Skip to main content
Set up an AI agent that handles first-level customer support — answering product questions, looking up customer data, and escalating complex issues.

Scenario

A SaaS company or service provider wants to automate their support hotline. The AI agent:
  • Identifies the customer and their issue
  • Looks up account data from the CRM
  • Answers common questions from the knowledge base
  • Creates support tickets for complex issues
  • Escalates to human support when needed

Setup steps

1. Write the prompt

2. Set up the greeting

3. Add knowledge base

Upload product documentation, FAQs, and troubleshooting guides:
  • Product FAQ — Common questions and answers
  • Troubleshooting Guide — Step-by-step fixes for known issues
  • Pricing & Plans — Current pricing, plan features, upgrade options
  • Account Management — How to reset passwords, update billing, cancel

4. Configure tools

HTTP Request: CRM Lookup
HTTP Request: Create Ticket
Extract Variable: Customer Email
Transfer Call + End Call — For escalation and call termination.

5. Configure forwarding slots

6. Set up post-call webhook

Configure a webhook to send call data to your CRM or helpdesk after each call:

Result

Most support inquiries are handled automatically — the agent answers questions, looks up customer data, and creates tickets. Complex issues are smoothly escalated to human agents with full context captured in the transcript and ticket.