This n8n workflow automates the collection, enrichment, and analysis of e-commerce product listings using Bright Data and AI, then delivers an HTML email report with the most competitive offers.
🚀 What It Does
- Pulls product titles from a Google Sheet.
- For each product, searches a Bright Data marketplace dataset (Google Shopping) for available listings.
- Extracts relevant fields: price, title, seller name, and listing URL.
- Sends this data to Google Gemini for AI-powered Markdown report generation.
- Converts Markdown to HTML and styles the output for better email rendering.
- Sends an email report for each product with the top 20 most affordable offers.
🛠️ Step-by-Step Setup
- Load product list from Google Sheets.
- For each product title, run a Bright Data filter request (case-sensitive match).
- Poll the snapshot status until it is ready.
- Retrieve snapshot content and clean the results with a Code node.
- Pass the results to Gemini (PaLM/Gemini Flash) for analysis and report generation in Markdown.
- Convert Markdown into styled HTML using Markdown + Code nodes.
- Send formatted email to a predefined recipient.
- Return to the loop and repeat for the next product.
🧠 How It Works
- Loop Control: SplitInBatches handles product-by-product processing.
- Snapshot Handling: Snapshot status is polled every 30s until success/failure.
- AI Formatting: Gemini summarizes listings and formats content.
- Error Handling: Failed snapshots produce a warning message and resume the loop.
📨 Final Output
- Each email contains:
- The product name
- A clean HTML of up to 20 sellers with lowest prices
- Links to listings
- AI-generated pricing summary
🔐 Credentials Used
- Bright Data account
- Google Gemini (PaLM/Gemini Flash)
- Google Sheets (OAuth2)
- SMTP Email (emailSend node)
⚠️ Important Notes
- Item title search is case-sensitive. Typos or casing mismatches may result in no results.
- Requires n8n-nodes-brightdata community node to be installed.