There is a tension in the policy field about the way that the policy process should occur and the manner in which evidence should inform policy decisions. This article presents findings from a qualitative study on the introduction of point-of-sale tobacco display bans in Australia to provide an insight into the way this public policy was introduced and the way that evidence was used and communicated to influence the policy outcome. This study highlights that tobacco control policy does not merely consist of a series of segmented and unconnected decisions. It is an ongoing process, whereby the acceptability of forms of knowledge has shifted over time. This has enabled the introduction of tobacco control initiatives on the basis of good judgement when there is a paucity of specific evaluation information to support a policy's introduction.