A method for testing a set of new game features for an online gaming application is provided. The method initiates with selecting a new game feature for testing from the set of new game features. Users are assigned to a test group or control group. Each of the test group users is provided a gaming application including the selected new game feature enabled. Each of the control group users is provided the gaming application without the selected new game feature enabled. Evaluation scores are generated based on gaming activity and social network activity. A score comparison report is generated comparing the evaluations scores of the test group and the control group. Based on the score comparison report, it is determined whether to continue testing one or more additional new game features from the set of new game features.