Success Factors of Game Crowdfunding Campaigns
A quantitative study about the Kickstarter platform
(Master Thesis)
MSc Entrepreneurship
Student :
Khrisna Aria Putra
Student number: 10750053 / 2553160
Date:
30/06/2016
Acknowledgments
I would like to start by expressing my biggest gratitude to the people who supported me during the process of writing this Master thesis. First of all, I would like to thank Dr. Yang Song for her time supervising and helping me during this thesis period. Furthermore, I would like to thank my family and my friends especially Rizky, Nirwan, Inez, Alifa, Annie, Ondrej, Keanu, Richard, Adnan, Via, Amanda, Sandy, Obi, Rira, and Andika for the support and encouragement throughout my studies.
Abstract
Crowdfunding is an innovative and relatively new financial method that connects entrepreneurs and investor through the internet. This method allows the entrepreneurs to search and collect funds to finance their campaigns/projects and this raise the necessary funds by relatively small contributions from a relatively large number of investors. Game industry is one of the main player in the crowdfunding phenomenon which has unique characteristics that are interesting to explore. The question of what factors influence the success of game crowdfunding campaigns, i.e, reaching the funding target, is very important. This study provide a comprehensive view in factors influencing game crowdfunding success by focusing on entrepreneurs and backers trust, social networks and motives. Using logistic regression method, this research analyze 1.686 projects initiated in the USA and held between January 2014 and 2016 from crowdfunding platform kickstarter.com. I find that the video, number of backers, rewards level and funding goal positively influence game crowdfunding campaign success. Interestingly, the question of the social networks of the entrepreneurs has no significant influence. These results provide valuable insight into the success factors of crowdfunding campaigns, and give entrepreneurs and funders some advice on crowdfunding.
Table of contents
1.
Introduction ...3
2.
Literature review ...6
2.1 Crowdfunding...6 2.1.1 Advantages...7 2.1.2 Disadvantages...7 2.1.3 Crowdfunding models...72.1.4 The use of reward based crowdfunding model in game industry ...8
2.1.5 Success factors in reward based crowdfunding...8
2.2 Trust...9 2.2.1 Internal assessment...10 2.2.2 External assessment...10 2.3 Social networks...11 2.4 Motives...11
3.
Methodology...13
3.1 Kickstarter...13 3.2 Data collection...13 3.3.Variables.... ...14 3.3.1 Dependent variable...15 3.3.2 Independent variables...15 3.3.3 Control variables...17 3.4 Pearson correlation...183.5 Research model...19
4.
Results ...22
4.1 Descriptive statistics...22 4.2 Regression analysis...255.
Discussions ...29
5.1 Trust...30 5.2 Social networks...31 5.3 Motives...325.4 Theoretical and managerial implications...33
6.
Conclusion ...36
6.1 Limitations and future research...37
List of tables and figures
Table 1 Key variables...14
Table 2 All projects...22
Table 3 Unsuccessful projects...23
Table 4 Successful projects...24
Table 5 Correlation matrix...25
Table 6 Regression model 1-2...26
Table 7 Regression model 3-4...26
Table 8 Regression model 5...27
Table 9 Classification table...27
Figure 1 Theoretical framework...18
Figure 2 Outline framework...30
1.
Introduction
In 2012, Double Fine Productions launched an astonishing campaign for their video game Broken Age on Kickstarter. They successfully raised $3.45 million with more than 87.000 backers within a single month, much more than the original pledged amount of $400,000. This project was funded by a form of micro financing: crowdfunding. Double Fine Productions were one of the earliest game developers tapping into the crowdfunding field and are to date, one of the highest-backed crowdfunding projects in the category of video games. In today’s society, many entrepreneurs face difficulties in attracting funding to support their projects or ideas. Several studies indicate that funding is critical for starting a new venture (Gompers & Lerner, 2004; Kortum & Lerner, 2000). However, as Schewinbacher & Llarralde (2010) pointed out, it is hard to attract early stage funding for new ventures. In the early phases of a company, funding is typically provided by the founder, his friends and family. If these funds are short, the venture faces a funding gap (Collins & Pierrakis, 2012). This situation has been worsen by the financial market crisis (Mach, Carter & Slattery, 2013). Traditional funding methods, for example bank loans or venture capital investments, are not fond of investing in early stage new ventures (Schwienbacher & Larralde, 2010). To address this dilemma, crowdfunding can thus be a novel and attractive alternative for entrepreneurs to generate funding without having to use traditional sources of investment (Mollick, 2014).
Crowdfunding itself is defined as an open call for the provision of financial resources in the form of a donation or in exchange for the product in the future or another form of rewards that supports the initiatives for specific purposes (Belleflamme, Lambert, & Schwienbacher, 2014). Crowdfunding, in which consumers act as investors, draws inspiration from crowdsourcing and microfinance (Ordanini, Miceli, Pizzetti & Parasurman, 2011). Intriguingly, crowdsourcing refers to employ the crowd to obtain ideas, feedback, and solutions to develop corporate activities (Kleemann, Voβ & Rieder, 2008). As a matter of fact, the concept of asking for small amounts from a large number of people (crowd) was not a new practice. In 1884, American Committee was requesting funds for the Statue of Liberty’s pedestal. The famous Joseph Pulitzer was once initiated a fundraising campaign in his journal, The New York World, asking for some aid to get the required amount of $100,000. Eventually, more than 160,000 donors contributed making a total of $101,091 in just 5 months.
A number of online platforms have emerged as a crucial intermediary between the entrepreneur and investors. In contrast to traditional funding, entrepreneurs are able to bypass venture capitalists and angel investors by raising money directly from individuals through crowdfunding via the internet (Schweinbacher & Larralde, 2010). In 2014, over 16 billion dollars was raised across the platforms,
which represents a 167% growth and the industry expects this number to be doubled in 2015 (Massolution, 2013).
With the emerging trend of crowdfunding, the game industry has experienced significant developments in recent years: 1) Crowdfunding has allowed smaller publishers to access financial funds to turn their ideas into reality, 2) The market is more evenly balanced between the smaller, independent, crowdfunded projects and established publishers, 3) Crowdfunding is an immense way to establish and maintain customer loyalty and develop greater understanding with customer needs.
According to Mollick (2013) there are four different models in crowdfunding; 1) the patronage model, 2) the lending model, 3) the reward model and 4) the equity model. In game industry, the most commonly recognized models are the reward model and the equity model. However, this study will focus on the reward model since this is perceived as the most promising model in which funders receive a reward for backing a project (Mollick & Kuppuswamy, 2014). Furthermore, the reward model is simpler than the equity model because it does not require investors to hold any financial claim of the company, the industry risk and return potential. Our analysis will be applied to the largest reward-based platform “Kickstarter” which has collected nearly $2 billion and hosted more than 60,000 projects since their launch.
Prior studies only covered the dynamics of crowdfunding (Mollick, 2014; Balboni, Kocollari & Pais, 2014) in general but did not distinguish them among industries. As a representation of creative categories, the game industry has unique characteristic that are intriguing to explore. There is an increase in the number of crowdfunding campaigns and the number of people that are willing to contribute. Data from Kickstarter recently revealed that games category ranks number one for funds raised and success rate. The data shows that once a project is able to raise 20% of funds, the probability to become successful climbs to 80%. As mentioned above, Kickstarter has raised approximately $2 billion and video games account for more than a quarter of the immense figure. The role of crowdfunding in the game industry will continue to grow without doubt.
This study addresses the knowledge gap in the context of the game industry and demonstrates our understanding of the factors that affect the success of game crowdfunding projects in particular.
Crowfunding is still growing and the phenomenon is relatively new in the academic literature. Hence, it is an interesting research topic because there is much more to be discovered. In order to gain further insights in crowdfunding this project examines the factors that make crowdfunding successful,
focusing on game projects, as they have the highest success rates on crowdfunding platforms. Accordingly, this project want to analyze and answer the following main research question:
- What factors affect the success of a game crowdfunding campaign?
In order to answer the main research question, the following sub questions have been developed: - To what extent does trust affect the success of crowdfunding campaign ?
- What role do social networks play in the outcome of the crowdfunding campaign? - To what extent do motives affect the success of crowdfunding campaign?
The results of this study can be beneficial for both crowdfunding platforms (Kickstarter) and entrepreneurs who are searching for new means of funding. The results of this thesis will help them understand how they can increase the awareness of crowdfunding in society. On the other hand, it will give entrepreneurs an advanced strategy of the factors they need to consider to be successful in crowdfunding campaigns.
This study consists of the following components. First, I will give a comprehensive review of the literature about crowdfunding and game industry to date. In the methodology section, I present my data collection process and the statistical methods and models I have used. Then, the results of the analysis will be discussed. Furthermore, in the discussion part, the theory and results of the study will be shown. Finally in the last part, the limitations of this study and recommendations for future research will be described. ..
2. Literature review
The objective of this section is to provide a review of the crowdfunding literature. I will first present the definition of crowdfunding also its advantages and disadvantages. Then, I will discuss the characteristics and the model of crowdfunding campaigns. It should be noted that crowdfunding is still a relatively young phenomenon compared to traditional financial forms such as bank or equity financing. Hence, the number of published academic literature on the topic is still limited.
2.1
Crowdfunding
Crowdfunding is an emerging approach to search funding for entrepreneurial ventures instead of using traditional methods for funding. Whilst it is widely seen as a relatively new phenomenon, crowdfunding is considered to be a financial instrument in transition.
The phenomenon has been illustrated by several researchers (e.g. Ordanini, 2009; and Mollick & Kuppuswamy, 2014), who adopt a similar definition of crowdfunding: namely that individuals make contributions to projects through the Internet.
According to Ordanini (2009), crowdfunding is a collective effort by people who network and pool their money together, usually via the internet, to invest in and support efforts initiated by other people or organizations. In similar vein, Mollick & Kuppuswamy (2014) define crowdfunding as ‘the efforts by individuals and groups social, for profit and cultural to fund their ventures by drawing on relatively small contributions from a relatively large number of individuals using the internet, without standard financial intermediaries’.
Crowdfunding borrows a concept from micro finance (Mollick, 2014), and also in fact related to micro lending (Vitale, 2013), a concept which refers to the idea of funding individuals who do not possess access to conventional financing from credit institutions (Armendariz & Morduch, 2010)
Technological advancements, the internet and the rapid growth of social media have made it easier for entrepreneurs to reach general public more efficiently (Schwienbacher & Larralde, 2010). Furthermore, it also helped crowdfunding to become a serious way to secure funding, as a result of following reasons by Agrawal (2013). On the crowdfunding platforms, entrepreneurs and investors can be matched at low search costs. In addition, investors get the option to fund small amounts of money which reduces capital risk.
2.1.1 Advantages
The biggest benefit for a start-up company to use crowdfunding is probably the fact that crowdfunding campaigns do not only help creators obtain low-cost financing for their project, it also allows them to raise public awareness and interest before the product or service is launched. It allows them to test the potential market with the potential customer (Schwienbacher & Larralde, 2010).
Secondly, compared to other forms of financing, crowdfunding remove the geographic barriers to investment, which meand that the georgraphical distance between the entrepreneur and the investor does not play any role (Belleflamme et al., 2013). Furthermore, crowdfunding allows crowd to participate in the creation and improving the design of a product, crowdfunders contribute to creating value for the creator. (Schwienbacher & Larralde, 2010)
2.1.2 Disadvantages
The biggest disadvantage in crowdfunding is that not all of the projects created reach their funding goal. Running a project or campaign requires a lot from the people who work on it. Companies or individuals may spend countless hours on preparing a campaign and working on their idea and still fail. Working directly with the end-consumers require a set of different skills such as knowledge about consumer marketing, social networks or project management to ensure that the product reaches all potential consumers (Steinberg, 2012).
Crowdfunding projects also demand creators to reach a larger audience and show transparency and openness of their running projects compared to other forms of financing. Thus, when the projects are open to public, it also means that the people inside creator’s company are also in the public eye, especially the creators of the projects. To have a successful campaign, creators have to invest and spend a lot of time to stay in contact with their audience, which can be very demanding and stressful (Steinberg, 2012).
2.1.3 Crowdfunding models
According to Mollick (2014), there are four different kinds of crowdfunding models to fund a project. Depending on the preferences of the entrepreneur, the investor - backer of the project - can choose the type of model. The first model is called the patronage model. In the patronage model backers do not expect a direct return for the donations, they are more interested in the social goal of the project. The second model is known as loan or lending money. In this model the backer of the project expects a return on the amount they invest. The third model is the reward-based model. This model is the most used one and the backers are early adopters of the product/service. The model offers different range of rewards to
the investor, for example, a first version of the product/service, discounted prices or a meet and greet with the entrepreneurs. Lastly, with the equity model, backers operate as investors. They are given an equity stake in a project as return for their investment. For the remainder of this project, I will go more in depth on the reward based crowdfunding model.
2.1.4 The use of reward based crowdfunding model in game industry
The reward-based model is the dominant type of crowdfunding when it comes to the funds raised and number of projects (Wilson & Testoni, 2014). As mentioned above, this model allows entrepreneurs to give backers a tangible reward, such as discounted prices of the product/service, “goodies” (e.g. a t-shirt or mug) or an early prototype of the product. There are two types of reward models of funding ( Cummings et al., 2014): flexible funding or known as All or Nothing (AON) and flexible funding or known as Keep it All (KIA). In the Keep it All model which is used by Indiegogo, the project initiator owns the amount of funding even when the target amount has not been reached. Contrary to the Keep it All model used by Kickstarter, when the project initiator in the All or Nothing model fails to meet the goal, the pledge made by the crowd will be canceled and no rewards will be given. In this study, the All or Nothing model will be used considering that this model are more likely to achieve their goal compared to the KIA model. The AON model reduces the risks of the crowd, and enables the AON utilizing firms to set higher goals, raise more money, and increases changes to reach their stated goal (Cumming, Leboeuf & Schwienbacher, 2014). .In the game industry, these models were the most commonly used. As I mentioned before, in crowdfunding the collective opinion of a group of individuals is considered to be better than the opinion of few experts. This idea can also be related to the game industry where the traditional model mostly involves a contract with a game publisher. In this traditional model, the game publisher is the major key to success as it helps the game developer’s career by taking care of the financing, production, promoting and distribution of the game. Reward based crowdfunding approach mainly benefits independent game developers who can turn to crowdfunding for a chance to realize their projects without the support of a publisher. By using a reward based crowdfunding approach, the game developers can test the market before making their product while generating some pre-orders at the same time.
2.1.5 Success factors in reward based crowdfunding
Mollick (2014) examines the underlying dynamics of project success and failure and concludes that social network size as well as the project’s quality relates to the project success. Mollick (2014) was one of the first researchers to date that tried to explain the factors influencing a crowdfunding project’s success through an exploratory, empirical study. Mollick (2014) works with a broad set of independent variables
such as the amount of funding requested, the existence of video, the number of comments and news updates, the presence of spelling errors and the number of backers. Mollick (2014) revealed that crowdfunding projects that signal a higher quality level are more likely to be funded, large numbers of friends on social networks are positively associated with success and geography was linked with the success rates of projects.
Wheat, Wang, Brynes, and Ranganathan (2013) found that the video is perceived as important by potential project backers. They also assert that a video is an important means to introduce the project’s owner or team. Other scholars, Crossetto and Regner (2014) used project characteristics such as duration, image count, blog entries and categories as independent variables to measure the project’s success. Furthermore, Xu et al. (2015) described variables such as reward level and good project quality can lead to funder satisfaction which create trust between entrepreneur and backer.
Mollick (2014) points to the important role of social networks in funding new ventures. Some scholars included social factors such as the existence of a link to personal Facebook page in the project description, the number of Facebook friends or the projects previously supported by the creator into the varibles which might contribute to a project’s success. (Balboni et al., 2014; Saxton and Wang, 2013; Zvilichovsky et al. 2014). Lu, Xie, Kong and Yu (2014) also argue that social network, especially in the early stage of project, can rapidly increase the probability of a successful project funding.
2.2
Trust
Trust is an important factor under conditions of uncertainty and risk, and has barely been addressed in the literature. Previous research by Donna, Thomas and Marcos (1999) identified the developing of trust as a fundamental and still an unresolved issue in the development of internet shopping. Trust has a direct impact on the customer purchase intention which particularly important problem in the case of crowdfunding, because consumers are not sure whether the project can be completed as planned.
In the context of crowdfunding, trust means that consumers believe that the project creator has the ability to complete the project and achieve the desired results. To gain trust, creator has to deliver a good information quality to the potential backers. In the process to gain trust, creator can transfer project information to potential backers via video, picture, text and so on. Higher levels of trust have shown to be positively correlated with consumer loyalty and increase on the customers willingness to share personal information needed for buying products (Sichtmann, 2007).
Trust can also be affected by the behavior of the backers. Research by Sher (2011) about political voting found that people who are uncertain of which party to vote for, tend to vote for the side which leads the
polls. This means that people tend to go with the flow. For crowdfunding, it would imply as the more backers a project has, the more backers will be attracted to join as well as they perceived the project is more trustable and credible compare to the others.
Wathen and Burkell (2002) create a model of how users judge the credibility of online information. Their model of credibility assessment is divided into three stages of user website interaction. First, a person will notice the sites direct visual appearance and presentation (e.g. colors, graphics, typography) when he/she entering a website, interface design, its usability and the general information are being judge in this stage. Second, more depth evaluation is made about the sites message (e.g. content, relevance, currency) as well as its source (e.g expertise , trustworthiness). At a third and final stage, the interaction of the sites content with the users cognitive state is assessed. External factors such as the prior knowledge, the need for information can have influence on the processing of the perceived information.
2.2.1 Internal assessment
Kickstarter suggests that the number one rule of success in a crowdfunding campaign is to have a video on the project page. Prior research confirms that crowdfunding success is positively related to project quality signals. Mollick (2014) and Balboni et al. (2014) mention that a video description in a crowdfunding campaign is considered as one indicator of a higher-quality project. Mollick identifies the lack of video as extremely negative, describing how “producing a video is a clear signal of at least minimum preparation”. In addition, Joenssen et al. (2014) and Li and Duan (2014) state that adding more images in a crowdfunding campaign could influence the project’s success positively. Assessing image and video can help potential backers to determine which projects that has more credible information.
2.2.2 External assessment
In crowdfunding, an individual who decides to invest in a project is called a backer. The backer can decide what kind of investment he/she wants to make. In a reward based model, the backer can choose from a so-called ‘reward scheme’ his preferred backing amount to invest into the project (Xiao et al., 2014). In a study by Kuppuswamy & Bayus (2014), the authors found that backers tend not to back up a specific project if the project already received substantial amount of money. Furthermore, they also found that in the early stage of the crowdfunding process, backers feel less personal/connected. Which leads to less funding contribution in the early stage project of crowdfunding campaign.
Staff pick is one of Kickstarter unique feature which help projects to become more popular and get more attraction from backers. As stated on the Kickstarter website, “when something sticks out as particularly compelling, whether it is a really fun video, creative and well-priced rewards, a great story, or an
exciting idea, we make the project a Staff Pick”. The benefit of being a staff pick is make the entrepreneurs more compelling to get the attention not only from the potential backers but also from journalist or online communities. These two factors can influence external parties in this case potential backers in assessing the credibility and the potential of the project.
2.3
Social networks
The social network of project creators is often the initial funding source of many campaigns and plays an important role in determining success. In venture capital financing activities, a common phenomenon Prior study by Mollick (2013), found that the number of Facebook friends of creators is positively correlated with the success of campaigns. Kickstarter allows user to connect its profile with Facebook. This feature gives additional information for backers to evaluate the credibility of entrepreneurs which can affect their decision. People can see the number of friend that project creators have and can go directly to the creators profile pages on Facebook.
Project creators frequently encouraged supporters to promote a project in social media. A previous study by Xu et al. (2014) and Li et al. (2016) found that the number of social promotion updates is positively correlated to the success of the campaign. In Kickstarter sites, project creators can embed a link of their social media profiles (e.g. Facebook page) to their project page ad updates. They could share the URL in their Facebook, and the pledge from their friends would help improve the probability of crowfunding success. With the development of social media and online social networks, crowdfunding became ways for entrepreneurs to raise funding through their social networks (Song and Boeschoten, 2015).
2.4
Motives
Apart from a present as a sign of gratitude, backers who support or fund a project in a crowdfunding reward model usually do not get anything in return for their donation. They do not have voting rights, profit shares or venture’s ownership in exchange for their contribution. The act of donating on a crowdfunding platform thus cannot be viewed as a pure form of economic exchange where goods are given in exchange for money or other goods (Bagozzi, 1975). Funds that are raised on crowfunding platforms therefore can be regarded as a gift or reward.
In Kickstarter, all projects offer different kinds of rewards depending on the level of funds that they have pledged. These reward tiers that the creators offer to backers are called the rewards level (Mollick & Kuppuswamy, 2014). Rewards levels can be an important feature for entrepreneurs to attract investors. Reward levels usually start small but can escalate to larger rewards depending on the scale of the projects. The proposed rewards can be either monetary or non-monetary. The rewards that are offered in
campaigns vary widely for each individual and are often only limited to the creator’s creativity. According to the research by Xu et al. (2014) the effect of revising the reward levels of a project is often bigger than revising the project content to turn a campaign into a success.
3. Methodology
In this section the methodology of this study is described. First the data collection method is described. Then, the second part of the section is the way the data variables is prepared for analysis. I discuss my process of analysis. This includes an explanation of my regression model and the variables of interest.
3.1
Kickstarter
Kickstarter is a crowdfunding platform that has been founded in 2009 and have been using the AON as their reward model. Kickstarter is currently the largest crowdfunding platform in the world and holds thousands of different projects. The projects on Kickstarter can be divided in 15 different categories. This includes technology, games film and music (Kickstarter, 2016). In addition of being the largest platform, Kickstarter is also thought to be the most commonly known crowdfunding platform for entrepreneurs, investors and researchers. Many researchers who conducted research on crowdfunding have used Kickstarter as their platform of choice (Agrawal et al, 2013, Belleflamme et al, 2013; Mollick, 2013). Following previous studies on crowdfunding, this study will also focus on the crowdfunding platform Kickstarter.
3.2
Data collection
The dataset is extracted using the secondary data source webrobots. Webrobots is a website that crawls and gather data about all and ongoing project from Kickstarter. However, not all variables needed for this study were available on Webrobots. A second web crawler from Kicktraq used the data extracted by the first web crawler, and crawled every individual Kickstarter campaign page directly. Kicktraq allowed all additional variables directly from Kickstarter campaign page which guarantee the quality of the data.
A total of 7,862 game projects on Kickstarter were collected between January 2014 until January 2016 , and initiated from 17 countries. In order to eliminate the effects of cultural differences, only projects launched in the USA where most projects were created were selected as samples . To prevent biases, this paper also eliminate canceled and live projects which results in total of 4,790 projects which already succeeded or failed in gathering enough funds.
Following prior studies by Mollick (2014), I exclude some projects from our analysis. First, we remove projects with a funding goal under $5,000 and successful projects which funders are less than 20 persons. Most of them usually are seeking funds for a one-time project and can get enough funding support from their family, relatives and friends , which are not in line with the concept of crowdfunding. Then, I also remove projects with funding goal higher than $1,000,000 and extremely successful projects with funding
level over than 50 percent, because such projects are considered as outliers (Mollick, 2014). There were 2,972 projects extracted from the dataset, consisting in 1,698 successful projects and 1,274 failed projects. Moreover, to increase the effect of social networks, projects with missing values of Facebook friends are also excluded. Finally, there were 1,686 projects extracted from the dataset, consisting in 713 failed projects (42.12%) and 973 successful projects (57.71%).
3.3
Variables
For each project, this thesis collected the following information based on the previous research by Mollick (2014) such as project information, duration, funding goal, funding raised, backers, staff pick, and project descriptions. However, because of the lack of data this thesis does not include geography and the project delivering time variables. Therefore, key variables in this study are described in table 1.
Table.1 Key variables
Variables
Descriptions
Funding_goal
The amount founders had collected using crowdfunding
Projects_duration
The number of days which the project runs, calculated by
the start date minus end date of the project
Backers
The number of individual funders supporting the project
Staff_pick
A dummy variable where the project get picked by the staff
and was feature on Kickstarter home page
Rewards_count
The number of reward levels project offers
Image
A dummy variable to record the existence of images in the
project description
Video
A dummy variable to record the existence of videos in the
project descriptions
Facebook_Shares
The number of projects shares in Facebook
3.3.1 Dependent variable
In order to answer the research question, I used percentage funded as the dependent variable, representing the success level of the project. I examined the percentage of funding entrepreneurs/creators received during their campaign periods. This variable is calculated from the proportion of money raised compared to the funding goal.
3.3.2 Independent variables
Image: Text has a great ability to describe projects or campaign in detailed manner. However, it is reasonable that based on text alone a project presentation cannot successfully catch the attention of potential backers/funders. In that sense, this project consider that pictures or graphics are a key element to catch attention in order to make the visitor become the potential funders. According to research by Danaher et al. (2006) graphics have a significant positive influence on webpage visit durations. Hence, if the visitors have a longer visit duration this will increase the probability that the projects is worth funding. As Li and Duan (2014) stated in their study that adding more images in crowdfunding campaign could influence the project success positively.
H1: The use of an image within project description positively influences the funding success of a crowdfunding game project.
Video: Besides texts and pictures, videos is the third key element of presenting a project. While pictures or graphics only show a snapshot in time, videos allow us to see the movements with audio information. In some projects, a pictorial snapshot is not enough for a visitors to understand and to be convinced that the project is worth funding. Mollick (2014) described that a video in a campaign description could be considered as a signal of higher quality. Furthermore, Jiang and Benbasat (2007) also found that video content are perceived as more useful by the visitor compared to websites with static-picture only. This leads to the conclusion that videos increase the number of potential funders and also positively correlated with the project success.
H2: The use of a video within project description positively influences the funding success of a crowdfunding game project.
Backers: Customers are more willing to believe in the choices of others in the stock market (Kremer & Nautz, 2013) or in online purchasing (Ye, Cheng, & Fang, 2013). This effects also occurs in the crowdfunding ecosystem, with the increasing number of individuals backing projects indicate the credibility and feasibility of the projects. As Koning and Model (2013) stated that larger numbers of contributors represented a strong signal of project quality and success probability. Thus, it leads to the conclusion that more backers support the project positively correlated to the project success.
H3: A higher number of backers positively influences the funding success of a crowdfunding game project.
Staff_pick: With this unique feature, Kickstarter frequently pick their favorite projects to be featured in the homepage of the website. This feature could help projects significantly, according to Mollick (2014), being featured by Kickstarter on their homepage increase the chance to achieved their funding.
H4: Being picked and featured by Kickstarter staff positively influences the finding success in a crowdfunding game project.
Facebook_friends: Creators who have more contacts find it easier to reach their goal. The connection to profile in Facebook provides additional information for backers and can potentially affect their decisions. Thus leads to the conclusion that the more Facebook friends creators had, more likely the projects will achieved success.
H5: A higher number of creator Facebook friend positively influences the funding success of a crowdfunding game project.
Facebook_shares: As stated by Mollick (2013), promoting a project on social media helps the funding process. Creators can share their projects through the Facebook Share button and pledge their friends to improve the probability of projects success.
H6: A higher number of Facebook shares by creator positively influences the funding success of a crowdfunding game project.
Rewards_count: Projects that only offers one reward option should not be even considered, because coming up with multiple rewards in different price categories allows people to pledge more. Xu et al. (2014) stated that, having more options encourages potential backers to turn a campaign into a success.
H7: A higher number of rewards level offered by the creator positively influences the funding success of crowdfunding project.
3.3.3 Control variables
This project also collected data for two control variables. The first one is crowdfunding funding goal. Funding goal is defined as the amount of money the entrepreneur seeks to raise during the crowdfunding project (Mollick & Kuppuswamy, 2014). Many factors can influence the results of the projects, thus entrepreneurs should select realistic funding goals. As Mollick (2013) found that raising too little capital may result in project non-delivery, while high funding goals make projects less likely to succeed.
H8: A higher amount of funding goal negatively influences the funding success of a crowdfunding game project.
The second control variable is crowdfunding project duration. Project duration describes the number of days entrepreneurs are allowed to raise funds for his/her project (Mollick, 2013). Kickstarter Cofounder Yancey Strickler explains that “More time does not create more urgency. Instead it makes it easier for backers to procrastinate, and sometimes they forget to come back at all.” Furthermore, a research by Kuppuswamy & Bagus (2014) also found that projects who have a shorter duration tend to attract more backers. They also added that most of the successful projects in crowdfunding campaigns have shorter durations.
H9: A shorter project durations positively influences the funding success of a crowdfunding game project.
In conclusion, this project hypothesize that variable trust such as images, video, backers and staff pick, social network of the creators and a better rewards level will increase the likelihood of higher fundraising percentage. Our theoretical framework is illustrated in Fig. 1.
Figure 1 Theoretical Framework
3.4
Pearson correlation
A bivariate analysis is conducted to find out whether the independent variables were correlate with each other. The Pearson correlation coefficient is an extensive method to check for correlations. As it is also a good way to determine the linear relationship between two variables. This variables can be either both continuous, or one continuous, and one dichotomous variable. The size of the value indicates the strength of the relationship. A value of -1 is define as a perfect negative correlation, it means that the two variables are move in opposite direction whereas value of is define as a perfect posstive correlation, meaning that two variables are move in lockstep. If the correlation is zero, the two variables can be define as no relationship (Pallant, 2007).
3.5
Research model
According to the literature review, the outcomes of a crowdfunding project can be affected by trust, creators’ social network and creator’s motives Following on the characteristics of crowdfunding, This project build on previous studies and add some potential factors into our analysis.
We want to find influential factors of project outcomes. Since project success is a binary variable, which is 1 if it’s successful project and otherwise 0, a logistic regression is an appropriate regression model to analyze it. It has been used in order to model the probability of success, defining success as the occurrence of the outcome coded with the number 1 whereas failure refers to the value of 0 (Menard, 2002).
The logistic regression model also estimates the probability of a certain event resulting as a function of a set of continuous and categorical independent variables. Because the dependent (Y) in this project is binary variable (1 = success, 0 = failure), which I try to explain using a set of independent (X) variables. Dependent variables that are binary are not suitable for analyzing with OLS as the probabilities are not linear since there is a minimum and maximum (Gortemaker, Hosmer, & Lemeshow, 1994).
The main assumptions of logistic regression can be summarized in the following. First, the dependent variable must be a dichotomy . Second, the assumption of independence (independence of irrelative alternatives) must not be violated. This means that the odds of one group must be unaffected by the presence or absence of other groups. Outliers which can significantly influence the results of the logistic regression must be excluded (Peng & So, 2002). In this research, significant outliers have been excluded following the previous research by Mollick (2014). In contrary to the OLS method, the logistic regression method does not assume a linear relationship between the dependent and independent variables, however it does assume that there is a linear relationship between the continuous variables and the log odds of the dependent one. Lastly, larger samples are needed when using logistic regression since it shown that as sample size increases the size of bias in logistic regression parameter estimates approaches zero (Menard, 2002).
The following logistic regression model:
logit(Ysuccess=1) =
α
0+ β
1Image+ β
2Video+β
3Facebook_Friends+ β
4Facebook_Shares+ β
5BackersTo test whether the model regression model is accurate and fits with the data set, the goodness of fit test is used. This test can be measured by using the pseudoR2attributed to Nagelkerke (1991) and also the log
likelihood chi-square.
Before explaining the logistic regression function in this research, it is crucial to highlight why linear regression model is not suitable for the analysis. A study by Bahovec et al (2013) describes that we should assume that Y represents the dependent variable and x (for j=1,k) represents the values of k independent for this same Y. Y is a binary variable coded as Y = 1 in terms of success and 0 is failed. The possibilities of the two events occurring are P(Y=1) =p and P(Y=0) = P-1. The expected value E(Y) would be the same as P(Y):
p
p)
-(1
*
0
p
*
1
E(Y)
And the multiple linear probability is:
k k 2 2 1 1
x
x
...
x
E(Y)
p
O
(β0 ...βk) are vectors of unknown parameters (Bahovec et al., 2013). As shown in the model above linear regression model has a flaw when it comes to predicting dichotomous outcome. Despite the probabilities can only take values within the range (0,1) the linear function can take on any value. Moreover, usually the relationship between the probabilities and the independet variables is non-linear, it contradicts with the basic assumptions of the linear regression model (Groβ, 2003). Subsequently, this project should re-illustrate this linear function into a logistic function.
) ' ( ) ( 0 11 0
1
1
1
1
xe
xe
p
x k k
This project could also observe the logit transformation of this function
'
1
1
1
1
1
In
p
-1
p
In
' 0 ' ' 0 0 0x
Ine
e
e
logit(p)
x x x
'
1
In
0x
p
p
This odds ratio can be interpreted as an increase or decrease value of Y, from one unit of change in X. In this case the odds ratio explains the log odds of the event occurring (Y=1, is the event of success) created by a one increase in the X (independent variable).
Furthermore, to determine the final logistic regression model this project applied stepwise backward elimination method. The stepwise backward elimination process is more accurate and less risky. It does not fail to identify relationships that already exist in the model (Menard, 2002). This method allowed the analysis to focused on a broad model including all the independent variables. Moreover, it also eliminate all non-significant variables in a gradual, stepwise process, based on a significance level of p<0.05 (Menard, 2002).
4.
Results
In this section, I am going to show the results of the study. In the first section, descriptive statistics are described. The second section are the results of the regression model using SPSS 22.0.
4.1
Descriptive statistics
Table 2 shows descriptive statistics for the main variables. Table 2 shows all the projects included in my sample; table 3 and table 4 indicate failed and successful projects respectively. According to table 2 out of 1686 observations examined, the average funding goal is 97.825 Dollars compared to the average of money raised, which is 44.394 Dollars. The highest amount collected is 3.327.757 Dollars and the highest funding goal is 100.000.000 Dollars.
One of the external assessment Staff Pick is defined in either 0 or 1. Zero means the project is not picked and featured by Kickstarter staff whereas 1 is defines as project featured by Kickstarter staff. On average 0.25 projects are featured by Kickstarter staff. However this holds a standard deviation of 0.436. Another project characteristics is the Number of Backers. Minimum number of backers for a project is zero. Maximum number of backers for project is 16.936. On average a project gets backed up 617.47 times by funders. This external assessment shows a large deviation standard of 1282.652 which means that there are a wide range of diversity among the amount of backers a project attracted.
Furthermore, the game motives examined in this study is the Rewards level. On average a game projects created on Kickstarter has 10.64 rewards level and 6.745 on standard deviation.
Table 2 All Projects
N Minimum Maximum Mean Std. Deviation
Image 1686 1 1 1.00 .000 Video 1686 0 1 .82 .385 FB_friends 1686 0 5000 684.79 850.679 FB_shares 1686 0 25300 641.20 1437.920 Rewards_count 1686 1 58 10.64 6.745 Duration 1686 7 60 32.45 9.046 Goal 1686 5000 100000000 97825.12 2442019.514 Pledged 1686 0 3327757 37966.68 120491.706 State 1686 0 1 .58 .494 Staff_pick 1686 0 1 .25 .436 Backers_count 1686 0 16936 617.47 1282.652 Valid N (listwise) 1686
In table 3 and 4, descriptive statistics for the unsuccessful and successful game projects are presented. Out of 1689 observations were categorized as 713 failed projects (42.12%) and 973 successful projects (57.71%). According to the table, unsuccessful projects have higher goals and duration than successful projects. Unsuccessful projects have an average of 34,49 days while successful projects have only 30.96 days on average. This somewhat in line with the results found by Kuppuswamy & Bagus (2014) which says that projects who have a shorter duration tend to attract more backers.
One of the game project quality conducted in this study is the video used in the project. Successful funded projects have an average of 0.92 video while unsuccessful projects have only 0.67 video. Another project quality is the Image used in projects. The number of average for successful and unsuccessful projects are same because all the projects were using images in their project campaigns.
Moreover, projects that linked with their Facebook account and shared their projects in Facebook are more likely to be successful than who does not. On average projects who are successful have 829.64 Facebook friends while unsuccessful projects only have 487.12 Facebook friends. Furthermore, the number of Facebook shares differs for successful and unsuccessful projects. The average number of Facebook shares on successful projects was ten times more higher than unsuccessful projects.
Having more rewards positively increase the projects chances to be successful. On average successful projects have 12.18 rewards level while unsuccessful projects have only 8.54 rewards level. When looking at the average numbers of backers that supporting successful projects and unsuccessful projects, a difference can be noticed. Successful funded project have an average of 1021.40 backers whereas unsuccessful project have only 66.23 backers. Furthermore, getting featured by Kickstarter staff also make projects are likely three times to be successful than who were not featured.
Table 3 Unsuccessful Projects
N Minimum Maximum Mean Std. Deviation
Image 713 1 1 1.00 .000
Video 713 0 1 .67 .469
FB_friends 713 0 4994 487.12 720.205
FB_shares 713 0 12400 194.69 794.424
Duration 713 7 60 34.49 10.569 Goal 713 5000 100000000 203585.73 3753922.105 Pledged 713 0 460657 3461.65 19175.661 State 713 0 0 .00 .000 Staff_pick 713 0 1 .12 .321 Backers_count 713 0 3157 66.23 206.219 Valid N (listwise) 713
Table 4 Successful Projects
N Minimum Maximum Mean Std. Deviation
Image 973 1 1 1.00 .000 Video 973 0 1 .92 .264 FB_friends 973 0 5000 829.64 908.222 FB_shares 973 0 25300 968.39 1693.690 Rewards_count 973 1 58 12.18 7.110 Duration 973 7 60 30.96 7.402 Goal 973 5000 800000 20325.30 34467.768 Pledged 973 5000 3327757 63251.46 152922.466 State 973 1 1 1.00 .000 Staff_pick 973 0 1 .36 .479 Backers_count 973 17 16936 1021.40 1560.333 Valid N (listwise) 973
Table 5 provides a correlation matrix for our main variables in the sample. The Pearson correlation coefficient is a statistic to measure how two variables are strongly related to one another. In the table, the coefficient only show a weak or a negligible relationship between the corresponding independent variables. Of those variables, as already expected, only funding_goal and project_durations has a negative correlations with the project outcomes. Whereas other variables such as, video, FB_friend, FB_shares, rewards_count, staff_pick and backers_count has positive correlations with the dependent variables. Variable image is excluded from the correlation because the variable is constant and have no significance effect.
Table 5 Correlation Matrix
4.2
Regression Analysis
The regression results in the following chapter derived in line with the methodology explained in chapter 3. Following the logistic regression analysis to predict project success as the dichotomous dependent variable, the initial regression model included all 10 independent variables. As I mentioned in the previous chapter, this thesis applied stepwise backward elimination approach in order to gradually adjusting the model of significant predictive variables (p<0.05). The backward elimination process eliminated four independent variables showing no significant predictive power, and finally leading to 5 evolving regression models as shown in the tables 6-8. The final logistic regression model (table 8) includes the final significant variables.
Table 6 Regression Model 1-2
Table 7 shows the logistic regression coefficient, standard error, Wald test, degrees of freedom, significance level and the odds ratio for each of the predictor variables of the final logistic regression model 5. Applying a 0.05 criterion of statistical significance video, rewards level, backers and controlling the funding goal have significant effects on the success of projects.
Table 7 Regression Model 5
Model 5
B S.E. Wald df Sig. Exp(B)
95% C.I.for EXP(B) Lower Upper Video 1.494 0.269 30.763 1 0.000 4.454 2.627 7.551 Rewards_count 0.060 0.018 10.544 1 0.001 1.061 1.024 1.100 Backers_count 0.014 0.001 248.646 1 0.000 1.014 1.013 1.016 Goal 0.000 0.000 192.432 1 0.000 1.000 1.000 1.000 Constant -2.116 0.293 52.264 1 0.000 0.121 Observations 1686 Chi Square 85.676 P 0,000 Pseudo R Square 0.807 2 Log-Likelihood 750.086 Classification Success 92
A test conducted with the final four variables versus a model with intercept only was statistically significant, Chi Square χ2 = 85.676, p<0.001. For the model fit, the pseudo R2 (Nagelkerke) display a
value of 0.807 which mean that the model account for 87% of the variability in the dependent variable. Looking at the classification table (table 8), the model was able to correctly classify 92% of the cases, based on a cut-off value of 0.5 and shows a sensitivity of 92.9% and a specificity of 90.7%.
Table 8 Classification Tablea
Observed Predicted State Percentage Correct Failed Successful Step 1 State Failed 647 66 90.7 Successful 69 907 92.9 Overall Percentage 92.0
The variable video shows a positive relationship with the funding success. With an odds ratio of 4.454 the subsequent interpretation is: “One video offered in project description, makes it 4.454 times more likely for a game project to attain success, holding all other variables constant”.
The number of rewards level available for the backers shows a positive impact on funding success. The variable display a positive coefficient and an odds ratio of 1.061 leading to the following interpretation: “One more rewards level available makes it 1.061 times more likely a game project to successfully achieve their goal, holding all other variables constant”.
The variable Backers shows a similar relationship with a positive coefficient and an odds ratio of 1.014: “One more backers supporting the project, makes it 1.014 more likely for a game project to successfully reach their goal, holding all other variables constant”.
The odds ratio and the coefficient of the funding goal also indicate a positive relationship with their chance to successfully funding their project, leading to interpretation: “Decreasing funding goals makes a game project 1.000 more likely to be success in their campaign, holding all other variables constant”.
In summary, the analysis of all variables confirms the hypotheses H2, H3, H7 and H8. In addition, it rejects the hypotheses H1, H3, H5, H6 and H9.
5.
Discussion
In this part of research, I am going to discuss about the contribution of my research. The results of logistic regression analysis supported four out of the nine hypotheses of this study. The variables which significantly contribute to a crowdfunding project’s success are video, rewards level, number of backers and funding goal. Two of these variables rewards level and number of backers were classified under the project characteristic whereas the video variable was classified under the project quality. All the variables in social network category were found not significant consistent in the empirical analysis.
Following from the results of the empirical analysis, figure 2 demonstrates the findings of my hypotheses. The following part will discuss about the empirical findings for each hypothesis in relevance to the existing literature within each attributes category.
Figure 2 Outline Framework
5.1 Trust
Based on the analysis conducted for internal assessment, variable image (H1) was found not to be significant and was exclude from the regression because all of the projects were found to have at least one image on their project. As a result, the variable is constant and have no effect to the outcomes of the projects. The previous literature shows various and even contradicting results on the use of images in crowdfunding projects (Mollick, 2014; Joenssen et al., 2014). Even though the result showed no significant influence on project success, the content and quality of image, that weren’t taken under consideration during this study, might have a much bigger influence over the mere existence of an image. As Li and Duan (2014) stated that adding more images in crowdfunding campaign could influence the project success positively.
On the other internal assessment, this study found that the project which has video on their campaigns (H2) were significant (p<0.001). The hypothesis summarize that one video makes it more likely for a
game project to achieve success. This finding is in line with the existing literature (Mollick, 2014; Balboni et al, 2014) who argued that including a video within the project is a measure of the project’s quality and therefore contributes to project success. The usage of a video offer certain advantages over a plain text project description. Furthermore, this finding also supports the previous finding by Jiang and Benbasat (2007) who found that video content are perceived as more useful by their visitor compared to websites with static-picture formats only. Therefore including or adding video into game project campaigns can significantly increase the chance of the project to be success.
Number of backers (H3), a variable of external assessment that was found to be significant in the study by Koning and Model (2013), was also proven significant (p<0.001) in the present analysis. The hypothesis concludes that having one more backers supporting the project makes a project more likely to achieve their funding. This finding also confirmed that Customers are more willing to believe in the choices of others in the stock market (Kremer & Nautz, 2013) or in online purchasing (Ye, Cheng, & Fang, 2013). Crowdfunding users are more likely to support a project when they have the reassurance that their funds will be used in desire manner. If their funds are used well for further maintenance and development, then their chances to gained success would be expected to increase accordingly. This effects also occurs in the crowdfunding ecosystem, with the increasing number of individuals backing projects indicate the credibility and feasibility of the projects.
The idea of getting picked by the Kickstarter staff and get featured in their homepage (H4) can increase the chance of being successful is based on the finding by Mollick (2014). According to the literature, being featured by Kickstarter on their homepage will help a project to get a higher percentage to achieve their funding. However, reflecting to the analysis, the variable showed non-significant results to funding success. In short, the analysis rejected the previous research found by Mollick (2014).
5.2 Social networks
As it has been argued in the literature review, the role of social network in the crowdfunding scene is fundamental. As Hekman and Brusee (2013), studied the impact of social networks on the projects launched in the crowdfunding platform Kickstarter. The results of the study showed that creators with more friends on Facebook but less dense network were more likely to succeed than fundraisers with less Facebook friends but a denser network (Hekman & Brussee, 2013). However, the results of analysis suggests that the number of creators Facebook friends (H5) in game project campaign was found to be not significant. This finding also rejects the previous finding by Mollick (2014) who stated that the number of Facebook friends of creators is positively correlated with the success of campaigns. In short, this finding contradict with what I thought before. This may lead into another interpretation of the analysis, it suggest
that the number of Facebook friends may associated with more early contributions but not associated with success. Moreover, Facebook friends can also easily be manipulated which can cause a bias in the analysis.
Furthermore, I also thought a project with more Facebook shares (H6) may have a big influence in the success of game project campaigns. However, reflecting to the variable Facebook shares, the analysis found that the number of Facebook shares also did not have a significant influence on funding success. This finding also rejects previous research conducted by Mollick (2013) , Xu et al. (2014) and Li et al. (2016). However, the variable does not take an account for the content and timing of shares, which might have a significant impact on the backers funding decision and ultimately on funding success.
5.3 Motives
Based on the analysis conducted, it has been confirmed that rewards level (H7) has shown a high level of significance (p<0.001). The hypothesis concludes that one extra rewards level makes a project more likely to achieve their funding. This finding is consistent with previous research conducted by Xu et al. (2014) who also suggest that having more reward option encouraged potential backers to turn a campaign into a success. Müllerleile and Joenssen (2014) described that the significance of this factor are based on the effects of price discrimination or differentiation. Price differentiation can be interpreted as “offering a homogeneous commodity at the same time to different customers in different prices (Machlup, 1955), is considered to have a positive contribution on sales and profitability (Phillips, 2005). Mollick (2014) explained reward based crowdfunding as a form of preselling. Therefore I can argue that the more rewards level available to the backers, the bigger the chances to include a various attractive offer at an attractive price. Moreover, the more carefully creators set up a rewards level, the wider the potential audience of backers can be reached, the higher the chances of game project campaigns to achieve success.
Other aspect that was relevant in the findings was the controllable variables. Based on results above, this project found that increasing funding goals (H8) were highly significant (P<0.001). Therefore, an increasing amount of funding goal has a negative impact with the success of game crowdfunding projects. The hypothesis indicates that by increasing the funding goals, projects are more likely to failed. Thus, it confirmed the previous research by Mollick (2013).
On the contrary, the analysis found that the effect of project durations (H9) was not significant to the success of game crowdfunding projects. This finding is inconsistent with the previous studies by Mollick (2014) and Kuppuswamy & Bagus (2014). Thus, it rejected their findings by the evidence in our study.
5.4 Theoretical and managerial implications
On a theoretical level, this study contribute to the broader view on success factor especially in game industry. Combining the existing literature with the empirical analysis, I am able to develop a conceptual model success factors in reward based crowdfunding projects in game industry. The empirical analysis tested various variables established by other scholars (Mollick, 2014; Kuppuswamy & Bayus , 2014; Joessen et al, 2014; Xu et al, 2014) and finally identified four significant factors. As illustrated in figure 3, this study summarizes the empirical findings Adding to the factors from prior research.
Figure 3 Conceptual Model - Success Factors
The new conceptual model consists of four variables which significantly influencing funding success such as, video, backers, rewards level and funding goal. As I already discussed in the previous part, adding a video increases the chances of a game project to attain success. From a research point of view, this finding also supports previous research conducted by Mollick (2014). For creators, this empirical finding makes it ultimately more likely for them to create a video to give a complete information and to engage with the potential backers. By using a video as a project information, it lets potential backers to see the founders face, witness a product being tested and see the inspiration behind a products idea which will
increase the credibility of the project. Thus, the higher the credibility of project, the more likely the project will gain trust from potential backers. This will ultimately lead to the higher chances of the project success.
From the perspective of the backers, by looking at the video created by the creator, it shows that the creator has a passion and creativity for their game crowdfunding campaign. In game industry, video has been an important part of promotion as they used it to show their new or upcoming product. In Kickstarter, videos are displayed prominently on the page. They are pretty much the first thing potential backers will see, and pressing Play button is the first logical step for users to take. There, backers can see and hear all the things about the project and build a powerful sense of empathy and trust. Thus, I presume that funders who already understand the environment of game industry will more likely support a project with video in their project information.
In addition to video, the model underlines the importance of rewards level offered by the creator. An increasing number of reward levels increases the amount of available options for the potential backers to choose from. From creator perspective, this increase in levels makes project more likely to offer various and attractive reward at an attractive contribution price for one particular backer, which increase more potential backer. A range of rewards level also showed that the creator has a plan to reached the customers and developed the project.
From a backers perspective, especially in the game category, I presume that rewards are the most important and main motives for them. It is because, the potential backers main motives to support a project is to get a product in exchange for their donation. By adding more rewards level, they will have more freedom and choice according to their personal needs and their value of money. This will eventually lead them to invest or donate more if they found more attractive offer.
Another important factor for game crowdfunding success is the funding goal of the project. A lower amount of funding goal , within a reasonable range (Mollick, 2013), can inspire the potential backers that the creator request on funding is based on realistic expectations. Hence, increasing the perceived feasibility of the project and positively influencing the funding decision by the backers.
Lastly, the model underlines the importance of number of backers. The role of backers in the crowdfunding scene is fundamental. Backers can detect something fishy about crowdfunding campaigns that are not well structured. From creators perspective, it is important to get more backers as early as possible to support the project. Before the campaign started, creators should start reach specific online or offline game community to get connected with them.. From backers perspective, if they already know the
product or the creator before the launched of the project, they will feel more connected and ultimately more likely to support the project.
Surely there is no right method or average crowdfunding project. The drivers for crowfunding success should always be configured in a way that they support the actual content of the project. There are projects that were very successful in raising money and exceeded their funding goal up to 10 times. However, there are also a lot of new projects started every day. To stand out, one’s project needs to be really innovative, unique and out of the box to catch backers attention. Other marketing stuff such as videos, images and rewards can only flourish when projects have such characteristics.
Overall, the conceptual model developed by the study can offer additional value for entrepreneurs and business owners especially in game industry across the crowdfunding environment. The conceptual model offers the entrepreneur an insight into the factors of crowdfunding game project success. The following recommendation should be seen as support for entrepreneur or business owner in setting up a reward based crowdfunding game project.
First of all, prepare a good quality of video which explain the inspiration behind the product, the value of the product and the future development of the product. It is important to make the video personal, it is not only about the creator, but it is also about the company. Explain why the project it is so excited and explain that the project need help to make it. This video will show your credibility as a founder and will increase the trust level between creator and potential backers.
Secondly, plan and create a wide range of rewards level which offer significant value to the potential backers and are positioned at an attractive contribution price. This will reach various number of backers needs and divide them based of the corresponding price.
Thirdly, carefully selecting and choosing the right amount of funding goal. The funding goal should be kept within the reasonable limits for both creators and potential backers. A transparent and reasonable amount of project funding shows the creator preparedness on planning and assures the potential backers of the feasibility and development of the project.
Last but not least, start early to reach potential backers. Connect with offline or online community related to the project. A good and solid network will bridge the gap between creators and potential backers. This will eventually help creators to reach more customers and create a good project based on their feedback.
