ImportanceManagement of cervical unknown primary squamous cell carcinoma (CUP) has evolved with the introduction of transoral robotic surgery (TORS). Objectives1. To describe the efficacy of TORS lingual and palatine tonsillectomy in identifying the primary site of malignancy. 2. To explore how the extent of surgery affects diagnostic yield. 3. To report margin status of TORS resections. Design, setting and participantsA retrospective, single-center cohort study utilizing a prospectively collected database of CUP patients in a high-volume tertiary referral center. Patient underwent operative laryngoscopy plus TORS as clinically indicated. Main outcomes and measuresPrimary end point was successful identification of the primary. The extent of surgery and margin status were also examined. ResultsFrom 2010–2016, 64 patients with CUP were treated. The primary tumor was found in 51 patients (80%). Fourteen patients (22%) were identified with operative laryngoscopy alone. Fifty patients underwent TORS lingual tonsillectomy ± palatine tonsillectomy with 37 primary tumors identified (74%). The primary was located in the lingual tonsil in 32 patients (86%) and palatine tonsil in 5 patients (10%, p<0.001). Negative margins were achieved in 19 patients (51%). The deep margin was the most commonly positive margin (47%, p=0.049). Conclusion and relevanceOperative laryngoscopy with TORS is efficacious, localizing the primary in 80% of patients. If a margin was positive, it was most commonly the deep margin. This study provides valuable information that can help standardize surgical technique, further increasing the diagnostic yield and decreasing the negative margin rate of TORS for CUP.