There are many more labs than that, and humans aren't designing unique scraping processes per domain.