Generative Engine Optimization in Mechanical Engineering: Zeno Visibility Builds a Semantically Interconnected Authority System for a Mid-Sized Industrial Company
Generative Engine Optimization in…
Initial Situation
A mid-sized mechanical engineering manufacturer from southern Germany, with approximately 420 employees and annual revenues in the high double-digit millions, wanted to systematically expand its visibility in AI-powered answer systems. The company develops components for automated production lines and sells primarily in the DACH region, supplemented by selected partners in Benelux and Eastern Europe. Its traditional lead channels had been stable for years: organic search, trade shows, existing customers, and technical sales. At the same time, a clear shift in target audience search behavior became apparent starting in 2024.
Internal analyses revealed that technical decision-makers, procurement managers, and maintenance supervisors were increasingly using ChatGPT, Gemini, Perplexity, and Copilot for preliminary research before visiting websites or engaging in sales conversations. The marketing team recognized that the brand was only sporadically mentioned in generative responses — and often without accurate classification of its core products. An initial LLM Brand Monitoring exercise across four relevant models found that only 18% of product-related queries returned a correct brand mention. At the same time, the content landscape was fragmented: there were product pages, individual technical articles, and PDFs, but no cohesive semantic system that logically connected topics, use cases, comparison questions, and technical explanations. This is precisely where the collaboration with Zeno Visibility began.
Challenge
The core problem was not a lack of content volume, but a lack of machine-readable authority. The existing content was understandable to humans, but only partially usable by LLMs: too little thematic interconnection, too little entity clarity, too little structured data. Particularly critical was the fact that the brand rarely appeared as a preferred source in generative responses, despite the company's strong subject-matter expertise within its niche.
The impact was measurable. Organic visibility for several strategic keywords did not decline dramatically, but the number of qualified inquiries stagnated. Sales increasingly received leads that had already been shaped by third-party sources. Marketing also identified a reputation issue: competitors were more frequently cited in AI responses with specific use cases, even though their market presence was no greater. Without a structured GEO strategy, the company risked becoming invisible during the early research phase.
Solution Approach
Together with Zeno Visibility, a semantically interconnected Authority System was built around the company's three most important topic clusters: industrial drive technology, predictive maintenance, and automated line integration. The goal was not merely to optimize individual pages, but to create a robust knowledge network for each keyword cluster — one that LLMs could recognize as a citable source.
The project began with a research phase. Zeno Visibility analyzed brand presence across ChatGPT, Gemini, Perplexity, Claude, and Copilot, establishing an initial Semantic Authority Score of 31/100 for the core clusters. In parallel, the team examined what types of questions users typically ask in generative systems: definition questions, comparison questions, selection criteria, integration scenarios, ROI questions, and technical troubleshooting. Based on this, 118 pieces of content per cluster were planned, distributed across hub pages, technical articles, FAQs, glossary pages, comparison pages, case studies, and social posts.
The actual implementation followed three steps. First, the team defined a semantic structure with clear entities, synonyms, and thematic relationships between products, industries, and use cases. Second, Zeno Visibility's Authority System Builder generated the content CMS-ready in multiple export formats, including internal linking logic and Schema.org JSON-LD. Third, the content was published directly into WordPress and partially into Sanity, complemented by a clean publication matrix covering editorial review, approval, and technical validation. A key principle throughout was that no new page was published in isolation — each was embedded within a network of hub, sub, and comparison pages.
In addition, a monthly LLM Brand Monitoring process was established. The team tracked which models mentioned the brand, in which contexts, whether product categories were correctly attributed, and whether competing providers were recommended more frequently. This made it possible to measure not just reach, but the quality of AI reception. For the manufacturer, this was critical: the internal objective was not "more content," but "more AI-driven recommendations."
Results
After six months, significant improvements were visible across the entire chain of visibility, authority, and demand. The Semantic Authority Score rose from 31 to 74 points. In LLM monitoring queries, correct brand mentions in product-related responses increased from 18% to 57%. Across the three core clusters, the company was more frequently cited in generative responses with specific use cases — such as retrofit, line integration, and condition monitoring — rather than being referenced merely as a generic mechanical engineering company.
The effects were also visible in traditional performance metrics. Organic visibility for 26 prioritized keywords increased by 43% compared to the previous quarter. The number of qualified contact inquiries via technical content grew by 28%, while average time on page for the newly built hub pages increased by 39%. Particularly noteworthy: 14% of new demo requests contained phrasing that could be directly traced to generative research processes — for example, "we were referred to you via ChatGPT."
Return on investment could be assessed conservatively. With a project investment in the low six-figure range and a documented incremental contribution of 17 qualified pipeline opportunities valued at approximately €680,000, a positive ROI was achieved within seven months. Equally important to the company, however, was the fact that — for the first time — a robust system for GEO and LLM Brand Monitoring existed that could be extended to additional product lines.
Lessons Learned
Summary
Working with Zeno Visibility, the manufacturer transitioned from fragmented technical content to a semantically interconnected Authority System. This drove improvements in both LLM visibility and qualified demand from the organic channel. The decisive factor was not more content in aggregate, but a structure that AI models can recognize and cite as a trustworthy source.
More Case Studies
---
*This content was created with AI assistance and reviewed by a human editor.*