Website Scraping & CRM Automation

Overview

A merchant platform helps parents discover extracurricular activities for their children during vacations. It aggregates nearby options, such as art classes, sports camps, or workshops, and allows parents to filter and book available slots easily.

Problem Statement

Information about extracurricular activities was scattered across hundreds of organizer websites, each maintained independently. Every year, new organizers published their offerings online, often on separate platforms or local pages. Manually collecting and updating this data for the merchant’s backend became time-consuming, inconsistent, and hard to scale.

Our Solution

We built an LLM-powered on-demand web scraper using n8n, designed to extract and update information from organizer websites whenever needed. The workflow runs weekly to capture new listings and detect changes, ensuring the merchant’s catalog stays current. It also includes validation nodes that automatically remove duplicates and outdated entries, maintaining clean and accurate data across the system. New or modified listings are parsed, processed, and synced directly to the merchant’s backend, keeping the experience seamless for parents browsing activities.

Outcome / Impact

A repetitive manual task that once took hours of effort each week is now an automated two-hour workflow. The merchant’s catalog remains accurate and up-to-date, enabling a smoother experience for parents and reducing operational overhead.

Tech Stack

• ➤ n8n
• ➤ LLM Integration for intelligent parsing
• ➤ HTTP Request and Webhook nodes
• ➤ Custom validation logic