The technique of distributing a set of natural entities into groups of related objects is called clustering. Trying to find a good search platform for data mining has incrementally become a key problem as most conventional search methodologies are still unable to contribute to the improvement of knowledge discovery. Spiral optimization (SO) seems to be an accurate functioning algorithm that uses natural processes including the spiralling pattern and pressurized to treasure an optimal result in a reasonable period of time. A novel SO is introduced in this paper to tackle the clustering dilemma. With the exception of the actual SO, that recursively spins the objects across the self-righteous centre, the developed system divides the population into many subgroups to maximize searching heterogeneity and thereby boost the result of cluster analysis. The procedure of k-means has also been utilized to improve the usefulness of the anticipated algorithm. To assess the proposed method’s accuracy, we introduce it to the problem of clustering and measure the findings through the optimization using spiral strategy and clustering using k-means algorithm. The outcomes illustrate that the suggested procedure is very optimistic.