screenshot 2026 03 06 at 2.43.20 pm

Building an AI Content Machine with n8n

The Content Consistency Problem Every Marketing Team Faces

Content marketing works — but only if you’re consistent. The AI Posts Content Machine I built with n8n changes the equation. It’s a 31-node automation workflow that researches trending topics, generates content ideas, scores and rewrites them using AI, and publishes across Instagram and X (Twitter) — fully automatically, on a schedule you set.


How the Workflow Works: End-to-End

Stage 1: Idea Generation and Research

The workflow starts with a Content Bank (Google Sheet or Airtable) where you seed topics and keywords. Every run, it pulls ideas and uses Tavily Search (7 parallel searches) to research what’s trending around each topic.

Stage 2: Post Drafting

The Pick One node selects the strongest content angle. An OpenRouter Chat Model node generates the initial post draft — optimized for the target platform’s format, character count, and engagement patterns.

Stage 3: AI Scoring and Rewriting

The Score and Rewrite node sends the draft to Claude with a detailed scoring rubric: hook strength, value density, platform-fit, and CTA clarity — all scored 0–10. If any score falls below 7, Claude rewrites that element automatically.

Stage 4: Scheduling and Status Management

Approved posts are stored in the Content Bank with status “Ready.” The Get Ready Posts node polls for ready posts and queues them for publishing. The Update Status node marks posts as “Published” after distribution to prevent duplicates.

Stage 5: Multi-Platform Publishing

Publishes to X (Twitter) via the X API, and to Instagram via a three-step process: Create Media Container → Wait for Processing → Publish Post.


What You Need to Set This Up

  • n8n (cloud ~$20/month or self-hosted)
  • OpenRouter API key for multi-model LLM access
  • Anthropic API key for Claude scoring
  • Tavily API key for real-time web search (free tier available)
  • X Developer Account with API access
  • Meta Business Account with Instagram Graph API access
  • Google Sheets or Airtable for the Content Bank

Estimated monthly cost: ~$35–65/month to run a fully automated content operation. Compare that to a junior content creator at ₹25,000–40,000/month.


The Human Layer

Spend 30 minutes per week seeding new topics. Review AI-drafted posts before they publish. Add brand-specific examples and proprietary data to your system prompt. The goal is a human-in-the-loop system — you retain creative direction, AI handles the production grind.


Download the Workflow

The complete AI Posts Content Machine workflow is available for download on my Resources page. Setup time: approximately 3–4 hours for a technical marketer. Payback time: day one.

Prajesh Meshram is a Senior Marketing Leader and marketing automation practitioner. Download his complete n8n workflow collection at the Resources page.

Tags: No tags

Add a Comment

Your email address will not be published. Required fields are marked *