Create Personalized B2B Outreach Emails with Tavily Research & OpenRouter LLM

This n8n workflow automates the creation of highly personalized B2B outreach emails by leveraging Tavily Research for company-specific insights and an OpenRouter Large Language Model (LLM) for crafting the email content. It streamlines the process of generating tailored emails from a list of prospects in a Google Sheet.

What it does

1. Triggers Manually: The workflow is initiated manually when you click 'Execute workflow'.
2. Reads Prospect Data: It reads prospect information (e.g., company name, contact person) from a specified Google Sheet.
3. Limits Processing (Optional): A "Limit" node is included, which can be configured to process only a subset of the data for testing or partial runs.
4. Loops Through Prospects: For each prospect, the workflow performs the following steps:
   • Researches Company: An AI Agent (LangChain) uses Tavily Search to gather relevant, up-to-date information about the target company.
   • Generates Email Draft: A Basic LLM Chain, powered by an OpenRouter Chat Model, uses the gathered research and a predefined prompt to generate a personalized B2B outreach email.
   • Parses Output: A Structured Output Parser extracts the generated email content in a structured format (e.g., JSON).
5. Writes Emails Back to Google Sheets: The generated personalized emails are written back to the original Google Sheet, or a new one, alongside the prospect data.

Prerequisites/Requirements

• n8n Instance: A running n8n instance.
• Google Sheets Account: With a spreadsheet containing your prospect data (e.g., company names, contact names).
• Tavily API Key: For the AI Agent to perform web research.
• OpenRouter API Key: To access the Large Language Model for email generation.

Setup/Usage

1. Import the Workflow:
   • Copy the provided JSON code.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the three dots next to "New workflow" and select "Import from JSON".
   • Paste the JSON and click "Import".
2. Configure Credentials:
   • Google Sheets: Set up your Google Sheets credential to allow n8n to read from and write to your spreadsheet.
   • Tavily: Configure a new "Tavily Search API" credential with your Tavily API Key.
   • OpenRouter: Configure a new "OpenRouter API" credential with your OpenRouter API Key.
3. Update Google Sheets Node (Node 18):
   • Select the Google Sheets node.
   • Choose your Google Sheets credential.
   • Specify the Spreadsheet ID and Sheet Name where your prospect data is located.
   • Ensure the "Operation" is set to "Read" initially.
   • Later, you'll configure a second Google Sheets node (or modify this one) to "Write" the generated emails back.
4. Configure AI Agent (Node 1119):
   • Select the "AI Agent" node.
   • Ensure the "Tool" is set to "Tavily Search" and select your Tavily credential.
   • Adjust the "Agent Prompt" to guide the research effectively.
5. Configure OpenRouter Chat Model (Node 1281):
   • Select the "OpenRouter Chat Model" node.
   • Choose your OpenRouter credential.
   • Select the desired LLM model (e.g., mistralai/mistral-7b-instruct:free).
   • Adjust "Temperature" and other parameters as needed.
6. Configure Basic LLM Chain (Node 1123):
   • Select the "Basic LLM Chain" node.
   • Craft the "Prompt Template" to instruct the LLM on how to generate the personalized email, incorporating data from the Google Sheet and the research results.
     • Example variables to use: {{ $json.companyName }}, {{ $json.contactName }}, {{ $json.researchResults }}.
7. Configure Structured Output Parser (Node 1179):
   • Define the "Output Schema" to specify the expected JSON structure for the generated email (e.g., { "subject": "string", "body": "string" }).
8. Add a Google Sheets Write Node (Not explicitly in JSON, but implied for saving results):
   • After the "Structured Output Parser" node, add another "Google Sheets" node.
   • Configure it to "Append" or "Update" rows in your spreadsheet, mapping the parsed email subject and body to appropriate columns.
9. Test and Activate:
   • Run the workflow in "Test" mode to ensure it's functioning as expected and generating correct output.
   • Once satisfied, activate the workflow.
10. Execute: Click "Execute workflow" to run the process and generate personalized emails for your prospects.

This workflow provides a powerful foundation for automating your B2B outreach, allowing you to scale personalization efforts efficiently.