Background: Observational studies on the prevalence of premonitory symptoms in people with migraine, preceding the headache pain (or aura) phase, have shown conflicting results. We conducted a systematic review and meta-analysis to estimate the prevalence, and relative frequency among clinic populations, of premonitory symptoms in people with migraine, overall and of the multifarious individual symptoms, and to review the methodologies used to assess them. Methods: We searched PubMed and Embase for studies published from database inception until 31^st of May 2022. Two investigators independently screened titles, abstracts, and full texts. We retrieved observational studies that reported the prevalence/relative frequency of one or more premonitory symptoms in people with migraine. Two investigators independently extracted data and assessed risk of bias. Results were pooled using random-effects meta-analysis. Our main outcomes were the percentage of people with migraine who experienced at least one premonitory symptom and the percentages who experienced different individual premonitory symptoms. To describe our outcomes, we used the terms prevalence for data from population-based samples and relative frequency for data from clinic-based samples. We also descriptively and critically assessed the methodologies used to assess these symptoms. Results: The pooled estimated prevalence in population-based studies of at least one premonitory symptom was 29% (95% CI: 8–63; I² 99%) and the corresponding pooled estimated relative frequency in clinic-based studies was 66% (95% CI: 45–82; I² 99%). The data from clinic-based studies only supported meta-analysis of 11 of 96 individual symptoms, with relative frequency estimates ranging from 11 to 49%. Risk of bias was determined as high in 20 studies, moderate in seven, and low in two. Conclusions: The substantial between-study heterogeneity demands cautious interpretation of our estimates. Studies showed wide methodological variations, and many lacked rigor. Overall, the evidence was insufficient to support reliable prevalence estimation or characterization of premonitory symptoms. More data are needed, of better quality, to confirm the existence of a distinctive premonitory phase of migraine, and its features. Methodological guidelines based on expert consensus are a prerequisite.