From a histopathologic perspective, head and neck squamous cell carcinoma (HNSCC) is a relatively straightforward diagnosis. However, the clinically useful information presently provided by pathologists is embarrassingly limited. Similarly, our ability to accurately identify the earliest premalignant lesions as well as predict which premalignant lesions are likely to progress to HNSCC is limited. Over the last decade, an explosive growth of knowledge related to the molecular biology of this disease has occurred, which is now being used to address these issues. For example, we now appreciate that there are multiple etiologies and multiple molecular mechanisms responsible for the development of HNSCC. New techniques have improved our ability to identify molecularly premalignant, but histologically benign lesions. Similarly, recent studies have been able to predict which premalignant lesions are likely to progress to HNSCC. In addition to having utility in the realm of early diagnosis, molecular diagnostics may have a profound impact on how we diagnose and report HNSCC. While still in the developmental stage, molecular protocols are being used to evaluate surgical margins, determine the location of unknown primary tumors, identify histologically undetectable lymph node metastasis, and predict which tumors are more likely to respond to a particular postsurgical adjuvant therapy.