All Case Studies
case-studyJune 18, 2026 ZENO Team 5 min read

Answer Engine Optimization in Healthcare: Zeno Visibility Builds Schema.org JSON-LD and Content Clusters for Greater LLM Visibility

Answer Engine Optimization in…

← All Cases

Initial Situation

RheinMed Diagnostik is a mid-sized healthcare group in the DACH region with 19 locations, around 1,050 employees, and a focus on radiology, laboratory diagnostics, and preventive medicine. For years, the company has published specialist content on diagnostic procedures, symptoms, preventive care, and patient education. The website had a solid organic reach, but visibility in AI search and answer systems was weak.

In spring 2025, the marketing team faced a clear finding: for queries such as “What does an MRI cost without a referral?”, “How do I prepare for a CT scan?”, or “Which blood values indicate iron deficiency?”, ChatGPT, Perplexity, and Gemini mostly mentioned competitors, specialist portals, or generic health platforms. The company’s own brand appeared only rarely as a citable source. At the same time, internal effort for content planning, medical approvals, and technical maintenance kept increasing.

The goal was therefore not just more traffic, but a reliable answer to the question of how RheinMed Diagnostik could become visible as a trusted source in generative systems. The project was set up as GEO Generative Engine Optimization, with a focus on semantic authority, structured data, and measurable LLM visibility.

Challenge

The core problem was not the amount of content, but the lack of semantic connections. The content was organized in topic islands, but not built as a complete authority system. There were individual guides, but hardly any robust hub pages, hardly any FAQ structures, and no systematically maintained Schema.org JSON-LD data.

As a result, the brand was difficult for answer systems to classify. The content answered questions, but provided too few machine-readable signals about entities, relationships, experts, service areas, and document types. In addition, internal linking was inconsistent, which weakened the topical authority of individual pages.

From a business perspective, this led to three effects: first, the share of qualified entry points from informational queries declined. Second, dependence on paid traffic increased. Third, the team lacked a measurement model to assess the impact of content on LLM visibility and recommendation frequency. In a regulated environment like healthcare, it was also important to structure medical statements properly and secure them editorially.

Solution Approach

RheinMed Diagnostik chose an architecture built from content clusters, Schema.org JSON-LD, and continuous monitoring of LLM presence. This was implemented with Zeno Visibility, which in this project was used not as a pure monitoring tool, but as infrastructure for AI Authority.

First, Zeno Visibility’s research engine analyzed the brand presence across ChatGPT, Gemini, Perplexity, Claude, and Copilot. Based on this, an initial Semantic Authority Score was defined and linked to 42 prioritized topic and prompt variants. At the same time, the team identified the 12 search intents with the highest potential for patient education and referring-physician communication, including MRI, CT, lab values, prevention, and pre-procedure education questions.

On this basis, the Authority System Builder generated a complete content system for each core keyword:

  • 8 hub pages with a clear topic architecture
  • 36 specialist articles and guides
  • 54 FAQ modules
  • 11 comparison pages and decision aids
  • 6 case studies and process pages for trust and transparency
  • All content was medically reviewed, tagged with clear entities, and embedded in an internal linking structure that clearly mapped topical proximity and relevance. In addition, Zeno Visibility automatically generated Schema.org JSON-LD for Article, FAQPage, Organization, MedicalClinic, and BreadcrumbList. This was especially important in the healthcare context because generative systems prefer structured, reliable signals when choosing sources.

    Technically, the rollout was implemented via WordPress and Contentful. For the editorial team, direct CMS exports were used to increase publishing speed and standardize approval processes. After twelve weeks, the new structure was live. After another eight weeks, optimization began based on the LLM monitoring data: content with low citation likelihood was expanded, FAQ wording was adjusted, and internal links were reassigned based on semantic proximity.

    Results

    After six months, a measurable effect on visibility and demand became apparent:

  • LLM mentions of the brand in relevant answers increased from 8% to 31% in the prioritized topic clusters.
  • Citations from the company’s own content in answer systems improved from 0 to 22% of the tested prompts.
  • Organic clicks on cluster pages increased by 67% compared to the same period last year.
  • Qualified appointment and contact inquiries from informational content entry points rose by 24%.
  • The Semantic Authority Score increased by an average of 38 points across all prioritized clusters.
  • The time to publish new specialist content dropped from an average of 8.5 days to 2.1 days.
  • The economic impact was also solid: thanks to reduced dependence on Paid Search and higher conversion from organic entry points, an estimated ROI of 3.6x on project costs was achieved within six months. What was especially relevant was that the brand was not only mentioned more often, but more often as a source with professionally relevant content. That shift is exactly what matters for GEO Generative Engine Optimization.

    Lessons Learned

  • Content volume does not replace authority. Only hub structures, internal linking, and consistent entities make content citable for LLMs.
  • Schema.org is not a detail in healthcare, but a foundation. JSON-LD improves not only readability, but also the attribution of expertise, services, and context.
  • LLM visibility must be measured. Without monitoring across multiple models, it remains unclear which content actually appears in answers.
  • Editorial and technical teams must work together. High visibility does not come from individual articles, but from a scalable authority system.
  • GEO needs operational processes. Anyone who wants visibility in answer systems must organize publishing, maintenance, and optimization as a repeatable process.
  • Summary

    With Zeno Visibility, RheinMed Diagnostik moved from isolated guide content to a systematic authority model. The decisive factors were structured data, semantically linked content clusters, and continuous monitoring of presence in AI answer systems. The result was not only greater reach, but measurably better visibility in generative search and answer environments.

    More Case Studies

    View all Case Studies →

    KIGEO Generative Engine Optimization