They try to do complete crawls of sites (mostly of user-generated content) that are known to go offline soon or at high risk of doing so, vs the Internet Archive crawler which as far as I know crawls everything every now and then.

If a siteowner is willing to hand over a data export to archive.org or another archival site they don't have to do that, but not many do.