Data Mining is an analytical process intended to explore the data in search of consistent patterns. Due to its wide spread applications in biomedical industry and publicly available genomic data, data mining has become upcoming topic in the analysis of gene expression data. Clustering is the first step in understanding the complicated biological systems. The objective of clustering is to organize the samples into intrinsic clusters such that samples with high similarity belong to same cluster. The significance of clustering gene profiles is two-fold. Firstly, it assists in diagnosis of the disease condition and secondly it discloses the effect of certain treatment on genes. In this paper, we propose a new method to cluster gene expression data that is solely based on the concept of hierarchical clustering with a different method to compute the similarity between datasets and merge the pairs. The experimental results on two microarray data show the correctness and competence of proposed technique.