The agent partially satisfied the task by assembling a comprehensive overview of Haufe Akademie from fragmented web sources and prior knowledge, but the site itself was poorly navigable for machine extraction. The homepage and most catalog pages returned incomplete or unreadable HTML; the agent relied heavily on web search results and cross-referenced external platforms (Capterra, SoftwareFinder) to answer questions about what they do, who they serve, and how they differ. Pricing remained intentionally opaque—no transparent pricing was discoverable on the site itself, forcing the agent to advise users to contact sales directly.

• ›Steps [3], [13], [15], [21] (the About, Corporate Learning, and Personal Development pages) returned valid 200 responses but HTML was truncated at 4400 chars in the fetch log—the agent could not extract full content from these pages and had to infer meaning from meta descriptions and page titles alone.
• ›The site structure is fragmented: the homepage (/index) uses old technology, while modern content pages (e.g., /ueber-uns, /corporate-learning) are hosted on Webflow subdomains with separate data attributes. The agent had to guess URLs like /seminare, /preise, /kataloge (all 404), indicating poor URL schema discoverability.
• ›Pricing information was entirely absent from the website—no seminar costs, no LMS pricing transparency, no cost-per-participant benchmarks. The agent only discovered LMS pricing (€1.49/employee/month) through external SaaS comparison platforms (Capterra, SoftwareFinder), not from haufe-akademie.de itself.
• ›The agent relied on 7 web searches (38% of fetches) to compensate for poor on-site navigation and content readability. External sources (Semigator, Capterra, SoftwareFinder) became the primary source of structured facts about offerings, audience, and differentiation—the site itself did not surface this metadata.
• ›Page [19] (a specific seminar detail page) returned HTML but was not readable; the agent could not extract pricing or content details and could not cite it as a source, indicating the site requires JavaScript rendering to be agent-readable.
• ›The agent correctly identified the dual business model (training services vs. LMS software) but noted this as a point of confusion—the website does not cleanly separate or position these two offerings, making the value proposition ambiguous for crawlers.