Since the 1960s, the use of large electronic text collections, or corpora, has evolved from educator-focused applications, such as building dictionaries and textbooks, to more learner-centred ones. This shift, exemplified by data-driven learning (DDL), empowers learners to autonomously explore linguistic inquiries using language data. This entry discusses the educational outcomes of DDL, particularly its effectiveness in second-language (L2) learning where authentic input is limited. DDL employs text-based corpora to provide authentic contexts for vocabulary and multiword units, aiding learners in recognizing language patterns and making incremental progress. However, challenges such as the relevance and comprehensibility of corpus data and the complexity of using concordancers need to be addressed, for example, by using tailored/simplified corpora and providing training opportunities for effective DDL implementation. Previous meta-analyses highlight DDL’s significant benefits for writing, translation and vocabulary acquisition. Today’s students are adept at using technology, such as Google, for language queries, and advances now allow for the integration of multimedia into corpora. This enhances language learning by providing textual, verbal and visual contexts. Further research is necessary to explore DDL’s full potential as a language learning tool.