AI Visibility
measured18/100
Emerging · in Education & LMS
counts brand mentions in AI answers, not source citations
AEO Readiness
fundamentals38/100
Weak fundamentals
Avg Prompt Score
4
across 115 prompts
AI Share of Voice
2%
across 106 prompts
Critical Issues
5
critical + high
How AI Visibility breaks down
in Education & LMS
Composite = 50% coverage + 30% engine breadth + 20% position, measured in the brand's strongest category. Position and citation depth join the score once the aggregation pipeline captures them. This is the measured Visibility score — separate from AEO Readiness, which scores fundamentals.
Cited rate · share of voice · engine consensus · sentiment, broken out by buyer-journey stage. Sentiment is the net positive−negative skew across engines that cited the brand at this stage.
Harvard University is likely to be cited for category-level “best online learning platform” discovery queries, but it is unlikely to be cited as a software purchase/evaluation option for employee training because harvard.edu does not present itself like a B2B SaaS product with crawlable SoftwareApplication/FAQPage evaluation surfaces. The single highest-ROI fix is to create first-party, crawlable evaluation content that directly answers the evaluation and conversion query patterns (especially pricing and “compare to” prompts) in a machine-extractable format.
Based on audit of harvard.edu · Aug 29, 2026