Background Functional recovery after a major depressive episode (MDE) requires both clinical remission and preservation of cognitive skills. As attentional deficit may persist after remission, leading to functional impairment, its role as a prognosis marker needs to be considered. Methods Five hundred eight depressed outpatients (DSM‐IV) were assessed at baseline for clinical symptoms (QIDS‐SR), social functioning (Sheehan Disability Scale, SDS) and attention through the d2 test of attention and the trail making test, simple tests, respectively, requiring to quote or to interconnect relevant items. All patients were treated by agomelatine, and examined 6 to 8 weeks after baseline to assess clinical remission (QIDS‐SR ≤ 5) and/or functional remission (SDS ≤ 6). Results At follow up, 154 patients (31%) were in clinical and functional remission. Shorter cumulative duration of prior depression, shorter present MDE, and two parameters of the d2 test of attention were predictive of such positive outcome, the number of omission mistakes (F1) being the only one still significantly predictive (P < .05) with a multivariate approach. F1 was unchanged after remission, patients with less than 11 mistakes had a 2.27 times increased chance to reach full remission, and a dose–response relationship was observed, with a regular increase of positive outcome for less mistakes. Conclusions The number of omission mistakes (F1) of the d2 test of attention was a stable marker, being predictive of, and with a dose‐effect for, clinical plus functional remission. It may constitute a specific marker of attentional deficit, involved in the resilience process that enables individuals to develop more adequate strategies to cope with everyday functional activities.