Automate LinkedIn Posts with n8n: OpenAI, Pexels & Google Sheets

Automatically generate, attach an image, and publish LinkedIn posts from your Google Sheets

This n8n workflow turns a daily Google Sheets “TO DO” topic into a complete LinkedIn post—using OpenAI to write the copy and Tavily to inform it, then Pexels to pick a matching landscape image, upload it to LinkedIn, publish, and finally log the generated content back into Google Sheets.

What this workflow does

• Runs on a daily schedule and reads the first Google Sheets row where the Status column is set to “TO DO”.
• Uses Tavily Search API to search the web for the row’s Topic and collects content from the top three results.
• Uses OpenAI (GPT-4) to synthesize those excerpts into a single LinkedIn post and outputs JSON including the post text, an image search query, and image alt text.
• Refines the visual search intent with OpenAI to produce a 2–5 keyword Pexels query, then pulls 10 landscape images from Pexels.
• Uses OpenAI (GPT-4o) to review the 10 Pexels image URLs, select the best match, and extract the chosen image URL.
• Downloads the selected image, uploads it to LinkedIn via the LinkedIn REST image upload endpoint, and publishes the post with the uploaded image.
• Writes the generated content back to Google Sheets for the matching row.

Use cases

• Maintain a simple content calendar: add a Topic to Google Sheets, set it to “TO DO,” and let n8n publish daily.
• SaaS operators and automation engineers can scale consistent LinkedIn posting without manual research and image selection.
• Teams can reuse the same workflow structure to operationalize “topic → research → post → image → publish.”

Technical details

• Google Sheets (OAuth): fetches “TO DO” rows and updates generated output.
• HTTP Request: supports Tavily and the LinkedIn REST image upload.
• OpenAI: GPT-4 for post synthesis and GPT-4o for Pexels image selection.
• Pexels: searches and evaluates landscape images; selects the best matching URL.
• LinkedIn: uploads the image and publishes the post.