The basic concept of clustering and its correlating research work is firstly present, a new algorithm based on least clustering cell (LCC) is proposed and analyzed which concerns the advantages and disadvantage of k-means and grid clustering algorithm. This algorithm is efficient in dealing with huge amounts of data and can make paralleled processing, which is proved to be correct, efficient and fast through application in customer relationship management. It overcomes the disadvantage of given the value k in k-means clustering algorithm and data dense in grid based clustering algorithm. Lastly the analysis and evaluation of the algorithm is given.