Triage support tickets by GPT-4o sentiment and route them to Slack

Created by

Last update

Last update 2 days ago

Who is this for

Customer support teams, customer experience managers, and operations teams that need to triage incoming support tickets in real time without manual classification. Ideal for SaaS companies, e-commerce businesses, and service providers handling high ticket volumes.

What this workflow does

This workflow automatically triages incoming customer support tickets by sentiment using OpenAI GPT-4o. A ticket arrives via webhook, is normalized, analyzed for anger intensity (scored 0-10), and routed to the correct Slack channel in real time. High-anger tickets are escalated immediately, mid-range tickets go to standard support, and low-anger tickets or feedback are logged separately.

How to set up

Add your OpenAI API credential to the AI sentiment analysis node
Add your Slack OAuth2 credential to all three Slack nodes
Update Slack channel names (#escalation / #support / #feedback) to match your workspace
Activate the workflow and copy the webhook URL for your support form integration
Send a test request to verify end-to-end execution

Requirements

OpenAI API account with GPT-4o access
Slack workspace with OAuth2 app installed (chat:write scope)
Three Slack channels: one for escalation, one for support, one for feedback

How to customize

Adjust the anger score thresholds in the Switch node to match your team's escalation policy. Modify the GPT-4o prompt to detect additional sentiment categories or extract custom fields from tickets. Add language detection for multilingual support.

Node List

#	Node Name	Type	Role
1	When Ticket Received	Webhook	Trigger on incoming HTTP POST request
2	Normalize Ticket Data	Code	Sanitize and validate ticket fields
3	OpenAI Sentiment Analysis	OpenAI	AI anger scoring and sentiment tagging
4	Parse Sentiment Output	Code	Parse JSON output and decide routing
5	Route by Sentiment Score	Switch	3-way branch: ≥8 / 4-7 / ≤3
6	Send to Slack #escalation	Slack	Alert for high-anger tickets
7	Send to Slack #support	Slack	Notification for mid-range tickets
8	Send to Slack #feedback	Slack	Log for low-anger tickets or feedback
9	Respond with Routing Outcome	Respond to Webhook	Confirm result to the caller

Total: 9 nodes (+ 5 Sticky Notes)

Sticky Note Compliance

#	Sticky Note Title	Color	Role
1	Main Sticky Note (Overview)	Yellow	Workflow overview, How it works, Setup steps, Customization
2	Receive and normalize	White	Covers webhook reception and data normalization
3	Analyze sentiment	White	Covers GPT-4o sentiment analysis
4	Parse and route	White	Covers JSON parsing and routing logic
5	Post to Slack channels	White	Covers all 3 Slack post nodes and webhook response

All sticky notes use H2 headings (## ) and follow n8n public guidelines.