Personality is typically defined as the consistent set of traits, attitudes, emotions, and behaviors that people have. For several decades, a majority of researchers have tacitly agreed that the gold standard for measuring personality was with self-report questionnaires. Surveys are fast, inexpensive, and display beautiful psychometric properties. A considerable problem with this method, however, is that self-reports reflect only one aspect of personality — people's explicit theories of what they think they are like. We propose a complementary model that draws on a big data solution: the analysis of the words people use. Language use is relatively reliable over time, internally consistent, and differs considerably between people. Language-based measures of personality can be useful for capturing/modeling lower-level personality processes that are more closely associated with important objective behavioral outcomes than traditional personality measures. Additionally, the increasing availability of language data and advances in both statistical methods and technological power are rapidly creating new opportunities for the study of personality at ‘big data’ scale. Such opportunities allow researchers to not only better understand the fundamental nature of personality, but at a scale never before imagined in psychological research.