Portfolio

FB & LinkedIn Auto-posting System Using Custom GPT Action

Project Overview

This project involved developing Custom GPT actions integrated with Make.com to automate social media posting. It streamlines content curation, creation, posting processes, and management into a single seamless workflow.

Problem & Goals

The goal was to automate the manual process of finding relevant news articles, curating content, publishing posts on Facebook and LinkedIn directly through a single GPT conversation, and tracking successful posts, improving efficiency and consistency.

Solution & Implementation

Custom GPT actions were created to trigger Make.com scenarios.
Make.com scenarios parse specified websites for the latest news articles and deliver curated content directly back into the GPT conversation.
Users select and edit content within GPT before it gets published automatically on chosen social media platforms.
Integrated the ability to post on Facebook pages either using the original article image or generating a new, engaging image through OpenAI’s DALL-E 3 module within the same scenario.
Successfully posted content is recorded and saved automatically in an Airtable base for easy tracking and analytics.
A fully integrated, single-conversation workflow handles all steps from content curation to posting and reporting.

Tools & Technologies Used

GPT (Custom GPT Actions)
Make.com (Integromat)
API integrations (Facebook, LinkedIn)
Airtable
OpenAI (DALL-E 3 module)

Results & Outcomes

Streamlined content workflow significantly, reducing time spent from content sourcing to posting.
Enhanced content relevancy and engagement across social media.
Improved user experience by managing all tasks within a single GPT conversation and automation scenario.
Enabled dynamic and visually appealing content creation for Facebook posts.
Improved record-keeping and analytics through integration with Airtable.

Screenshots

Make.com scenario
Custom GPT main screen
Fetch articles
Post prep
Post creation
Image and make a post