jeffbullas.com

Use AI to Build the Business and the Life, You Actually Want. Practical insights on AI, identity, and growth for entrepreneurs who are done playing small. One email a week. No noise.

Home › Playbooks › How to Build Your Own AI-Powered Content Repurposing Factory

How to Build Your Own AI-Powered Content Repurposing Factory

Jeff Bullas

Content repurposing is the secret sauce behind every successful content operation — from solo creators to massive corporate teams. In this tutorial, you’ll learn how to set up an AI-powered system using n8n, Apify, and Claude Sonnetthat transforms a single long-form asset (like a YouTube video) into multiple, engagement-optimised posts for social media platforms.

This system is highly customisable, letting you adjust target platforms, tone, and writing style to fit your specific needs.

Step 1: Secure Your Free n8n Workflow Template

Start by getting the n8n automation template (free).

  • Locate the Template — Inside, head to the “YouTube n8n resources” tab.
  • Download the JSON File — This contains the full n8n workflow template so you can clone the system into your own n8n instance.

Step 2: Set Up the Automation’s Trigger

Every automation needs an entry point. Here, it’s a YouTube video URL.

  • Simple Form Trigger — The setup uses a form trigger inside n8n.
  • Input Field — A required text field accepts the YouTube video URL.
  • Optional Adaptation — If you use blogs or other content instead of YouTube videos, modify the trigger to scrape that data source instead.

Step 3: Scrape YouTube Data Using Apify

The key is extracting the video details and transcript.

  • Apify Scraper — Use Apify, a full-stack scraping tool with “actors” (mini scrapers).
  • Select the YouTube Scraper — Specifically, the streamers YouTube scraper.
  • Make the API Call:
    • Use the run sync get data set items method for a single response.
    • Pass the YouTube URL into the start URLs array.
    • Enable subtitle scraping to pull the full transcript.
  • Authentication Setup:
    • In n8n, set up credentials with Header Auth.
    • Create a credential (e.g., “Apify header auth”), using Authorization as the key.
    • Copy your Apify API key from console.apify.com → Settings → API and integrations.
    • Paste it into n8n and apply it to all Apify requests.

Step 4: Process and Format Scraped Data

Once Apify responds, you’ll want to clean and organise it.

  • Review Output — Includes title, URL, thumbnail, and metadata.
  • Extract Transcript — Focus on the subtitles array, which holds English subtitles in SRT format.
  • Use a Set Field Node — Structure the essentials for later use:
    • Title
    • URL
    • Full Transcript (indexed with timestamps)

Step 5: Fan Out for Multiple Platforms

With your transcript ready, branch out to create assets for multiple platforms.

  • Multiple Paths — Example: one branch for Twitter posts, one for LinkedIn posts.
  • Unlimited Options — Add more outputs, such as Instagram carousels, newsletter sections, or TikTok scripts.

Step 6: Repurpose Content per Platform

Each platform has its quirks. Here’s how to tailor content for maximum engagement.

A. Curate Platform-Specific Examples

  • Gather top-performing posts from your brand.
  • Create a “Set Field” node (e.g., Set Twitter Examples) and paste them in.
  • Use these as training examples for the AI.

B. Build a Comprehensive LLM Prompt

  • Craft a Master Prompt — Define clear roles for the LLM.
    • Twitter: “You are a world-class social media copywriter and viral growth hacker for Twitter.”
    • LinkedIn: “You are a world-class B2B content strategist and LinkedIn growth expert.”
  • Prompt Instructions:
    • Analyse the transcript to identify core insights.
    • Study examples for structure and tone.
    • Generate multiple options (e.g., 3 tweets or 3 LinkedIn posts).
    • Apply proven structures from examples.
    • Respect platform constraints:
      • LinkedIn: professional tone, minimal emojis.
      • Twitter: casual, short, free-form, emoji-friendly.

C. Execute the LLM Chain

  • Feed your prompt into the LLM chain call node.
  • Get back the requested number of options.

D. Review & Share via Slack

  • Save the generated content in its own field.
  • Split options and send them into a Slack channel.
  • Manually review and copy the best one to your scheduler.
  • (Optional) Automate further by adding an LLM to auto-select and publish via API.

Final Thoughts

By following these steps, you’ll have a robust AI-powered content repurposing factory ready to transform long-form assets into tailored, platform-specific posts at scale.

This workflow saves time, multiplies your content reach, and keeps your promotional engine running without burning you out.

Share this post: