Background: Regional differences in cervical cancer survival have been reported in several countries. They may result from disparities in access to early diagnostic services, timely referral or appropriate treatment. Estimates of survival by stage at diagnosis could help to distinguish whether lower-than-expected survival is due to late-stage diagnosis or sub-optimal management, and to inform health-policy makers for resource allocation. In this retrospective cohort study, we aim to provide a detailed and up-to-date analysis of cervical cancer survival in Saudi Arabia by stage and region, and to explore whether any differences in survival between regions are due to differences in stage at diagnosis. Methods: Data on all women diagnosed with invasive cervical cancer during 2005–2016 were obtained from the Saudi Cancer Registry (SCR). Vital status and date of death if dead were ascertained by linking the registry records to vital registration data in the National Information Centre (NIC) of the Ministry of Interior. Women for whom no death record existed on the day of record linkage were considered to be alive. We estimated age-standardised five-year net survival using the Pohar-Perme estimator for women diagnosed during 2005–2010 and 2011–2016. Survival was also estimated by region and stage at diagnosis, and by region stratified by stage. Results: Age-standardised 5-year net survival did not change in Saudi Arabia between 2005–2010 [59.2%; 95% confidence interval (CI) 52.7–65.7%] and 2011–2016 (59.7%; 54.7–64.6%), or in any of the regions, except Makkah, where there was a 19% increase in survival for women diagnosed during 2011–2016 compared to 2005–2010. Survival for women diagnosed at a distant stage was substantially lower in the Eastern Region than in other regions. Conclusions: Cervical cancer survival has remained largely unchanged. Higher survival could be achieved by improving early diagnosis and access to high-quality treatment.