This paper examines the possibilities that big data-based approaches offer to language research. In a nutshell, the term “big data” makes reference to the massive amount of data that users generate in their digital interactions and whose great volume and heterogeneous nature typically requires a specialized treatment. The chapter starts by reviewing the main characteristics of big data, and then focuses on the possible problems arising from the use of big data in linguistic analysis. The following section offers a review of specific studies that apply this big-data approach to the study of multimodality: an approach to language study that includes not only the verbal component but also multimodal aspects such as gestures or intonation. The paper concludes with a review of the advantages and problems of using this type of data.