Randomized controlled trials showing a significant benefit are met with enthusiasm because they may change the standard of care for patients who share the clinical and pathophysiologic characteristics of trial participants. Nonetheless, a well-designed and fully executed trial with neutral or negative findings also represents a critically important investigation deserving careful scientific scrutiny. In this paper we propose a 10-step approach to the interpretation of neutral or negative trials to exclude important methodological issues before concluding that the treatment really does not work. We will discuss this approach using the most classic trials of the past and some notable examples among superiority trials (mostly phase 3 trials) published over the last years.