While big data has the potential to make a significant contribution to international development, that potential is currently constrained by a number of barriers. Systematic analysis of those barriers is rare, so this paper applies the design-reality gap model to identify and evaluate barriers to effective use of big data in one context: the Colombian public sector. The model provides a structured framework that exposes a broad set of barriers, and also helps highlight priority areas for action to accelerate the application of big data. The design-reality gap model can also be seen to provide the basis for related analyses such as readiness for big data, and risk identification for big data initiatives in developing countries.