Grey Wolf Optimizer (GWO) is a very efficient metaheuristic inspired by the hierarchy of the Canis-lupus wolves. It has been extensively employed to a variety of practical applications. Crow Search Algorithm (CSA) is a recently proposed metaheuristic algorithm which mimics the intellectual conduct of crows. In this paper, a hybrid GWO with CSA, namely GWOCSA is proposed which combines the strengths of both the algorithms effectively with the aim to generate promising candidate solutions in order to achieve global optima efficiently. In order to validate the competence of the proposed hybrid GWOCSA, a widely utilized set of 23 benchmark test functions having a wide range of dimensions and varied complexities, is used in this study. The results obtained by the proposed algorithm are compared with ten other algorithms in the literature for verification. The statistical results demonstrate that the GWOCSA outperforms other algorithms including the recent variants of GWO called Enhanced Grey Wolf Optimizer (EGWO) and Augmented Grey Wolf Optimizer (AGWO) in terms of high local optima avoidance ability and fast convergence speed. Furthermore, in order to demonstrate the applicability of the proposed algorithm at solving complex real-world problems, GWOCSA is also employed to solve the feature selection problem as well. The GWOCSA as feature selection approach is tested on 21 widely employed data sets acquired from the University of California at Irvine (UCI) repository. The experimental results are compared to the state-of-the-art feature selection techniques including the native GWO, EGWO and AGWO. The results reveal that the GWOCSA has comprehensive superiority in solving feature selection problem which proves the capability of the proposed algorithm in solving real-world complex problems.