International Journal of Scientific Research and Engineering Development

More than 50% of startup companies fail in the initial four years. Further, three out of every four venture-backed firms fail. The algorithm proposed in this paper will help to predict the success of a startup company based on financial and managerial variables. This prediction will help investors to get an idea whether the investing in a startup will be successful or not? Apart from implementing a model consisting of all the factors mentioned below and predicting the success of a startup company, various other models will be created representing various milestones achieved by the company. This paper will help startup companies to know which factors are essential for getting an investment. The algorithm will be based on more than 15,000 companies’ data collected from crunchbase.com. The financial variables include: investments in each funding rounds, valuation after each round of funding, current market value, total funds, investments and acquisitions by the company, financial background of key people and the managerial variables includes: Number of employees, competitors, location, age of the company, founders background, burn rate and various news articles on the company scrapped from internet. A variety of methods will be used to determine the best model such as random forest, text parsing, logistic regression, decision tree and survival analysis.