I Built a Voice Agent That Calls Every New Lead (n8n + Vapi)

I Built a Voice Agent That Calls Every New Lead (n8n + Vapi)

🎙 Nate Herk 👥 964K 📅 January 12, 2026 ⏱ 26 min 👁 56K 📄 tutorial 🧭 2026-08-28
Available in: English (current) Français

Keywords

voice agentn8nVapilead qualificationautomation

Summary

The video presents a comprehensive tutorial on building an outbound voice AI agent using n8n for workflow automation and Vapi for voice interaction. The system is designed to automatically call new leads after form submission, qualify them, and collect structured data. The creator demonstrates a live call, showing how the agent gathers information such as motivation, urgency, budget, and past experience. The workflow includes data normalization, API calls to Vapi, polling for call completion, and logging results to Google Sheets. The tutorial covers configuration of the Vapi assistant, including system prompt design, structured outputs, and variable substitution. The creator emphasizes best practices like introducing the agent as AI and handling voicemail. The video also mentions limitations of Vapi’s built-in phone numbers and suggests using Twilio for scaling. The tutorial is practical and aimed at users with some technical background, providing templates and resources for replication.

145 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides high practical value by offering a complete, replicable system with a live demonstration. The argumentation is solid, as the creator explains each step logically, from wireframe to implementation, and justifies design choices (e.g., data normalization, polling). The use of a real example with a mock lead adds credibility. However, the argumentation is largely based on personal experience and lacks comparative analysis or independent validation of the approach’s effectiveness.

Scientific Rigor, Source Quality, Title Accuracy

The video references official Vapi API documentation and uses standard practices like bearer token authentication. The creator mentions using Claude for code generation but does not provide direct sources for that. The title accurately reflects the content. The description includes links to the creator’s courses and tools, which are promotional but not misleading. The video does not cite external scientific sources, but as a tutorial, it relies on practical demonstration and documentation.

158 words

Title / Content Match

The title accurately reflects the content: the video demonstrates building a voice agent that calls new leads using n8n and Vapi.

Quality & Reliability

7/10

The video provides a detailed, step-by-step tutorial with a live demo, clear explanations of the workflow, and references to official documentation. However, it lacks independent verification of claims and relies heavily on the creator's personal experience, with some promotional content.

Chapters

Cited Sources

  • Vapi API Documentation — Referenced for creating and getting calls, and configuring API requests.
  • n8n — Used as the automation platform for the workflow.
  • Vapi — Voice agent platform used for the AI agent.
  • Twilio — Mentioned as a provider for phone numbers to scale outbound calls.
  • Claude — Used to generate the code for the n8n code node.

Concurring Sources

  • Vapi API Reference — The video's API calls align with the official Vapi documentation.
  • n8n Documentation — The workflow uses n8n nodes as described in the official documentation.

External References

Contribution & Novelties

The video offers a practical, end-to-end blueprint for building an outbound voice AI agent, combining n8n and Vapi. It demonstrates a real-world use case (lead qualification) and provides reusable templates and code. The approach of using structured outputs and dynamic variables is a notable technique for extracting actionable data.

Pour aller plus loin :

96 words

Radar Profile

The radar profile shows high scores in information quantity and technical level, reflecting the detailed tutorial nature. Quality and reliability are slightly lower due to reliance on personal experience and lack of external validation. The overall profile indicates a practical, hands-on resource with strong technical depth.

Reliability 7/10