In recent decades, consumer credit has experienced phenomenal growth largely due to accurate automated credit scoring models that aided lenders decisions. The predictive power of credit scoring models is vital to the profitability of financial institutions, as they help to prevent losses and to maximize profits. Nowadays, there are a large variety of credit scoring models utilized to address the problem of credit risk estimation. Thus, this paper proposes to explore the fundamental literature and provide an overview of the credit scoring context and methods. Some industry standard techniques, the mostly used classification models and a popular ensemble approach are presented, in a way that their strengths, weaknesses and relevant details are highlighted. Machine learning models demonstrate better results than the statistical techniques. Also, recent studies have been massively researching the use of ensemble methods, which consistently show superior predictive performance over the single classifiers in terms of average accuracy.