Clustering is considered as widely used data mining practices. Clustering is the technique of dividing entire dataset in certain clusters created on the comparable characteristics of the instances. On the foundation of the likeness between the instances of data, grouping or clustering the instances of the large database regardless of its size is considered as significant chunk of data mining. There are plentiful approaches of clustering but this book mainly focuses on improving k-Means clustering algorithm. This method clusters the input dataset in quantified number (k) of groups. This method is verified to be very efficient when while dealing with small data, but for huge data, it fails in time complexity; it takes time more than usual. This work mainly aims comparison of k-means clustering scheme with ranking method to speed up the comprehensive running time for k-Means clustering algorithm. The experimental results clearly confirms that the new technique is more time efficientthan the old-style k-Means clustering method.