Video continues to be used in many basic communication courses as a way for students to self-evaluate speechmaking. In this study, students (N = 71) presented speeches, viewed the video recordings, and produced self-generated feedback. Comparing student's self-estimated grades from the self-evaluation against earned grades resulted in composite scores that were organized in three estimator categorizations: under-, accurate-, or overestimators. The estimators' feedback was analyzed by new software, the Meaning Extraction Helper, to identify specific themes. The common content words that emerged provided themes that were then inductively categorized. Results indicated unique categories/subcategories for each estimator type and minor subcategory similarities across estimators. Findings discuss estimator perceptions influencing student understanding of their speaking skills.