Chronic liver disease is a major contributor to global morbidity and mortality and may result from a wide range of insults, including non-alcoholic fatty liver disease (NAFLD), alcohol liver disease (ALD) and results in a number of complications including cirrhosis, liver failure and hepatocellular carcinoma (HCC). Despite its prevalence and burden, there remains a number of major unmet needs in the diagnosis and management of liver disease including accurate risk stratification for HCC, the need for better diagnostic tools and for novel therapeutic interventions. We highlight 3 examples of liver disease (NAFLD, ALD and HCC) where the generation and analysis of big data using high throughput technologies may impact the development of clinically useful biomarkers (objectively measured characteristic that describes a normal or abnormal biological state in an organism by analysing biomolecules) and novel therapies in liver disease. In the five research papers presented, we show how big data can impact the development of predictive biomarkers in alcoholic steatohepatitis (publication 1) and bariatric surgery for patients with NASH and morbid obesity (publication 2). We present an example of therapeutic discovery of anti-fibrotic agents using transcriptomics and gene signatures (publication 3). Finally we present 2 papers assessing the utility and cost-effectiveness of molecular-based risk stratification of HCC risk in the context of HCC surveillance (publication 4) and identifying clinical features associated with HCC molecular subclasses aiding the discovery of novel potential biomarkers of HCC molecular subgroups (publication 5). We discuss our ongoing projects including using transcriptomic meta-analysis of liver tissue of animal models and human NAFLD to identify the “ideal” animal model of NAFLD for translational purposes and therapeutic discovery. Another project focuses on the identification of novel biomarkers for stratifying HCC risk in NAFLD subjects to better target patients at highest risk of HCC with more intensive interventions. We conclude by discussing caveats and limitations of the use of big data, in particular areas where human input will remain irreplaceable.