Objectives Small nucleolar RNAs (snoRNAs) form clusters within the genome, representing a mysterious category of small non-coding RNAs. Research has demonstrated that aberrant snoRNAs can contribute to the development of various types of cancers. Recent studies have identified snoRNAs as potentially valuable biomarkers for the diagnosis or/and prognosis of cancers. However, there has been a lack of comprehensive reviews on prognostic and diagnostic snoRNAs across different types of cancers. Methods We conducted a systematic search of various databases including Google Scholar, Medline, Cochrane, Scopus, PubMed, Embase, ScienceDirect, Ovid-Medline, Chinese National Knowledge Infrastructure, WanFang, and SinoMed with a time frame reception to December 30, 2022. A total of 49 relevant articles were included in our analysis, consisting of 21 articles focusing on diagnostic aspects and 41 articles focusing on prognostic aspects. Pooled odds ratio, 95% confidence intervals (CIs), and hazard ratio (HR) were utilized to evaluate clinical parameters and overall survival (OS), respectively. Result The findings indicated that area under the curve, sensitivity, and specificity were 0.85, 75%, and 80% in cancer, respectively. There was a possibility that snoRNAs had a positive impact on the diagnosis (risk ratio, RR = 2.95, 95% CI: 2.75-3.16, P = 0.000) and OS (HR = 1) in cancer. Additionally, abnormally expressed snoRNAs were associated with a positive impact on OS time for chronic lymphocytic leukemia (HR: 0.88, 95%Cl: 0.69-1.11, P < 0.00001), colon adenocarcinoma (HR: 0.97, 95%Cl: 0.91-1.03, P < 0.0001), and ovarian cancer (HR: 0.98, 95%Cl: 0.98-0.99, P < 0.00001). However, dysregulated snoRNAs of colon cancer and colorectal cancer had a negative impact on OS time (HR = 3.01 and 1.01 respectively, P < 0.0001). Conclusion The results strongly suggested that snoRNAs could serve as potential novel indicators for prognosis and diagnosis in cancers. This systematic review followed the guidelines of the Transparent Reporting of Systematic Review and Meta-Analyses (PROSPERO register: CRD42020209096).