Speech and gesture are two vital components of communication. Gesture itself provides an external support to speech, potentially promoting comprehension of a spoken message. The question of whether gesture promotes comprehension is not new, with research dating back to the 1970s. However, when gestures are most beneficial to comprehension is poorly understood. This meta-analysis explored 2 questions: whether and when gestures benefit comprehension of verbal information. We examined the effect sizes of 83 independent samples. Within each sample, a learner's comprehension was measured when gestures accompanied speech, compared with speech alone. Across all samples, gesture had a moderate, beneficial effect on comprehension when either produced or observed by a learner. Further stratified tests revealed that gestures significantly benefitted comprehension under a variety of circumstances, dependent on the type of gesture used, the information provided by gesture, the function of the gesture, the age of the learner, and the way comprehension was measured. The function of the gesture moderated the magnitude of the effect, with studies investigating the effect of producing gestures on comprehension yielding significantly larger effect sizes on average than studies investigating the effect of observing gestures on comprehension. The results from the current meta-analysis have theoretical and practical implications for gesture-related research and highlight new avenues for future studies.