A method for generating a social graph based on payment data is provided. The method includes extracting, from one or more databases, customer-based information for various entities and previous transaction information conducted by the various entities; generating data nodes for the various entities; determining a score for each of the data nodes with respect to a user based on the extracted customer-based information and the previous transaction information, in which the score indicating a proximity level with the user; generating a relationship model for the various data nodes with respect to the user based on the determined score; and identifying, based on a search performed for an input provided by the user on a user interface, one or more data nodes as potential candidates for conducting a transaction; and providing the potential candidates based on the proximity level with respect to the user.