AI Social Media Content Scheduler With n8n And Open AI

Most marketing teams spend Sunday evenings or Monday mornings manually rewriting the same core idea three different ways for three different platforms. A LinkedIn post needs professional framing and a question hook to drive comments. A Twitter post needs to be tight and punchy under 280 characters. Instagram needs warmth, emojis, and hashtag strategy. That’s 20–30 minutes per topic, every week, often for 10–15 topics — easily 4–6 hours that disappears before any real strategy work begins.

This workflow connects your existing Notion content calendar to GPT-4o and Buffer. Once a piece of content is marked “Ready to Post” in Notion, the automation handles the rest: generating three platform-native versions, scheduling them through Buffer with your preferred timing, and logging everything to Google Sheets so you can track what was published and when. The $0.02-per-post cost comes almost entirely from GPT-4o API usage — each three-post generation uses roughly 600-800 output tokens.

The real value isn’t just speed — it’s consistency. The system prompt enforces your brand voice, target audience framing, and CTA structure every single time. You stop getting LinkedIn posts that read like tweets, and Instagram captions that forget their hashtags.

Prerequisites

1. An active n8n instance (cloud or self-hosted; n8n Cloud free tier works fine for this workflow)
2. An OpenAI API key with access to GPT-4o (usage-based billing; estimated $0.40–$0.80/month for 20 posts/week)
3. A Buffer account connected to your LinkedIn, Twitter/X, and Instagram profiles
4. A Notion database with at minimum these properties: Topic (text), Target Audience (text), Key Message (text), Scheduled Date (date), Status (select with a “Ready to Post” option)
5. A Google Sheet with headers: notion_page_id, linkedin_text, twitter_text, instagram_text, scheduled_date, buffer_post_ids, logged_at

Setup Guide

1. Import the n8n blueprint — Download the JSON blueprint and import it via n8n’s workflow import screen (top-right menu → Import from File).
2. Connect Notion — In the Notion node, authenticate with your Notion integration token and select your content calendar database. Set the filter to Status = Ready to Post.
3. Add your OpenAI credential — In the OpenAI node, add your API key. Confirm the model is set to gpt-4o. Leave temperature at 0.7 for creative variation.
4. Customize the system prompt — Replace [TARGET_AUDIENCE] defaults with your brand’s typical audience if it rarely changes. Add any brand voice rules (e.g., “never use exclamation marks on LinkedIn”).
5. Connect Buffer — Generate a Buffer access token from your Buffer developer settings. Paste it into the three HTTP Request nodes. Update the profile_id fields with your actual Buffer channel IDs (find these via GET /profiles in the Buffer API).
6. Link Google Sheets — Authenticate the Sheets node with your Google account and point it to your logging spreadsheet and sheet tab name.
7. Set the schedule trigger — Default is Mon–Fri at 7:00 AM in your local timezone. Adjust in the Schedule Trigger node settings.
8. Test with one Notion entry — Create a test row in Notion with Status = “Ready to Post”, run the workflow manually using the “Execute Workflow” button, and verify all three platforms receive scheduled posts in Buffer.

Who This Is NOT For

• Teams posting more than 30 pieces of content per week (you’ll want a more robust queue management system with approval layers built in)
• Brands in highly regulated industries (pharma, finance, legal) where every post requires human legal review before scheduling — this workflow skips that gate
• Anyone who needs image generation as part of post creation (this workflow handles text only; pair it with a DALL-E or Midjourney workflow if visuals are required)
• Teams already using an all-in-one social media platform like Sprout Social or Hootsuite that has built-in AI features — the added complexity may not justify the cost savings