Increase customisation of website scraping.

For instance, I want to scrape only the subdomain of a website. But currently it will scrape everything which I don't want and waste training credits and it's a hassle to delete all the unnecessary info. Also is there a way to automate this?