Workflow automation is the backbone of modern business operations. But most automations are "dumb" — they move data from point A to point B without understanding what that data actually means. Today, we're changing that for n8n users with the release of the official Sentor AI node.
Why precision matters in automation
Traditional sentiment analysis often fails in automated workflows because it provides a single score for an entire message. If a customer says "I love the product but the shipping was slow," a basic tool might just label it "Neutral."
Sentor's entity-based analysis goes further. It extracts each entity (product, shipping) and scores them independently, giving your automation the granularity it needs to route, alert, and act correctly.
Introducing the Sentor AI Node for n8n
The Sentor AI node connects directly to Sentor's Analysis API from within your n8n workflows. No custom HTTP requests, no JSON wrangling — just drag, drop, and configure.
What you can automate:
- Pull new reviews from Google, Trustpilot, App Store, or your own database
- Send each review to Sentor for entity-level sentiment analysis
- Route critical negative feedback to your Slack channel immediately
- Log all analysed results into a spreadsheet or database
- Trigger alerts when a specific topic (e.g. "WiFi", "staff") crosses a negative threshold
Setting Up Your First Workflow
1. Get Your Sentor API Key
Sign up at dashboard.sentor.app and grab your API key from Settings → API Keys.
2. Create an n8n Workflow
Start with a Schedule Trigger (runs every hour) or a Webhook Trigger (fires on new review submission).
3. Add an HTTP Request Node
Configure it to call the Sentor Analysis API:
POST https://sentor.app/api/predicts
x-api-key: YOUR_API_KEY
Content-Type: application/json
{
"docs": [{
"doc_id": "{{ $json.id }}",
"doc": "{{ $json.review_text }}",
"entities": ["product", "shipping"]
}]
}4. Parse the Response
Sentor returns structured JSON with per-entity sentiments and scores. Use n8n's Set node to extract the fields you need:
{
"results": [{
"doc_id": "review_001",
"predicted_label": "positive",
"details": [{
"sentence_text": "I love the product but shipping was slow",
"predicted_label": "positive"
}]
}]
}5. Route Based on Severity
Add an IF node: if sentiment is negative, route to a Slack notification. Otherwise,
log to your spreadsheet and move on.
Real-World Results
Teams using this setup typically see:
- Zero manual review triage — everything is categorised automatically
- Faster response times — critical issues surface in Slack within minutes of a review being posted
- Trend detection — weekly summaries show which topics are rising or falling
Advanced: Competitive Benchmarking Pipeline
Pull competitor reviews from public sources, run them through Sentor, and compare your topic scores side by side — all automated, updated daily.
"We went from manually reading 200 reviews a week to having our Slack channel tell us exactly what to fix — in real time." — Early access customer
Get Started
The Sentor n8n integration is fully documented in our n8n Integration Guide. You can also explore our full API Reference to build custom workflows in any language.