Mastery is the state in which a learner can reliably recall and apply the knowledge or skill that a learning unit was designed to develop, and it is the outcome that all other metrics — completion rate, watch-time, engagement — are ultimately meant to predict but cannot directly measure. It is operationalised through assessment: a post-test, a performance task, a scored simulation, or a spaced-repetition recall check administered after a delay. The key conceptual distinction is between completion and mastery: a learner can finish every video and pass every embedded quiz with fluent-seeming answers and still fail a delayed retention test two weeks later — a phenomenon called the fluency illusion. In adaptive learning systems mastery gates progression: the learner is routed to additional or alternative content until their assessment performance crosses a defined threshold. At that point an xAPI statement with the verb "mastered" or "passed" is sent to the LRS, recording the achievement in a form that downstream systems such as HR or compliance platforms can query. The threshold itself is a design choice with real consequences — set too low it is a rubber stamp, set too high it creates frustration and drop-off. Mastery data, aggregated across a cohort, is the most honest answer to the business question "did the training work?" and the metric that justifies learning investment to stakeholders.
