BACKGROUND Recurrence of oral squamous cell carcinoma (rOSCC) after primary treatment is associated with poor survival outcomes. Salvage treatment with further surgery, radiotherapy and chemotherapy has high morbidity, making patient selection crucial. However, in the recurrence setting, reliable stratification data are scarce. Decision-making in this context is consequently complex. We investigated factors influencing overall survival after rOSCC. METHODS Retrospective cohort study of patients with rOSCC (n = 83) at the Queen Elizabeth Hospital Birmingham, UK between 2006 and 2016. Associations with overall survival were analysed using univariate and multivariate analyses to identify important clinical prognostic indicators. RESULTS Overall survival at 1 year was 32.5% and at 5 years was 18.1% after a median follow-up of 7.4 months. Multivariate analysis identified four independent predictors of overall survival following rOSCC: size of primary tumour (HR 2.077; 95% CI 1.034-4.172), extent of recurrent disease (HR 3.286; 95% CI 1.545-6.991), history of moderate alcohol consumption (HR 0.351; 95% CI 0.162-0.763), and close or positive margins at primary resection (HR 1.955; 95% CI 1.063-3.595). CONCLUSIONS We identified four key factors that help prognostication and risk stratification of rOSCC. Given the high morbidity associated with salvage treatment, we recommend that the multidisciplinary team (MDT) and the patient weigh these factors carefully when considering further treatment. Further investigation of the biology underlying these oncophenotypes may contribute to better patient stratification.