Graph Contrastive Learning (GCL) methods primarily employ random augmentations, such as node and edge dropping and feature masking, to generate contrasting views. However, these perturbations often lead to arbitrary information loss, disrupt structural integrity, and degrade the quality of representation. Although recent adaptive augmentation techniques aim to address these limitations, they introduce substantial computational overhead and reinforce dataset-specific biases, thereby restricting generalization. To overcome these challenges, we propose a neighborhood commonality-based augmentation strategy that captures critical node and edge information while aligning node representations by preserving both local and global structural dependencies. Our method achieves a favorable trade-off between computational efficiency and performance, avoiding the complexities of adversarial or eigen-decomposition-based augmentation strategies. Furthermore, we provide a rigorous theoretical justification for our approach and empirically validate its effectiveness through experiments.