Abstract:

Crowdfunding has emerged as an alternative financing method for project owners who have difficulties in finding funds. Project owners submit their projects to crowdfunding platforms, request funds from individuals and put their projects into action with the funds they collect. After the projects are published on the platform, the project can be updated, funders can make comments, and thus the project’s success can be affected. Therefore, in the present study, it is aimed to determine the success of crowdfunding projects at the beginning. Unlike the studies in the literature, only the numerical variables that the project owner can change at the beginning of the project are included in the analysis. In the dataset, there are 8 variables with a total of 4758 projects submitted to Kickstarter.com, a reward-based crowdfunding platform. According to the results of the logistic regression analysis, it was determined that description length, full description length and number of videos did not affect success of a project, while goal amount of funding, number of frequently used questions, the number of awards, the number of images and the project duration affected the success of the project. The classification rate of the proposed model was %75.6. In addition to the logistic regression analysis, t-test and correlation analyzes were also applied to the data and the results were interpreted.

Keywords: