Elevations in blood lactate concentrations have been studied in sepsis and other disease states for decades and are well known to be associated with increased mortality. Many studies have also demonstrated the prognostic accuracy of serial lactate levels, and some have suggested that lactate clearance may be a useful therapeutic target for resuscitation. Lactate measurements have therefore gained an increasingly prominent role in sepsis definitions, screening protocols, management guidelines, and quality measures over the past two decades. The heavy emphasis on lactate monitoring, however, has also generated controversy and concerns. Lactate is not specific to infection and its frequent use for sepsis screening and diagnosis may therefore trigger unnecessary broad-spectrum antibiotic use in some patients. Because hyperlactatemia does not always reflect fluid-responsive hypoperfusion, titrating resuscitation to lactate clearance can also lead to unnecessary fluid and volume overload. More broadly, there is a lack of high-quality evidence demonstrating that initial and serial lactate monitoring leads to better patient-centered outcomes. Indeed, a recent randomized controlled trial comparing resuscitation strategies based on lactate clearance versus normalizing capillary refill time showed no benefit and potential harm with lactate-guided therapy. In this article, we review the basic pathobiology of lactate metabolism and delineate why the traditional paradigm that hyperlactatemia reflects tissue hypoxia is overly simplistic and incomplete. We then review the evidence behind the diagnostic, prognostic, and therapeutic uses of lactate monitoring and place this in the context of evolving sepsis diagnosis and management guidelines.