Recent empirical work suggests that small price changes are relatively common. This evidence has been used to criticize classic menu-cost models. In this paper, we use scanner data from a national supermarket chain and micro data from the Consumer Price Index to reassess the importance of small price changes. We argue that the vast majority of these changes are due to measurement error. We conclude that the evidence on the prevalence of small price changes is much too weak to be used as a litmus test of nominal rigidity models.