Introduction Greater occipital nerve block (GONB) is a minimally invasive procedure frequently used in patients with chronic migraine (CM); however, the quality of the evidence supporting its use is still unknown. Therefore, we aimed to conduct a systematic review, meta‐analysis and quality assessment of GONB local anaesthetics combined or not with corticosteroids to prevent CM. Methods We searched Medline, Scopus and Web of Science up to October 2020. We included randomized control trials (RCT) and observational studies assessing GONB without language restrictions. Two researchers selected the studies, extracted the data and evaluated the risk of bias independently. The primary outcomes measured to assess efficacy were the change from baseline in the intensity and frequency of headache in the intervention group compared to placebo at a onetime point. We performed a meta‐analysis with random effect models, and we evaluated random errors with trial sequential analysis (TSA). We assessed the risk of bias (ROB) with the ROB2 tool and the certainty of the evidence with GRADE. Results We identified 2864 studies in the databases and included three RCTs for quantitative synthesis. Most ROB assessments were ‘high risk’ or ‘some concerns’. GONB reduced the intensity of headaches at the end of the first month (MD: −1.35, 95% CI: −2.12 to −0.59) and the second month (MD: ‐2.10, CI 95%: −2.94 to −1.26) as well as the frequency of headaches (first month: MD: −4.45 days, 95% CI: −6.56 to −2.34 days; second month: MD: ‐5.49, 95% CI −8.94 to −2.03 days). Corticosteroids did not show a significant decrease in the frequency of headaches during the first month of treatment (MD: −1.1 days, 95% CI: −4.1 to 1.8, p = .45). Included trials reported similar adverse events between groups. The exploratory TSA showed inconclusive results. Overall, the quality of the evidence was very low because of the substantial risk of bias and imprecision. Conclusion The limited evidence available shows that GONB with local anaesthetics could reduce headache frequency and intensity compared to placebo, while adding corticosteroids did not show additional benefits. GONB was safe with a similar number of minor adverse events. However, our confidence in these estimates is very low since the evidence is based on a few trials, with a small sample size and a significant risk of bias. In addition, the exploratory TSA was inconclusive, so we need larger and specific trials.