Generate Personalized Sales Emails with LinkedIn Data & Claude 3.7 via OpenRouter

This n8n workflow automates the process of generating highly personalized sales emails using prospect data from Google Sheets and leveraging the Claude 3.7 large language model via OpenRouter. It's designed to help sales teams create tailored outreach messages efficiently.

What it does

This workflow performs the following steps:

1. Manual Trigger: The workflow is initiated manually, allowing for on-demand execution.
2. Retrieve Prospect Data: It reads prospect information (e.g., Name, Company, LinkedIn URL) from a specified Google Sheet.
3. Loop Over Items: Each row (prospect) from the Google Sheet is processed individually.
4. Extract LinkedIn Profile: For each prospect, it makes an HTTP request to a LinkedIn profile scraping service (e.g., ProxyCrawl, Bright Data, or a custom scraper) to retrieve detailed public information.
5. Filter for Valid Data: It checks if the LinkedIn scraping was successful and returned valid data. If not, the item is filtered out.
6. Prepare Data for LLM: It processes the scraped LinkedIn data and prospect information, preparing it as a structured input for the language model.
7. Generate Personalized Email Draft: It uses a "Basic LLM Chain" node with the "OpenRouter Chat Model" (configured for Claude 3.7) to generate a personalized sales email draft based on the prospect's data and LinkedIn profile.
8. Parse LLM Output: A "Structured Output Parser" node extracts the generated email subject and body from the LLM's response.
9. Aggregate Results: The generated email drafts (subject and body) are aggregated back with the original prospect data.
10. Send Email (Optional): The workflow can be extended to automatically send the personalized email drafts using Gmail.

Prerequisites/Requirements

To use this workflow, you will need:

• n8n Instance: A running n8n instance.
• Google Sheets Account: To store your prospect data.
• LinkedIn Scraping Service: An API key or access to a service that can scrape LinkedIn profiles (e.g., ProxyCrawl, Bright Data, or a self-hosted solution). The HTTP Request node needs to be configured with the appropriate endpoint and authentication.
• OpenRouter Account & API Key: To access the Claude 3.7 language model.
• Gmail Account (Optional): If you intend to send emails directly from the workflow.

Setup/Usage

1. Import the Workflow: Download the provided JSON and import it into your n8n instance.
2. Configure Credentials:
   • Google Sheets: Set up a Google Sheets credential to allow n8n to read your prospect data.
   • HTTP Request (LinkedIn Scraper): Configure the HTTP Request node with the URL and authentication details for your chosen LinkedIn scraping service.
   • OpenRouter Chat Model: Set up an OpenRouter credential with your API key.
   • Gmail (Optional): If using, configure a Gmail credential.
3. Update Google Sheets Node:
   • In the "Google Sheets" node, specify the Spreadsheet ID and Sheet Name where your prospect data is located. Ensure your sheet has columns for at least "Name", "Company", and "LinkedIn URL".
4. Review LLM Prompt:
   • Inspect the "Basic LLM Chain" node. The prompt is crucial for generating effective emails. Adjust it to fit your sales strategy and desired email tone.
5. Test the Workflow: Run the workflow manually using the "When clicking ‘Execute workflow’" trigger to ensure all steps are configured correctly and data flows as expected.
6. Review and Send Emails: The workflow will output the generated email subjects and bodies. You can then manually review these emails before sending them, or connect the "Gmail" node to automate sending.