The present meta-analysis aimed to summarize the extent to which second language vocabulary is learned from the most frequently researched word-focused activities: flashcards, word lists, writing, and fill-in-the-blanks. One hundred effect sizes from 22 studies were included in meta-regression analyses and administered separately for the observations measured with meaning-recall and form-recall tests. The results revealed that the average percentage learning gains were 60.1% and 58.5% on meaning-recall and form-recall immediate posttests. These gains dropped to 39.4% and 25.1% on delayed meaning- and form-recall tests, respectively. These results suggest that learning through word-focused tasks is far from guaranteed. Moreover, the percentage learning gains among the different activities ranged from 18.4% to 77.0% on immediate posttests and from 23.9% to 73.4% on delayed posttests indicating that there is much variation in efficacy among the activities. Moderator analyses revealed that learners' place of study and direction of learning affected learning.