Research on Computer Assisted Language Learning (CALL), Mobile Assisted Language Learning (MALL), and Computer Mediated Communication (CMC) has informed us that the adaptation of new technologies helps in overcoming some of the challenges faced in language classrooms; such as the limited classroom time (Cardoso & Collins, 2016). In light of CALL, MALL, and CMC research, the purpose of this corpus driven study was to investigate the potential of the Instagram platform in learning English as a foreign language (EFL): particularly whether the type of Instagram post (vocabulary or grammar) had an effect on the amount of learners’ EFL output, the output accuracy, and the amount of feedback the learners received. The data was collected from authentic EFL use from the comments section of 15 Instagram accounts that were targeting the Saudi learners as their population. A total of 140 comments were analyzed (70 for vocabulary and 70 for grammar). A non-parametric Mann-Whitney test was carried out and indicated a statistical significance Z (140)= -2.38, p. = 017 for output, with a relatively small effect size (d = .438), showing that vocabulary posts elicited more output from the commenters. However, post types did not have any influence on learners’ output accuracy and the amount of feedback they received. From a pedagogical standpoint, teachers are encouraged to use social media as means for creating language practice opportunities, and as a source of extra input outside the classroom.