For our purposes Common Crawl's corpus was missing too many websites (possibly due to robots.txt configs of websites) Also we needed some deep coverage which CC could not provide.