Background: Negative lymph node (NLN) count has been reported to associate with the prognosis of various cancers. This study aims to reveal the prognostic value of NLN count in breast cancer. Methods: Clinical characteristics of patients were collected from the Surveillance, Epidemiology, and End Results (SEER) database. The X-tile program was used to determine the optimal cutoffs for NLN count. Univariate and multivariate analysis were used to assess the risk factors for breast cancer-specific survival (BCSS). Results: The X-tile program identified that cutoff value of 2 and 10 could divide the patients into high, middle and low risk subgroups. According to multivariate analysis, patients with NLN count <= 1, over 60 years old, being black, higher tumor grade, higher T or N stage, negative hormone receptor, no radiotherapy or no chemotherapy would more likely suffer poor survival outcome. Subgroup analysis showed that NLN count could still predict survival independently. Conclusions: NLN count is a potentially effective predictor of breast cancer and is a good supplement for N stage and TNM stage. Combining NLN count with other prognostic factors will be a better predictor for the survival of breast cancer patients.