Personalized Hotel Reward Emails for High-Spenders with Salesforce, Gemini AI & Brevo

This n8n workflow automates the process of identifying high-spending hotel guests from Salesforce, generating personalized reward emails using Google Gemini AI, and sending them via Brevo (formerly Sendinblue). It helps hotels nurture their most valuable customers with tailored offers.

What it does

1. Triggers on Salesforce Data: Listens for new or updated "Contact" records in Salesforce.
2. Filters High-Spenders: Checks if the guest's "Total Spend" in Salesforce exceeds a predefined threshold (e.g., $5000).
3. Generates Personalized Email Content (Gemini AI):
   • If the guest is a high-spender, it uses a Google Vertex Chat Model (powered by Gemini AI) to generate a personalized reward email.
   • The AI is instructed to create a compelling email based on the guest's name, total spend, and a suggested reward (e.g., a free night, upgrade, or spa credit).
   • It then structures the AI's output into a JSON format using a Structured Output Parser for easy extraction of the email subject and body.
4. Sends Email via Brevo: Dispatches the personalized reward email to the high-spending guest using Brevo.
5. Updates Salesforce (Optional - not explicitly in JSON but implied by common use cases): While not explicitly shown in the provided JSON, a common next step would be to update the Salesforce contact record to log that an email was sent or to track the reward offered.

Prerequisites/Requirements

• n8n Account: A running n8n instance (cloud or self-hosted).
• Salesforce Account: With appropriate permissions to read Contact records.
• Google Cloud Project: With the Vertex AI API enabled and a service account for the Google Vertex Chat Model. This will likely involve setting up a Google Cloud credential in n8n.
• Brevo (Sendinblue) Account: With an API key for sending emails.
• Basic LLM Chain Node: Requires the @n8n/n8n-nodes-langchain package to be installed in your n8n instance.
• Structured Output Parser Node: Also from the @n8n/n8n-nodes-langchain package.

Setup/Usage

1. Import the Workflow:
   • Download the provided JSON file.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the three dots menu (...) in the top right and select "Import from JSON".
   • Paste the workflow JSON or upload the file.
2. Configure Credentials:
   • Salesforce Trigger:
     • Click on the "Salesforce Trigger" node.
     • Select or create a new Salesforce OAuth2 credential. Follow the n8n documentation for setting up Salesforce credentials.
     • Configure the "Object" to Contact and set the "Operation" to Watch for New or Updated.
   • Google Vertex Chat Model:
     • Click on the "Google Vertex Chat Model" node.
     • Select or create a new Google Cloud credential (e.g., Service Account Key). Ensure the service account has permissions to use Vertex AI.
     • Configure the Model Name (e.g., gemini-pro).
   • Brevo:
     • Click on the "Brevo" node.
     • Select or create a new Brevo API credential.
3. Configure Nodes:
   • If Node (ID: 20):
     • Adjust the condition to match your high-spender logic. The current setup likely checks a numeric field for a value greater than a specific amount (e.g., {{ $json.totalSpend > 5000 }}).
   • Code Node (ID: 834):
     • This node likely prepares the prompt for the AI. Review and adjust the prompt to fine-tune the email generation to your specific needs and brand voice.
   • Basic LLM Chain (ID: 1123):
     • Ensure the prompt template and variables are correctly mapped from the previous "Code" node.
   • Structured Output Parser (ID: 1179):
     • This node is crucial for extracting the subject and body from the AI's response. Verify its schema definition matches the expected JSON output from the AI.
   • Brevo Node (ID: 820):
     • Configure the "Operation" to Send Email.
     • Map the recipient email address (e.g., {{ $json.email }}) and the subject and body from the "Structured Output Parser" node.
     • Set your sender email address.
4. Activate the Workflow: Once all credentials and configurations are set, save and activate the workflow.

Now, whenever a Salesforce Contact record is created or updated and meets your high-spender criteria, a personalized reward email will be generated and sent automatically!