This quantitative study examined the effect of word familiarity within collocational components on the rate of learning collocations by comparing mixed-familiarity collocations, where only one word is known to the learner, with high-familiarity collocations, where both component words are known. For this purpose, 27 participants from a pool of 97 available Iranian EFL learners were selected through two pretests. Pretest 1 assessed the participants’ prior knowledge of mixed-familiarity collocations whereas Pretest 2 was meant to ensure that they had no familiarity with high-familiarity collocations. Each participant received their own specific list of six mixed-familiarity and six high-familiarity collocations (along with their counterparts translated into Persian and an English exemplar sentence) and two exercises as the instructional materials for intervention. The results of immediate and delayed posttests, analyzed via a two-way repeated measures ANOVA, revealed that learning collocations with only one known component word or those containing both familiar component words were equally effective (η = .917). EFL teachers might confidently introduce collocations even when one component word is known to learners, as partial familiarity might not hinder acquisition. The findings support early integration of collocations into vocabulary instruction to enhance fluency.