Cervical lymph node metastases in oral squamous cell carcinoma (OSCC) are key predictors of disease specific survival. It was therefore the aim of this study to evaluate how much imaging is minimally needed for reliable and efficient identification of cervical lymph node metastases. In this retrospective cross-sectional study, results (metastasis yes/no) of ultrasound (US), computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET) were compared to the final histopathological results of the corresponding neck dissection (ND) specimens (metastasis yes/no). A score was calculated to account for cervical lymph node size, shape, clustering, peripheral enhancement, hilus sign, architecture, blood flow, and central necrosis. Sensitivity and specificity were analyzed for each imaging technique separately. In 164 patients diagnosed with OSCC, 96 underwent uni- or bilateral ND (122 ND in total). One hundred percent sensitivity was achieved by CT+MRI, MRI+PET, US+CT+MRI, US+MRI+PET, CT+MRI+PET, and US+CT+MRI+PET. The highest specificity was realized by US with 79% (95% CI [0.698–0.890]). Specificity for CT+MRI and PET+MRI was 51% (95% CI [0.359–0.665]) and 70% (95% CI [0.416–0.984]), respectively. Regarding 100% sensitivity with acceptable specificity, the combination of CT+MRI or PET+MRI appeared to be suitable for staging cervical lymph nodes in primary OSCC.