Today, startups are turning into the driving forces of the economic growth throughout the world. During the last decade, number of start-ups has grown dramatically all over the world. The very uncertain and unstable nature of startup environments in the initial stages of their activity has made it difficult to analyze and interpret information in order to evaluate their success. Therefore, investors and venture capital funds need a suitable tool to predict startups’ success rate. Considering the time complexity and computationally intensive nature of the problem, predicting startups’ success rate at the beginning of their activity with this approach requires the use of a quantitative model. In this study, the success of a start-up company is defined as their ability to reach the next stage of investment within a certain period of time. The ability to predict success is regarded as a valuable competitive advantage for venture capitalists who are looking to invest in startups. In fact, their first choices for investment are companies with faster growth potential than the competitors. Finally, the proposed model enables investors to move one step ahead of the competitors. To address this challenge, the present study provides a solution to predict the startups’ success rate in the future. Specifically, first, by collecting the data from three different sources, it examines and extracts factors affecting the startups’ success rate. Then, using machine learning methods, it presents a model for predicting the startups’ success rate, which has improved significantly compared to the previous models