Abstract Phenotypic sequences are a type of multivariate trait organized structurally, such as teeth distributed along the dental arch, or temporally, such as the stages of an ontogenetic series. However, unlike other multivariate traits, the elements of a phenotypic sequence are arranged along a vector, which allows for distinct evolutionary patterns between neighboring and distant positions. In fact, sequence traits share many characteristics with molecular sequences. We implement an approach to estimate rates of trait evolution that explicitly incorporates the sequence organization of traits. We apply models to study the temporal pattern evolution of cricket calling songs. We test whether songs show autocorrelation of rates (i.e., neighboring positions along a phenotypic sequence have correlated rates of evolution), or if they are best described by rate variation independent of sequence position. Our results show that models perform well when used with sequence phenotypes even under small sample sizes. We also show that silent regions of the songs evolve faster than chirp regions, which suggests that macroevolutionary changes are faster when associated with axes of variation less constrained by multiple sources of selection. Our approach is flexible and can be applied to any multivariate trait with units organized in a sequence-like structure.