Research on the implementation of CLIL at the onset of primary school is limited and has largely overlooked the role of other sources of individual differences. This study investigated the effects of the CLIL approach to English learning, together with the effects of out-of-school exposure to the language through media and other sources of individual differences, in a sample of Grade 1 students in Catalonia (Spain) using a longitudinal design. Participants (N = 176) from 14 different schools completed a test battery at the beginning and end of Grade 1 that assessed receptive and productive English skills. Results revealed that abilities at the onset of Grade 1 were the best predictor of abilities at the end of the year, and that CLIL was not associated with additional advantages in the students that followed the approach. In addition, certain characteristics of the linguistic and family background of participants predicted additional gains during the academic year: participants who engaged in more English extracurricular activities and participants with more educated mothers performed better at the end of Grade 1.