The present paper reports in parallel the findings of the two phase III trials that evaluated the efficacy of agomelatine in older depressed patients. It describes how the particular methodological innovations (particularly in relation to patient selection, design and accuracy of diagnosis of depression) introduced in study 2 have improved the quality of recruitment of patients and the assay sensitivity. Study 1 lacked assay sensitivity, and among the many differences with study 2, the inclusion of unexpected mildly ill patients could have inflated the placebo response. The increased demands on investigators in study 2 appear to have reduced the placebo effect and showed a robust benefit of agomelatine. The two agomelatine studies offer the opportunity to discuss hypotheses that have been raised to explain the low level of response of older patients to available antidepressants.