New drugs that retain potency against multidrug/extensively drug-resistant strains of Mycobacterium tuberculosis, with the additional benefit of a shortened treatment duration and ease of administration, are urgently needed by tuberculosis (TB) control programs. Efforts to develop this new generation of treatment interventions have been plagued with numerous problems, the most significant being our insufficient understanding of mycobacterial metabolism during disease. This, combined with limited chemical diversity and poor entry of small molecules into the cell, has limited the number of new bioactive agents that result from drug screening efforts. The biochemical, target-driven approach to drug development has been largely abandoned in the TB field, to be replaced by whole-cell or target-based whole-cell screening approaches. In this context, the properties of a good drug target are unclear, since these are directly determined by the ability to find compounds, using current screening algorithms, which are able to kill M. tuberculosis. In this review, we discuss issues related to the identification and validation of drug targets and highlight some key properties for promising targets. Some of these include essentiality for growth, vulnerability, druggability, reduced propensity to evolve drug resistance and target location to facilitate ready access to drugs during chemotherapy. We present these in the context of recent drugs that have emerged through various approaches with the aim of consolidating the knowledge gained from these experiences to inform future efforts.