In a ‘post-truth’ era in which personality and opinion trump evidence and reason, the need for frankness in debates about the use and boundaries of science and policy is high. We welcome the reflective and nuanced approach to behavioural science in policy-making in Sanders, Snijders and Hallsworth's (2018) piece. Despite our support for the approach in this paper, we suggest that there are deeper issues than are currently acknowledged. Our critique tackles three issues: the empirical, the normative and the political. In the first section, we examine what counts as ‘behavioural’ and how this label is used to legitimate a range of policy activities. We then look at randomised controlled trials in the next section, highlighting the extra-scientific dimensions of the empirical ‘What Works’ revolution. Finally, we question some ontological assumptions that drive empirical research and its translation into policy, asking where the collective is to be found in behavioural public policy.