Recommendation algorithms are best known for their use on e-commerce web sites, where they are used to create additional business opportunities by suggesting additional products and services. Generally, the recommendation are created by collating feedback from various users who have purchased the same or different products, as well as comparing the features of the products themselves. While recommendations have been most successful in domains like retail, due the availability of large volume of feedback, it is challenging to implement in domains where there is no such prior information or the information available is very small in volume.  This paper presents how a hybrid recommender system was leveraged in insurance domain, by integrating an attribute based recommendation system with preference-based recommendation system.