Papertrail solves four problems with text data:
It sits on filesystems where nobody sees it, and is not easily consumable via the Web, command-line, API, or email, let alone by whole teams.
One needs to know what to look for. There’s no easy way to see “everything new or strange,” yet that is what’s most useful.
Related data is split across multiple apps, directories, and systems. They can’t easily be managed by content (username, IP address, message ID) instead of by source.
The single datacenter (“behind the firewall”) is dead. Everyone distributes datacenters, servers, clouds, and slices now.