Digital flashcard users typically must choose between creating their own flashcard content or using freely available flashcard sets. The latter is more convenient and saves time, but is it more effective for learning? We conducted six experiments, each involving the use of user-generated or premade flashcards to learn material drawn from educational text passages, followed by a 48-hr delayed criterial test. Different approaches to generating content and variations in the quality of premade content were also examined. Across experiments, user-generated flashcards improved memory relative to premade flashcards (an estimated advantage of d = 0.45, 95% CI [0.25, 0.66]), and in most cases, enhanced performance on application questions (an estimated advantage of d = 0.29, 95% CI [0.12, 0.45]). These results suggest that generating one's own flashcards enables productive learning processes that enhance memory and comprehension. Accordingly, digital flashcard users may benefit from eschewing premade versions in favor of making their own.