Although enhanced priority-setting for investments in health research for development is essential to tackling inequalities in global health, there is a lack of consensus on an optimal priority-setting process. In light of the current surge in tuberculosis (TB) research investment, we use TB as a case study.We investigated two critical aspects of a research prioritisation process, namely the criteria that should be used to rank alternative research options and which stakeholders should be involved in priority-setting. We conducted semi-structured interviews with 24 key informants purposively selected from four distinct groups - academia, funding bodies, international policy or technical agencies, and national disease control programmes. Interview transcripts were analysed verbatim using a framework approach. We also performed a systematic analysis of seven diverse TB research prioritisation processes.There was consensus that well-defined and transparent criteria for assessing research options need to be agreed at the outset of any prioritisation process. It was recommended that criteria should select for research that is likely to have the greatest public health impact in affected countries rather than research that mainly fills scientific knowledge gaps. Some interviewees expressed strong views about the need - and reluctance - to make politically challenging decisions that place some research areas at a lower priority for funding. The importance of taking input from stakeholders from countries with high disease burden was emphasised; such stakeholders were notably absent from the majority of prioritisation processes we analysed.This study indicated two critical areas for improvement of research prioritisation processes such that inequalities in health are better addressed - the need to deprioritise some research areas to generate a specific and meaningful list for investment, and greater involvement of experts working in high disease-burden countries.