Creating high-quality, location-specific pages is a big challenge for content and SEO teams. But Niklas Buschner and the team at Radyant has done just done that using several AirOps workflows. They were able to create several pages for cities and states in Germany, that proved invaluable to users. Niklas joined the AirOps team to share their workflows and thought processes behind how they did it.

Top 5 Takeaways

1. AI workflows can be used for high-quality, location-specific content. Niklas's team published 247 unique, high-ranking pages using AirOps automations.‍
2. Combining Perplexity with other LLMs keeps it accurate and authoritative. Layering Perplexity for research with Claude for writing delivered reliable, customized content. Watch below for why they decided on which LLMs to use. ‍
3. Human review remains critical for quality, even with advanced automation. Manual review steps caught errors and improved content standards, especially for images.‍
4. Work in batches and test. There is no need to publish hundreds of articles at once. Test in small batches, see the outputs and then proceed from there. ‍
5. Custom content agents and feedback loops significantly accelerated their operations. Niklas's content agent workflow slashed production time while embedding internal linking and editorial review.

Best practices and key learnings

Content engineering opens new doors, but success depends on combining automation with smart strategy and hands-on oversight. Here's what stood out:

Start small, test, and scale with confidence

Launching in small batches lets you catch issues early and avoid wasted effort.

• Begin with 5–10 test pages to validate workflow quality before scaling up.
• Use early results to refine prompts, logic, and model selection.
• Monitor both rankings and real traffic to measure impact.

"We always start with, like, 5 to 10. So we run the whole thing for 5 to 10 pages at first, and we build it out step by step..." — Niklas Buschner

Leverage the right AI models for each task

Not all LLMs are created equal. Use their strengths where they matter most.

• Use Perplexity for up-to-date, authoritative research.
• Deploy Claude for nuanced content creation.
• Assign simpler admin or formatting tasks to GPT models.

Build editorial review and feedback loops into every workflow

Every automation still needs human oversight and checks:

• Require manual review steps before publishing live.
• Implement feedback loops to catch and correct errors or mismatches.
• Prioritize quality at every stage, especially for visuals and structured data.

How to put this into practice

Instead of relying on generic automation, Niklas's approach shows that combining the right models, thoughtful workflow design, and human-in-the-loop review delivers results that stand out in search and drive real business outcomes.

To apply these strategies, content teams should pilot small-scale workflows, choose AI models based on task strengths, and never skip manual review—especially for high-impact or edge-case content. This balance of automation and oversight helps brands scale faster without sacrificing accuracy or authority.

Future-proofing your content operations with AirOps

Scaling content with AI isn't just about speed—it's about building systems that adapt, improve, and deliver results at scale. By combining automation, smart model selection, and built-in quality checks, teams can move faster and smarter.

Want to see how AirOps can help your team build, launch, and optimize content workflows like these? Book a call to get started.

‍