Task-based language assessment (TBLA) has generated interest since the early 1990s, primarily in conjunction with the ongoing development of task-based language teaching (TBLT) and the pursuit of developing appropriate testing models for this approach to pedagogy (Norris, 2002, 2009). However, tasks also offer considerable advantages for language assessment, beyond their obvious relevance within TBLT classes and programs. In fact, major innovations in the general domain of language assessment over the past two decades have occurred in conjunction with the introduction of tasks into assessment design, largely in response to the need for tests that better represent examinees’ abilities to use the language (Mislevy, Steinberg, & Almond, 2002), but also because tasks offer a meaningful space for language teachers, testers, learners, and others to examine, understand, and improve language learning endeavors (Van Gorp & Deygers, 2013). This article reviews the considerable range of current uses for TBLA, illustrating different types of assessment with concrete examples and highlighting distinct roles for tasks as a basic unit of analysis in test design, interpretation, and intended consequence. Ultimately, it argues that tasks offer a fundamental, though not exclusive, foundation for useful language assessment, and that task-based assessment, though challenging, is probably worth the effort.