Build a Multi-Channel Sales Agent (No-Code)

Key Concepts

• No-code AI sales agent
• Lead research and qualification
• Personalized communication (voice call, WhatsApp, email, LinkedIn)
• CRM integration
• Relevance AI (no-code AI agent builder)
• Make.com (automation platform)
• Agent prompting and SOP (Standard Operating Procedure)
• Tool usage within AI agents
• Voice agent with personalized script
• Lead qualification and follow-up automation

System Overview

The video demonstrates how to build a no-code AI sales agent using Relevance AI and Make.com. This agent automates the process of researching leads, making personalized calls, qualifying them, updating the CRM, and sending follow-up messages through various channels. The system is designed to improve efficiency in both inbound and outbound sales funnels.

Step-by-Step Process

1. Trigger: A lead fills out a form on the website (Typeform).
2. Data Transfer: Make.com receives the form data and triggers the Relevance AI agent.
3. Lead Research: The agent uses tools to find the lead's and company's LinkedIn profiles and scrapes their data.
4. Personalized Call Script Generation: Based on the research, the agent generates a personalized call script.
5. Voice Call: The agent uses a voice calling tool to call the lead, deliver the personalized script, gather more information, and qualify the lead based on their budget.
6. Decision Making: Based on the call outcome (qualified or unqualified), the agent decides on the next steps.
7. Follow-up Communication:
   • Qualified Leads: The agent sends personalized follow-up messages via WhatsApp, email, and LinkedIn, including a booking link.
   • Unqualified Leads: The agent may refer the lead to another agency or partner.
8. CRM Update: The agent updates the CRM (e.g., HubSpot) with all the information gathered during the process.

Detailed Breakdown of the System

1. Lead Form and Trigger (Typeform & Make.com)

• The process starts with a lead filling out a form on the website.
• Make.com is used to trigger the Relevance AI agent because it has native integration with Typeform.
• The Make.com blueprint will be available in the free community.

2. Relevance AI Agent Setup

• Relevance AI is used as the primary platform for building the AI agent.
• The agent has a prompt and seven tools.
• The template of the agent solution will be available for free in the community.

3. Agent Prompt

• The agent prompt defines the agent's role, objective, and SOP.
• Role: World-class inbound lead agent.
• Objective: Efficiently manage new leads, research them, qualify them, and engage them through multiple channels.
• SOP: A step-by-step guide for the agent to follow, including finding LinkedIn profiles, researching leads and companies, making voice calls, sending follow-up messages, and updating the CRM.
• The prompt includes extra instructions and rules, such as only using the "send LinkedIn invite" tool if a LinkedIn profile was found.
• The prompt also includes examples of input and output scenarios to guide the agent's actions.

4. Flow Builder

• The flow builder is used to visually represent the SOP and ensure the agent follows the correct sequence of actions.
• It includes conditions, such as using the "find LinkedIn" and "research lead" tools first.

5. Tools

The agent uses seven tools:

• Find LinkedIn and Research Lead:
  • Uses a Google search module to find the lead's LinkedIn profile.
  • Scrapes the LinkedIn profile to gather information.
  • Uses an AI step to summarize the lead's information.
  • Outputs the lead summary, company LinkedIn URL, and lead LinkedIn URL.
• Research Company:
  • Scrapes the company's website and LinkedIn profile.
  • Uses an AI step to create a company summary.
  • Handles the scenario where the company LinkedIn URL is not found.
• Voice Call:
  • Generates a personalized call script based on the lead and company research.
  • Uses a built-in voice calling tool to call the lead.
  • The call script includes a personalized line and a pre-defined structure.
  • The agent is instructed on tone of voice and how to handle common questions.
  • The tool uses the best AI models for voice agents.
  • The call details (transcript, recording, status) are extracted and sent back to the agent.
• Send Email:
  • Generates a personalized email based on the call outcome and other details.
  • Uses a built-in email module to send the email.
• Send WhatsApp:
  • Generates a personalized WhatsApp message based on the call outcome and other details.
  • Uses a built-in WhatsApp module to send the message through a personal WhatsApp number.
  • Requires scanning a QR code to connect the WhatsApp account.
• Send LinkedIn Invite:
  • Sends a LinkedIn invitation with a personalized message.
  • Requires the lead's LinkedIn username (the last part of the LinkedIn URL).
  • The LinkedIn Action tool is used to send the invitation.
  • The video warns against overusing this tool to avoid being blocked by LinkedIn.
• Update CRM (HubSpot):
  • Updates the CRM with all the information gathered during the process.
  • Uses the HubSpot API call module to create a new contact with the relevant properties.

6. Voice Call Tool Details

• The voice call tool uses a system prompt to guide the AI voice assistant.
• The prompt includes an objective, specifics, tone of voice instructions, and a script.
• The script includes mandatory lines that must be recited verbatim.
• The prompt also instructs the agent on how to handle common questions, such as if the user asks if they are an AI.
• The video notes that the Relevance AI voice agent does not have a knowledge base or tool calling capabilities during the call.

7. LinkedIn Invite Tool Details

• The LinkedIn invite tool requires the lead's LinkedIn username.
• The video demonstrates how to find the LinkedIn username from the LinkedIn URL.
• The tool uses the LinkedIn Action module to send the invitation with a personalized message.
• The video warns against overusing this tool to avoid being blocked by LinkedIn.

Key Arguments and Perspectives

• Efficiency: The AI sales agent automates many tasks, saving time and improving efficiency.
• Personalization: The agent personalizes communication based on lead research and call outcomes.
• Qualification: The agent qualifies leads based on their budget and project needs.
• Integration: The system integrates with various tools, including Typeform, Relevance AI, Make.com, WhatsApp, and HubSpot.
• No-Code: The system can be built without coding knowledge.

Notable Quotes

• "This agent in my case it's getting triggered from my lead form on my website."
• "I've set up almost the entire system in relevance AI."
• "Your role is critical to B AI or AI automation agency by effectively researching leads understanding their automation needs qualifying them and initiating communication through various channels you help Drive our business growth."
• "Every line that begins with this symbol is mandatory and must be recited vertin meaning it has to exactly say that line like it is."

Technical Terms and Concepts

• No-code: A development approach that allows users to build applications without writing code.
• AI agent: A software program that can perform tasks autonomously.
• Lead: A potential customer.
• CRM: Customer Relationship Management, a system for managing interactions with customers.
• API: Application Programming Interface, a set of rules that allows different software systems to communicate with each other.
• Web scraping: The process of extracting data from websites.
• Agent Prompting: The process of designing and writing instructions for AI agents.
• SOP (Standard Operating Procedure): A set of step-by-step instructions for performing a task.
• Variables: Placeholders for data that can change.
• Modules: Reusable components that perform specific tasks.

Logical Connections

• The lead form triggers the entire process.
• The research tools provide data for the personalized call script.
• The call outcome determines the follow-up communication strategy.
• The CRM is updated with all the information gathered throughout the process.

Synthesis/Conclusion

The video provides a detailed guide on building a no-code AI sales agent that automates the entire lead management process, from initial contact to CRM update. By leveraging Relevance AI and Make.com, users can create a system that researches leads, personalizes communication, qualifies prospects, and streamlines sales operations. The key takeaways are the importance of a well-defined agent prompt, the strategic use of tools, and the integration of various communication channels for effective lead engagement. The system offers a practical solution for businesses looking to improve their sales efficiency and drive growth through AI automation.