Experts are frequently called upon to serve as instructors in their disciplines. However, studies indicate that their unaided explanations contain significant inaccuracies and omissions that negatively impact the effectiveness of instruction. Cognitive task analysis (CTA) is an effective tool for eliciting, analyzing, and representing expert knowledge in a more accurate and complete manner. CTA-based instruction is consistently found to be more effective than unaided instructional explanations provided by experts. However, prior studies of the efficacy of CTA-based instruction have included a confound between the source of content (CTA) and the instructional design decisions made by designers. The current study utilizes content analysis of the instruction from the CTA and non-CTA conditions of an ongoing experment to assess the relative contributions of instructional content and instructional design to the overall effectiveness of CTA-based instruction. Results indicate that CTA’s ability to elicit more specificity in instructionally relevant information accounts for significantly more variance than the instructional design decisions related to representing instructional content inprocedural or conceptual formats.