AI Social Media Autoposter Blueprint for Make (Free)

Cross-posting the same text to LinkedIn and X is how you end up sounding wrong on both. This Make blueprint keeps a single content queue in Google Sheets but writes different copy per platform — a 150-word LinkedIn post with a hook, and a sub-270-character X post — then publishes and marks the row done.

What this scenario does

• Google Sheets — Watch Rows picks up new rows in your “Content Queue” tab: topic, key points, link, target platform, status
• Router splits by the Platform column
• LinkedIn route: OpenAI rewrites the idea as a 120-200 word LinkedIn post (hook first, line breaks, 3-5 hashtags) → LinkedIn — Create Share Update publishes it → Sheets row updated to posted-linkedin with a timestamp
• X route: OpenAI compresses the same idea to a hard 270-character limit, max one hashtag → Buffer — Create Update queues it on your X profile → row updated to posted-x

Each route has its own system prompt because the platforms genuinely need different writing: the LinkedIn prompt bans “I’m excited to share,” the X prompt bans hashtag spam. Both output bare post text with no commentary, so nothing needs cleanup before publishing.

Prerequisites

• Make account
• Google connection and a spreadsheet with a Content Queue tab (columns: Topic, Key points, Link, Platform, Status)
• OpenAI connection
• LinkedIn connection (personal profile; switch the module to organization mode for a company page)
• Buffer connection with your X profile attached (Buffer sidesteps X API pricing for small senders)

How to import

1. Download the blueprint from the box at the top of this page.
2. In Make: new scenario → More (⋯) → Import Blueprint → upload the JSON.
3. Re-link the five connection placeholders and select your real spreadsheet and sheet in all three Google Sheets modules.
4. Set the scenario schedule (every 15 minutes is plenty) and run once with a test row per platform.

What to customize

• Voice and rules in both OpenAI prompts — the defaults write for a “B2B automation consultancy”; describe your actual brand voice and paste in two example posts for much better mimicry
• Platform values — the router matches linkedin and x case-insensitively in column D; add a third route (e.g., Threads or Mastodon) by cloning a branch
• Posting target — replace Buffer with the native X module if you have API access, or point LinkedIn at a company page
• Status strings written back to column E, if your sheet uses dropdowns

The full content-calendar setup around this scenario is covered in the social media posting automation tutorial.

Cost per run

Each post is one small gpt-4o-mini call — a few hundred tokens, costing a rounding error as of mid-2026 (think well under a cent even for daily posting on both platforms). Make operations are the real budget line: roughly 4 operations per post, so 60 posts a month is ~240 operations, fine on any plan.