AI-Powered Content Research, Creation, and Distribution Pipeline for Omnius
How n8n, AI agents, and a human-in-the-loop strategy automated an entire content marketing agency's core service delivery.
40-60%
Manual Work Reduced
5+
Platforms Integrated
Built-In
Human Oversight
Background
Omnius is a B2B SEO & AI Search marketing agency based in Europe, specializing in SaaS and Fintech companies. They've built a reputation for driving serious organic growth, taking clients from zero to millions of organic visitors through strategic content and search optimization.
As a content-driven agency, Omnius faces a familiar paradox: the more clients they win, the more content they need to produce. And content is not just blog posts anymore. It's LinkedIn carousels, Twitter threads, YouTube summaries, WordPress articles, and AI-optimized snippets designed to show up in ChatGPT and Perplexity answers.
Omnius needed a system that could handle the research, ideation, drafting, and distribution, while keeping humans in control of quality and brand voice.
The Challenge
The Topic Discovery Problem
For an agency serving multiple B2B clients, content research is not a one-time task. It is a continuous operation that has to be good.
- Multi-Client Research Load: Each client operates in a different niche. Research for one does not apply to another.
- Real-Time Relevance: The best topics have a short shelf life. Trending conversations need immediate action.
- Format Intelligence: It is not just what to write, but how. A topic might be a blog post for one platform, a carousel for another.
- The Expertise Asymmetry: Strategists are SEO experts, not industry experts. They need to get up to speed on niche topics quickly.
The Content Scaling Problem
- Platform-Specific Adaptation: Manually reformatting content for LinkedIn, WordPress, and X is a bottleneck.
- The Approval Bottleneck: Email chains and document hunting for approvals slow down publishing.
- Visual Asset Creation: Manual design requests create a queue and add delays.
- Distribution Fragmentation: Logging into multiple platforms to publish is inefficient.
The Solution
We built an Agentic Content Army. A multi-stage n8n workflow that automates the entire content pipeline from topic discovery to published posts, with human review at the critical quality checkpoint.
The workflow scrapes YouTube and X, analyzes content using AI to identify trending topics, high-performing formats, content gaps, and unique angles. This replaces hours of manual research with deep, automated competitive intelligence.
Informed by the research, a "Post Writer" agent drafts content tailored for each platform (LinkedIn, WordPress, etc.), maintaining brand voice and strategic positioning.
Drafts are compiled into a Google Doc for easy review. Upon approval, the workflow automatically generates visuals and publishes the content across all designated channels.
Impact
The Content Army did not just save time; it multiplied the quality and strategic value of the research itself, transforming how Omnius delivers value to its clients.
| Activity | Before | After |
|---|---|---|
| Topic Research | 2-3 hours/week manual scanning | Automated, comprehensive analysis |
| Research Quality | Limited to manual review | Patterns from 10x more sources |
| Content Drafting | 4-6 hours per batch | AI-generated drafts in minutes |
| Visual Creation | Manual design requests & revisions | Auto-generated with content |
| Review Process | Messy email chains, scattered feedback | Single doc with context, one-click approval |
| Publishing | Manual login to each platform | Automated multi-channel distribution |
| Total Weekly Hours | 10-15 hours | 4-6 hours (review) |
"What impressed me the most was their very practical problem-solving approach. They did not just build workflows for the sake of it. They actually dug deep into our suboptimal processes, understood where things were breaking or wasting time, and then fixed them in a very structured and thoughtful way."
- Matija Golubović, Founder at Omnius
Why Human-in-the-Loop Matters
- Quality Assurance: AI drafts, but humans ensure brand voice, accuracy, and strategic alignment.
- Client Trust: The review gate prevents publishing incorrect or off-brand content.
- Research Validation: Humans validate whether AI-identified trends are relevant to specific clients.
- Continuous Improvement: Each review generates feedback that improves the AI agents over time.
Next Steps
- Analytics Feedback Loop: Automatically feed content performance back into the research phase to create a self-improving system.
- Multi-Client Scaling: Template the workflow for rapid deployment across Omnius's full client roster.
- Predictive Topic Intelligence: Use historical data to predict which topics will trend before they peak.
Technology Stack
| Component | Technology |
|---|---|
| Workflow Engine | n8n (self-hosted) |
| AI Models | OpenAI GPT-4, DALL-E |
| Data Collection | YouTube API, Apify |
| Memory Layer | Simple Memory |
| Publishing | WordPress API, LinkedIn API |
| Review | Human-in-the-Loop with Google Docs |
| File Processing | Convert to File |
n8n LabLet Us Help You Win Tomorrow’s Automation Landscape
Start eliminating manual bottlenecks with custom n8n automation. Let us architect the system for you.
Our Team
Ready to help