Today, increased level of customer awareness caused themto access to the other suppliers easily and they can get their servicesfrom the competitors with similar or even better quality and same price.Therefore, focusing on customers and preventing them to leave, has beenthe most important strategy for any company. Researches have shownthat retaining former customers is cheaper than attracting new ones. Inthe proposed model in this article we first identified important factorscausing customers in insurance industry, to have a specific behavior byusing a k-means clustering algorithm, and then we tried to predict thefuture behavior of them by a logistic regression. Our model accuracy is98%. Manuscript profile