The "gate control theory of pain" of 1965 became famous for integrating clinical observations and the understanding of spinal dorsal horn circuitry at that time into a testable model. Although it became rapidly clear that spinal circuitry is much more complex than that proposed by Melzack and Wall, their prediction of the clinical efficacy of transcutaneous electrical nerve stimulation and spinal cord stimulation has left an important clinical legacy also 50 years later. In the meantime, it has been recognized that the sensitivity of the nociceptive system can be decreased or increased and that this "gain control" can occur at peripheral, spinal, and supraspinal levels. The resulting changes in pain sensitivity can be rapidly reversible or persistent, highly localized or widespread. Profiling of spatio-temporal characteristics of altered pain sensitivity (evoked pain to mechanical and/or heat stimuli) allows implications on the mechanisms likely active in a given patient, including peripheral or central sensitization, intraspinal or descending inhibition. This hypothesis generation in the diagnostic process is an essential step towards a mechanism-based treatment of pain. The challenge now is to generate the rational basis of multimodal pain therapy algorithms by including profile-based stratification of patients into studies on efficacy of pharmacological and nonpharmacological treatment modalities. This review outlines the current evidence base for this approach.