Recent evidence shows that COL3A1 promotes the progression of many types of cancer. The purpose of our study is to explore the correlation between COL3A1 and the prognosis of patients with head and neck squamous cell carcinoma (HNSCC) and its potential mechanism. We initially screened the differentially expressed gene COL3A1 in The Cancer Genome Atlas (TCGA) database, and the association between the expression level of COL3A1, prognosis, and the clinical parameters of HNSCC patients was verified. A nomogram was constructed according to the multivariate analysis results. Next, a heatmap of COL3A1 co-expressed genes was constructed in TCGA database. The TargetScan database is used to explore the microRNAs (miRNA) related to COL3A1. The starBase database was used to explore and predict the long non-coding RNAs (lncRNAs) that the candidate miRNAs might bind to. Finally, the potential mechanism of action was investigated using Gene Set Enrichment Analysis (GSEA). COL3A1 expression is elevated in HNSCC tumor tissues, and HNSCC patients with high COL3A1 expression have worse prognostic factors. COL3A1 was positively correlated with the central carbon metabolism-related proteins: epidermal growth factor receptor (EGFR), phosphoglycerate mutase 1 (PGAM1), hexokinase 3 (HK3), and phosphofructokinase, platelet (PFKP). The TargetScan database showed that the best candidate miRNA for binding to the three prime untranslated region (3'UTR) end of COL3A1 mRNA was hsa-miR-29b-3p, which was negatively correlated with COL3A1. The starBase database showed that the lncRNA X Inactive Specific Transcript (lncRNA XIST) was the best candidate upstream non-coding RNA for regulating hsa-miR-29b-3p. GSEA showed that COL3A1 may be involved in the poor prognosis of HNSCC by participating in carbon metabolism, glucose metabolism, oxidative stress, and the Wingless-Type MMTV Integration Site Family (Wnt) and vascular endothelial growth factor A-vascular endothelial growth factor receptor 2 (VEGFA-VEGFR2) pathways. Low COL3A1 expression can be employed as a new HNSCC predictive biomarker, and the prognosis of HNSCC patients with lower COL3A1 expression can be greatly improved. At the same time, we found that the lncRNA XIST/miR-29b-3p/COL3A1 axis may regulate the central carbon metabolism of HNSCC and is associated with poor prognosis. These findings point to a potential target for developing HNSCC anticancer therapies.