A high-quality body of evidence supports the use of aspirin in reducing sporadic and hereditary adenomatous polyps, and numerous observational studies point to a reduction in colorectal cancer (CRC) risk. However, using aspirin as an adjuvant therapy in established CRC was until recently inconceivable. Now, evidence from both observational and clinical trials of aspirin for other indications suggests that aspirin initiation after (or before) the diagnosis of CRC improves CRC-specific mortality. These exciting findings need to be confirmed in prospective randomized trials that are underway. The recent failure of adjuvant irinotecan, bevacizumab, and cetuximab clinical trials compels us to reconsider our assumptions and paradigms for treating CRC. In this Review, we summarize clinical and preclinical evidence supporting aspirin use in established CRC and outline a framework for better understanding aspirin activity in the pathogenesis of CRC. We describe the data supporting adjuvant aspirin in resected CRC, including the issues of dose, duration and toxicity, and discuss potential biomarkers that may help better select patients for aspirin therapy.