Client review collection & sentiment analysis with HighLevel, GPT-4o, Gmail & Slack

📘 Description: This automation streamlines client review collection and sentiment summarization for Techdome using HighLevel CRM, Azure OpenAI GPT-4o, Gmail, Slack, and Google Sheets. It starts by pulling recently won deals from HighLevel, then generates and sends AI-written HTML review request emails with built-in Google Review and feedback form links. After waiting 24 hours, it fetches the client’s reply thread, summarizes the sentiment using GPT-4o, and posts a clean update to Slack for team visibility. Any failures—API errors, empty responses, or data validation issues—are logged automatically to Google Sheets for full transparency and QA. The result: a fully hands-free Client Appreciation + Feedback Intelligence Loop, improving brand perception and internal responsiveness.

⚙️ What This Workflow Does (Step-by-Step)

▶️ When Clicking ‘Execute Workflow’ (Manual Trigger) Allows on-demand execution or scheduled testing of the workflow. Initiates the fetch for all newly “Won” deals from HighLevel CRM.

🏆 Fetch All Won Deals from HighLevel Retrieves all opportunities labeled “won” in HighLevel, gathering essential client details such as name, email, and deal information to personalize outgoing emails.

🔍 Validate Deal Fetch Success (IF Node) Checks each record for a valid id field. ✅ True Path: Moves ahead to generate AI email content. ❌ False Path: Logs the event to Google Sheets under the error log sheet.

🧠 Configure GPT-4o Model (Azure OpenAI) Initializes the GPT-4o engine that powers all language-generation tasks in this workflow—ensuring precise tone, correct formatting, and safe structured HTML output.

💌 Generate Personalized Review Request Email (AI Agent) Uses GPT-4o to create a tailored, HTML-formatted email thanking the client for their business and requesting feedback. Includes two clickable CTA buttons:

⭐ Google Review Link: 📝 Internal Feedback Form: Google Form link for in-depth feedback Each email maintains Techdome’s friendly, brand-consistent voice with clean inline CSS styling.