The Kafka method (now applied through Kafka Field Labs) was initially developed in the Netherlands as a practical approach to shift services to become more citizen centred. We started with action research, to learn about bureaucratic dysfunction in detail from the user perspective and to help tackle it at the same time. Action research can be described as a family of research methodologies that pursue action (or change) and research (or understanding) at the same time (Dick 1991). Ideally this is done through cyclical or spiral processes, which alternate between action and critical reflection. While our developmental process has been messier than that, we have continuously made efforts to refine methods, data and interpretation in the light of our developing understanding of bureaucratic dysfunction. Action research requires proximity to the object of study (Reason and Bradbury 2004, Stringer 2007): only when you engage with the subject matter will it reveal its nature and inform you on how best to study it. We found this highly applicable to the diagnosis of bureaucratic dysfunction.