The search for robust, clinically useful markers for major depression (MD) has been relatively unproductive. This is unfortunate because MD is one of the largest socio-economic challenges for much of the world and the development of reliable biomarkers for MD could aid in the prevention or treatment of this common syndrome. In this editorial, we compare the approaches taken in the search for biomarkers for MD to that of the more successful searches for biomarkers for tobacco use, and identify several substantive barriers. We suggest that many of the existing clinical repositories used in these biomarkers searches for MD may be of limited value. We conclude that in the future greater attention should be given to the clinical definitions, characterization of confounding environmental factors and age of subjects included in studies. © 2014 Wiley Periodicals, Inc.