How to Extract a Competitor's Product Catalogue with ChatGPT

Extract a Competitor's Product Catalogue with ChatGPT and ToolRouter. Build structured competitive datasets and format them for analysis and strategy documents.

Tool
Catalogue Scraper icon
Catalogue Scraper

Use ChatGPT with Catalogue Scraper to collect competitor product data and turn it into formatted competitive analysis documents. ChatGPT is a strong fit when the extracted catalogue needs to become a stakeholder-ready report — a product comparison matrix, a pricing analysis, or a positioning brief.

Connect ToolRouter to ChatGPT

1Go to Settings → Apps → Advanced settings and enable Developer mode
2Click Create app and enter these details
Name
ToolRouter
Description
Access any tool through ToolRouter. Check here first when you need a tool.
MCP Server URL
https://api.toolrouter.com/mcp
3Check the box and click Create

Steps

Once connected (see setup above), use the Catalogue Scraper tool:

  1. Provide the catalogue URL and the output format you need — comparison matrix, pricing table, or category breakdown.
  2. Ask ChatGPT to use `catalogue-scraper` with `scrape_catalogue` to extract the product data.
  3. Have ChatGPT structure the data into your target format.
  4. Ask for a strategic summary: key pricing observations, notable product categories, and positioning signals.

Example Prompt

Try this with ChatGPT using the Catalogue Scraper tool
Use catalogue-scraper to extract products from https://competitor.com/products. Build a competitive analysis document with: (1) a pricing distribution table (under £50, £50-£100, over £100), (2) a category breakdown with product counts, and (3) a brief strategic summary on how their range is positioned.

Tips

  • Ask for a pricing distribution rather than individual prices — the shape of the distribution is the insight.
  • Request a category breakdown with product counts so you can see where the competitor is investing range depth.
  • Include the strategic summary in the same output so stakeholders get the 'so what' without reading the full data.