Music​has been a form of entertainment for many years. The music industry is constantly putting forth the effort to improve the quality of our music. It would be an interesting exercise to predict that the song makes it into top charts from a mathematical perspective. The proposed model in this paper predicts the success of a song based on its audio features and listeners' perspective factors that will be of great help to music producers. Violating the features of the song to predict whether the song will be hit before it is released can greatly help producers to increase their profits and reduce the risks they take in​producing songs. The audio features can be processed using machine learning algorithms like Logistic Regression, Support Vector Machines, Random Forest, Naive Bayes Classifier and patterns may be identified in the processed data, which can finally be combined with listeners' perspective features to effectively predict hit songs.