# Make.com: Turn One Blog Post into 12 Assets Automatically

> A Make scenario that reads a new WordPress post and generates a LinkedIn carousel, X thread, YouTube script, email intro, and 6 social snippets — all in your voice.

**Platform:** Make  
**Category:** content-creation  
**Difficulty:** Intermediate  
**Tags:** make, content, marketing, social, openai  
**Updated:** 2026-05-20  
**Source:** https://promptfoo.tech/workflows/make-blog-to-multichannel-repurposing

## Overview
Most content teams write once and distribute badly. This Make scenario turns a single published blog post into a full multichannel package the moment it goes live, using a voice guide to keep tone consistent and a review queue so nothing ships without a human eye.

## How it works
1. Trigger: WordPress webhook fires when a post is published with tag 'repurpose'.
2. A Router splits the flow into six parallel branches — one per output format.
3. Each branch calls OpenAI with a format-specific system prompt and a shared voice guide.
4. Outputs land in a Notion 'Distribution Queue' database with status = 'draft'.
5. A Slack notification pings the content lead with a link to review.

## Benefits
- One 90-minute writing session yields a week of distribution.
- Voice consistency across every channel because the guide is shared.
- Nothing ships without human approval — AI drafts, humans decide.

## Use cases
- Solo creators repurposing long-form into shorts.
- B2B marketing teams with a weekly cornerstone post.
- Agencies producing distribution kits per client per week.

## Step-by-step
### Step 1: Author the voice guide
One markdown file: forbidden phrases, sentence-length preferences, hook style, CTA rules. This is the highest-leverage artifact in the whole scenario.

### Step 2: Set up the WordPress webhook
Use WP Webhooks or a lightweight custom hook on publish. Filter by tag so drafts don't fire.

### Step 3: Build one branch first
Start with the LinkedIn post — highest ROI per output. Get quality right, then clone the branch for each format.

### Step 4: Structure Notion as a queue
Columns: Channel, Draft, Status, Scheduled, Published URL. Status filters make review painless.

### Step 5: Add scheduling downstream
Once approved, a second Make scenario pushes to Buffer, Typefully, or Beehiiv on the schedule your team already uses.

### Step 6: Review the misses weekly
Any draft that never gets approved is signal. If LinkedIn drafts keep getting rejected, the LinkedIn prompt needs work — not the voice guide.

## Example
```
Blog post title: 'Why we killed our roadmap'. LinkedIn draft opens with: 'We deleted our public roadmap last week. Support tickets dropped 30%. Here's what we replaced it with →'
```

## FAQs
**Can this fully automate publishing?**

It can. Don't. One tone-deaf auto-post undoes a year of trust. Keep the human review step.

**How many formats before diminishing returns?**

Six is the practical ceiling. More formats means more low-quality drafts nobody ships.

**Which OpenAI model?**

GPT-4o for hooks and threads, GPT-4o-mini for snippets and email intros. Split saves ~70% on tokens.

---
Licensed under CC-BY-4.0. Attribution: PromptFoo.tech