Abstract Background We aimed to evaluate the prognostic significance of preoperative neutrophil‐to‐albumin ratio (NAR) in oral squamous cell carcinoma (OSCC). Methods A total of 622 patients with surgically treated OSCC were enrolled. NAR was defined as the absolute neutrophil count divided by the serum albumin level in peripheral blood before the radical surgery. Cox proportional hazards model were used to discover survival outcome‐associated factors. Results The optimal cut‐off of NAR to predict overall survival (OS) was determined to be 0.1. In Cox model, high NAR was identified as an independent negative prognosticator of OS, cancer‐specific survival, and recurrence‐free survival (adjusted hazard ratio: 1.503, 1.958, and 1.727, respectively; all p  < 0.05). The NAR‐based nomogram accurately predicted OS (concordance index: 0.750). Conclusion Our study suggests that preoperative NAR is a convenient and effective prognostic marker for OSCC and NAR‐based nomogram can be a promising prognostic tool in clinical setting.