For more than a decade the humanitarian community has been mandated to mainstream gender in its response to crises. One element of this mandate is a repeated call for sex-disaggregated data to help guide the response. This study examines available analyses, assessments and academic literature to gain insights into whether sex-disaggregated data are generated, accessible and utilised, and appraised what can be learned from existing data. It finds that there is a gap between policy and practice. Evaluations of humanitarian responses rarely refer to data by sex, and there seems to be little accountability to do so. Yet existing data yield important information, pointing at practical, locally-specific measures to reduce the vulnerability of both males and females. This complements population-level studies noting the tendency for higher female mortality. The study discusses some possible obstacles for the generation of data and hopes to spur debate on how to overcome them.