ai-content-pipeline

name	ai-content-pipeline
description	AI content repurposing pipeline turning one piece into multiple formats with workflow and quality gates.

AI Content Repurposing Pipeline

The Core Principle

Create once. Distribute everywhere. The goal is not to write 10 separate pieces of content — it is to extract maximum value from one high-effort, high-quality source piece.

The pipeline flows from long-form → medium-form → short-form → micro-format.

---

The Full Repurposing Map

SOURCE: Long-form content (blog post, podcast, webinar, YouTube video)
│
├── MEDIUM-FORM
│   ├── LinkedIn article (condensed, 800–1200 words)
│   ├── Email newsletter (intro + 3 key insights + link)
│   └── YouTube video script (if source was written)
│
├── SHORT-FORM
│   ├── LinkedIn text post (hook + 3 insights + CTA)
│   ├── Twitter/X thread (6–10 tweets)
│   ├── Instagram carousel (8–10 slides)
│   ├── TikTok / Reels script (45–60 seconds)
│   └── Facebook post (1–3 paragraphs)
│
└── MICRO-FORMAT
    ├── 5 standalone tweets (one insight each)
    ├── 5 quote graphics (pull quotes from source)
    ├── 1 infographic (key stats or framework as visual)
    ├── 3 email subject line options (for A/B testing)
    └── FAQ expansion (for SEO, support docs)

---

Source Content Types and Their Yield

Source Format	Richest Derivative Formats
Long blog post (1500+ words)	Thread, carousel, newsletter, LinkedIn post, shorts
Podcast episode (30–60 min)	Blog post, newsletter, 3–5 Reels scripts, thread
Webinar or long YouTube video	Blog, carousel, 5+ short clips, email sequence
Interview or Q&A	Quote cards, Twitter thread, FAQ article
Case study	LinkedIn post, carousel, testimonial card, email story
Research/data report	Infographic, thread, carousel, press release

---

The AI Repurposing Workflow

Step 1 — Prepare the Source

Before prompting, prepare your source content:

• Transcribe audio/video (use Whisper, Otter, Riverside, or similar)
• Clean the transcript: remove filler words, fix speaker labels
• Tag the highest-value sections (insights, stats, stories, frameworks)

Step 2 — Extract the Core Message

Prompt:
"Read this content carefully: [paste content]

Extract:
1. The single most important insight (1 sentence)
2. The top 5 supporting points (1 sentence each)
3. The most quotable sentence (verbatim)
4. The primary audience and their problem being solved
5. The actionable takeaway for the reader"

Use this extraction as the foundation for all derivative formats. Every piece should point back to the same core message.

Step 3 — Generate Per Format

Use format-specific prompts (see prompts below), always including:

• The extracted core message
• The target audience
• The core takeaways
• Platform constraints (length, format)

Step 4 — Review and Calibrate

AI output will be generic without calibration. For each generated piece:

1. Add one personal detail, story beat, or specific data point not in the original
2. Adjust the opening hook — AI hooks are often weak
3. Ensure the CTA is appropriate for the platform
4. Read aloud to catch awkward phrasing

Step 5 — Publish and Distribute

Stagger publication across 1–2 weeks. Do not publish all derivatives on the same day.

Suggested stagger:
Day 1:   Publish source (blog post / podcast / video)
Day 1:   Send email newsletter featuring the source
Day 2:   LinkedIn article or text post
Day 3:   Twitter/X thread
Day 5:   Instagram carousel
Day 7:   TikTok / Reels
Day 10:  Quote cards (pick best-performing from early posts)
Day 14:  Repurpose best-performing derivative again (quote-tweet, re-share)

---

AI Prompts Per Format

Blog Post → LinkedIn Post

Here is a blog post: [paste]

Write a LinkedIn post:
- Line 1: Bold hook that works without reading the post
- Lines 2–8: 3 key insights from the post, one paragraph each
- Final 2 lines: Takeaway + question for comments
- 3–5 hashtags
- Under 1300 characters total
- Do NOT mention "In this post I wrote..." or link to the article

Blog Post → Twitter Thread

Here is a blog post: [paste]

Write a 8-tweet thread:
- Tweet 1: Hook that makes the thread worth reading (bold claim or curiosity gap)
- Tweets 2–7: One key point per tweet, under 280 chars each
- Tweet 8: Summary + "Save this thread if useful"
- Number as 1/ 2/ 3/ etc.

Podcast Transcript → Instagram Carousel

Here is a podcast transcript: [paste]

Write a 10-slide Instagram carousel:
Per slide: headline (under 8 words) + body copy (under 25 words) + visual suggestion
Slide 1: Hook — the most surprising or counterintuitive insight
Slides 2–9: One actionable point per slide, ordered logically
Slide 10: Summary + CTA to follow or save

Any Source → Email Newsletter

Here is content I want to turn into a newsletter section: [paste]

Write a newsletter section:
- Subject line: 3 options (curiosity, benefit, contrarian)
- Preview text: 1 option (40 chars)
- Opening: 2 sentences connecting to reader's situation
- Body: 3 key insights, each with a 1-sentence takeaway
- Closing: 1 sentence CTA (read full piece, try something, reply with answer)
- Tone: [conversational/professional]

---

Quality Gates

Before publishing any derivative piece, verify:

Accuracy gate:

• No facts from the source were distorted in AI generation
• Statistics and quotes are verbatim, not paraphrased incorrectly

Value gate:

• This piece delivers standalone value (reader doesn't need to see the source)
• The CTA is appropriate for someone who hasn't seen the source

Voice gate:

• At least one personal addition that AI could not have generated
• Reads like you wrote it, not like a tool generated it

Platform gate:

• Correct length for the platform
• Correct format (hashtags on LinkedIn, thread numbers on Twitter, etc.)
• Hook is rewritten (AI default hooks are predictable)

---

Repurposing Cadence

Content Volume	Recommended Cadence
1 long piece/week	Full pipeline, staggered over 7–10 days
1 long piece/month	Full pipeline, staggered over 2–3 weeks
Daily short content	Reverse: pull quotes from past long pieces

---

Tools Stack Reference

Task	Tools
Transcription	Whisper (local), Otter.ai, Riverside, Descript
AI writing	Claude, GPT-4, Gemini
Carousel design	Canva, Adobe Express, Figma
Scheduling	Buffer, Hypefury, Typefully, Later
Link tracking	UTM parameters (manual), Bitly, Dub.co
Newsletter	Beehiiv, Substack, ConvertKit