Optimization is necessary to solve almost all real-world problems. Though there are several optimization techniques developed by previous researchers, society requires more because of the “No Free Lunch” theory, which explains the need of more metaheuristic algorithms. It is hard to frame one algorithm to figure out all types of problems. This paper proposes a Clustering-Induced Opposition-based learning Grasshopper Optimization Algorithm (CIOGOA) that makes clusters from all the grasshopper populations and then changes a specific dimension for each cluster. The idea behind this approach is, all grasshoppers of a swarm may not change their position in arbitrary direction. Only small groups from each swarm according to their proximity may move in a specific direction. Hierarchical clustering methods are used to create these groups. We applied our algorithm over 22 different types of optimization functions from CEC2005 real-parameter benchmark functions. To evaluate our proposed algorithm’s performance, we compare it with four metaheuristics, swarm intelligence algorithms along with basic grasshopper optimization algorithm (GOA). A statistical test, p-val test is done to show the superiority of our proposed algorithm over other metaheuristic algorithms.