Designing a forecasting model to identify cross-selling opportunities using data mining techniques in insurance companies

 The Iranian insurance industry is one of the economic leverage of the government, which, along with the Central Bank, plays an essential role in economic development. One of the major concerns of the insurance industry is the development of sales in companies, and one of the most effective ways to develop sales in a business organization is cross-selling. This increases the customer more tied with the organization, and it will cost much less than acquiring  new customers for the business, so the profitability of the organization will also increase. The main purpose of this research is to design a forecasting model using data mining techniques, with the help of which we can first identify potential customers of our target product and then offer them a marketing proposal to buy our product.

The insurance product that we focused on was life insurance because its sales are lower than other insurance products and it has a good profit for insurance organizations. Therefore, in this study, first important indicators for identifying a life insurance customer were identified through a literature review and then interviews with insurance industry experts, and then using the XGBoost classification algorithm, the forecasting model was designed, trained, and evaluated through which we can identify potential life insurance customers among the property insurance clients of the insurance company (Vehicle Insurance) and offer them to buy life insurance.