Decision making is central to drug development, and at its core consists of evaluation of information to select a course of action based on development goals, risk awareness and tolerance, and overall strength of evidence. Incorrect decisions often have significant consequences that may not be easily remediated. For example claiming futility or efficacy based on an interim effect, or terminating/continuing a phase 2 program to the next stage of pivotal trial development. Often statistical significance drives perspective on whether a trial justifies further development or not. This talk will examine a Bayesian decision making paradigm that weighs evidence from Phase 2 data against predefined states of a Target Product Profile (TPP) and provides a flexible user-defined framework for decision making. In addition, implications for study design will be discussed in relation to study operating characteristics, sample size, and futility stopping. The paradigm will be contrasted with a traditional hypothesis testing approach geared towards statistical significance under reasonable control of type 1 and type 2 errors. Advantages and potential limitations will be discussed.