The major clinical presentations seen by critical care physicians are sepsis and acute respiratory distress syndrome (ARDS), both of which are heterogeneous clinical syndromes rather than specific diagnoses. The current diagnostic criteria provide little insight into the mechanisms underlying these heterogeneous syndromes and minimal progress has been made with regard to the development of therapies, despite many large randomized controlled trials being undertaken. This review outlines the advances made in improved characterization of critically-ill patients, using ARDS as an exemplar, and highlights the need for this improved patient characterization to be coupled with mechanistic science to develop therapies that target specific pathomechanisms.