As one of the financial industry, the insurance industry is now facing a vast market and significant growth opportunities. The insurance company will generate a lot transaction data each day, thus forming a huge database. Recommending insurance products for customers accurately and efficiently can help to improve the competitiveness of insurance company. Data mining technologies such as association rules have been applied to the recommendation of insurance products. However, large policyholders' data will be calculated when it being processed with associate rule algorithm. It not only requires higher cost of time and space, but also can lead to the final rules lack of accuracy and differentiation. In this paper, a recommendation model for insurance products based on consumer segmentation is constructed, which first divides consumer group into different classes and then processed with associate rule algorithm. The empirical results show that our proposed method not only makes the consumption of association rules analysis reduced, it has also got more effective product recommendation results.