Metacognition is a broad term that means different things to researchers in different sub-areas. A major contribution of Anastastia Efklides is to bring together disparate approaches in metacognition under one theoretical perspective. In this paper, we examine the concept of fluency and how it has been employed in metacognition research. Fluency-based judgments are generally considered to be the primary source of inaccuracy of metacognitive judgments as well as the primary reason why metacognitive control goes astray in self-regulated learning. We discuss how and when fluent processing influences metacognition, when fluency leads to accurate judgments and when it leads to illusions.