ETL Pipeline: Tweet Collection, Sentiment Analysis & Slack Alerts

This n8n workflow implements a complete ETL pipeline for processing tweets. It starts with a Cron node triggering daily execution to search Twitter for specific hashtags like #OnThisDay. Collected tweets are inserted into MongoDB for raw storage, then analyzed using Google Cloud Natural Language API to determine sentiment scores and magnitudes. A Set node extracts these key metrics, which are then inserted alongside the original tweet data into a Postgres database for structured querying and reporting.

Following data enrichment, an IF node filters tweets based on sentiment positivity. Positive tweets (high score) are posted to a designated Slack channel with formatted messages including the tweet text, score, and magnitude for immediate team visibility. Negative or neutral tweets are routed to a NoOp node, ensuring only actionable insights are highlighted. This setup prevents alert fatigue while capturing valuable social sentiment.

Benefits include significant time savings on manual social media monitoring, scalable data pipelines for growing tweet volumes, and integration of AI-driven insights into business workflows. Use cases span social media analytics, brand monitoring, customer sentiment tracking, and content curation—ideal for marketing teams, PR agencies, or any operation needing real-time positive buzz detection. The workflow's modularity allows easy customization for different hashtags, databases, or notification channels.