In general trips frequently entail several stages varying in mode, duration, and other factors. In some way travelers aggregate their satisfaction with the stages to satisfaction with the whole trip. In this paper we address the question of how this aggregation is made. We use data from a Swedish survey measuring satisfaction with commutes to and from work and with the stages of the commutes. We test several aggregation rules for their goodness of fit to the observations. Our results show that a normatively correct averaging rule that takes into account the relative durations of the stages out-perform heuristic aggregation rules such as the peak-end, summation, and equal-weight averaging rules. We note that this does not exclude that the heuristic aggregation rules apply to other trips than repetitive commute trips.