XGlass Portal Scraping

A client from France, operating a car windshield business, needed to manually look up data for each vehicle from the XGlass portal. The challenge was that the portal updates its data periodically—often after several months—forcing the client to repeat the same manual data entry process. I built an automated solution to extract and structure this data, eliminating repetitive work and enabling better decision-making.

Outcomes

The Problem

The client relied on manually searching the XGlass portal to retrieve windshield data for each vehicle. Since the portal updates its data periodically, the same process had to be repeated every few months. This resulted in significant time loss, human error, and inefficiency, especially as the dataset continued to grow.

Challenges

• No API access — data was only available through a dynamic web interface
• Frequent structural changes in the portal required adaptable scraping logic
• Handling large volumes of data efficiently without breaking the scraper
• Ensuring clean and structured output for direct use in the client’s CRM
• Avoiding detection and maintaining stable scraping sessions

Solution

I built a robust web scraping pipeline using Selenium to automate browser interactions and handle dynamic content, combined with BeautifulSoup for efficient HTML parsing. The scraper runs periodically to extract updated windshield data directly from the XGlass portal. The extracted data is processed and cleaned using Pandas, then structured into a standardized format for easy analysis. Finally, the data is exported as CSV and synced to Google Sheets, allowing the client to seamlessly integrate it into their CRM system for price comparison and operational use. This automation replaced a fully manual workflow with a reliable, repeatable process that delivers up-to-date data with minimal effort.

Tech Used

Demo

Sample Data set