Although there is a large and increasing body of research on the use of corpus data by language teachers and learners, the language teachers in question are also the researchers reporting on their use of corpora with their own learners. There is thus a gap, which this plenary speech addresses, between this research and the practice of language teachers who are not corpus linguists. After a brief account of the successful integration of corpus data in the production of language learning resources such as dictionaries, grammars and course books, the potential benefits of having direct access to corpora by teachers and learners are discussed. Following this, a number of publications on the use of corpora in the language classroom, both in higher education and at secondary level, are examined, focussing on how easy or difficult it would be for teachers to replicate the use of corpus data in their own classrooms. This leads to a final section where possible solutions to the research–practice gap are considered, involving language teacher education, publishers of course books, and research involving the use of corpora in the classroom by language teachers who are not also corpus linguists.