To handle with the problem of so many-answers replied from an online Web database in response to a relative query, this paper proposes a unique approach to rank the similar query results. Depending upon the on the database workload, we hypothesize how much the user worries about each and every attribute and assigns a specific weight to it. Then, based on the relationship among of different attribute values, a user satisfaction degree ranking method develops, which ranks all the answer tuples according to their satisfaction degree level for the initial query asked. Then, for the tuples with the same satisfaction degree, each undefined attribute value is assigned with a score according to its “importance” to the user, which is used for differentiating these tuples. Finally, the results of preparatory experiment shows effectiveness of the ranking techniques used for better user satisfaction in web-database systems.