects on it, of which nearly half belong to the Performing arts. It is therefore a good source to do an observational study of the characteristics of these projects. Case studies are commonly used in exploratory research, to help the researcher get a better understanding of the context and the activities in it (Saunders, Lewis, & Thornhill, 2011) . The platform as a whole and the collective self-imposed category of Performing arts serves as the case under scrutiny.

The nature of the research will be descriptive; the goal is to compare the categories belonging to Performing arts collectively, comparing the means, medians and distributions of the different variable values to projects that do not belong to this category.

Saunders & Lewis(2012, p.111) define descriptive studies as “research

designed to produce an accurate representation of persons, events or situations.”.

Descriptive research can be chosen when basic information does not exist in the literature (Shields & Tajalli, 2006: p.324) . Such is the case of the categorization of Performing arts. It also typically answers a “what” question. An advantage of this

Performing*arts* (Dance,(Music,(Musical( Theater,(Theater)( Others* (All(other(categories,(excl( Photography)( Variables:( Goal((€)( Raised((€)( Percent((%)( Donators((#)( Updates((#)( Likes((#)( Tweets((#)( Differences* All*projects* PloDed(for(relaFve(size(of(category(on(the(plaIorm( ( ( ( ( ( ( ( (

research method is that it is very accurate, yet a disadvantage is that is doesn’t address the causality and relations underlying an observed phenomenon; it simply serves to aid the researcher in gaining a better understanding of the phenomenon. Follow up research can then address any observations that seem necessary to further examine (Shields & Rangarajan, 2013: p.71) .

5.1 Research'design'

Based on the variables that are publicly available on the Voordekunst platform, Figure 6 shows the framework for analysis of project variables. The wordcloud in the figure represents all 24 categories on the platform, scaled to represent the number of projects in each category.

Taking this as a starting point, the analysis will then further dive into the differences between Failed and Successful projects between categories as well as the differences between Failed and Successful projects within each category as is portrayed in Figure 7.

5.2 Data'collection'

Data was crawled from the platform using Import.io9, a free data scraping program, which once ‘trained’ (i.e. given a few pointers as to what type of data is found on what

9 www.import.io Variables:! Goal!(€)! Raised!(€)! Percent!(%)! Donators!(#)! Updates!(#)! Likes!(#)! Tweets!(#)! Performing0arts0 (Dance,!Music,!Musical! Theater,!Theater)! Others0 (All!other!categories,!excl! Photography)! Successful0 Failed0 Successful0 Failed0 Differences0 Differences0 2.#Comparison#within#category# Differences0 Failed0 Performing0arts0 (Dance,!Music,!Musical! Theater,!Theater)! Others0 (All!other!categories,!excl! Photography)! Differences0 Successful0 Performing0arts0 (Dance,!Music,!Musical! Theater,!Theater)! Others0 (All!other!categories,!excl! Photography)! 1.#Comparison#between#categories#based#on#outcome#

part of a webpage) automatically collects the desired data and allows for it to be downloaded into an Excel spreadsheet.

Several crawls were performed in order to retrieve data from different levels of the website; namely the project summary pages, the project pages, and the creator pages. Using Excel the results from these crawls were then stitched together using the VLOOKUP command to form one file encompassing all crawled variables. Projects that had not yet finished were eliminated from the overview, leaving me with 1,091 projects.

One limitation was that the variable number of Tweets was embedded in such a way that the Import.io software was unable to obtain its values. This variable is therefore excluded from further analysis. Another limitation is that data on the projects’ duration, a variable that in many reports is found to be correlated to the success of a project, is not visible on the website. Nevertheless, for the other variables the crawl resulted in a large amount of data covering over a thousand projects.

6 ANALYSIS'

6.1 Success'rates'across'categories'

Overall, the different categories on Voordekunst are all very successful, with success rates ranging between 70 – 90% (Figure 8). In Figure 8 we see that the most successful projects on Voordekunst are the Performing Art categories (Dance, Music, Musical Theater, and Theater) as well as Photography.

Figure 8: Success rates across categories with at least 25 finished projects. Percentage after each category refers to the size of the category on the whole platform.

Figure 8 has been created to also reflect the amount of projects that raised a certain percentage of their targeted goal for each category. We can see that of the failed projects (those raising less than 80% of their goal) the majority never even got past the 25% mark, and only very few projects received much more than they had anticipated (the 125%+ category). This is consistent with other findings (Colombo et al., 2013; Hekman & Brussee, n.d.; Mollick, 2014)

6.2 Descriptive'statistics'

In order to get a scope of how the Performing arts score on the different variables that were crawled from the project pages, Figure 9 presents a summary table with means and standard deviations for each variable. It was chosen to exclude Photography from the comparison to Performing arts so that the contrast with the less successful project categories becomes more visible.

From Figure 9 we can see that Performing arts projects have lower Goal amounts (€717 on average), raise slightly less than Other projects (€183), and on average they raise 90% of what they set out to fund (whereas Other projects raise 84% on average). On the number of donators both categories are pretty much tied, with an

0% 20% 40% 60% 80% 100% Dance (2%) Photography (8%) Music (29%) Musical theater (5%) Theater (13%) Film & video (9%) Interdisciplinary (7%) Visual arts (10%) Letters (5%) 0 - 25% 25 - 50% 50 - 80% Adjusted (80 - 100) Success (100 - 125%) 125%+