This paper addresses two long-standing goals in software testing: making mutation-based testing practical and software testing overall more effective, predictable, and consistent. To that end, this paper proposes a novel mutation-based approach that guides testing effort based on test goal utility.