The advent of next generation sequencing (NGS) technologies has advanced our understanding of the intrinsic biology of different gastrointestinal (GI) tumor types. The use of novel, more efficient sequencing platforms has improved turnaround times of sequencing results. This is providing real time opportunities to put precision medicine to the test. A number of early phase clinical trials are testing targeted therapies in unique molecularly characterized subsets of patients (baskets). While basket studies are gaining momentum, treatment failures serve to remind us that shifting from a histology-driven to a histology-agnostic approach is unlikely to be a failure-free strategy for a number of tumor types as recently learnt from vemurafenib failure in BRAF mutated metastatic colorectal cancer (mCRC). GI malignancies are clinically and molecularly heterogeneous. Unfortunately, development of biomarkers of response to therapy as well as targeted therapies for GI adenocarcinomas has fallen behind compared to other malignancies. Trastuzumab is the only FDA approved targeted therapy for GI adenocarcinomas for which a biomarker of response (HER2 amplifications) is available. In addition, RAS mutations are known to predict lack of response to epidermal growth factor receptor (EGFR) inhibitors in advanced colorectal cancer (CRC) patients. However, NGS has recently revealed that a number of actionable genetic aberrations are present at low prevalence across different GI malignancies. Prospective randomized clinical trials will determine whether matching actionable aberration with targeted therapy will contribute to improve survival in patients with GI malignancies. Here, we review current evidence for targeted therapies in GI malignancies, as well as application and pitfalls of NGS including tissue testing and liquid biopsies.