So my idea is to have an entirely new section dedicated to administrating/auditing of the CM system. I have envisioned this section to be accessible from the burger menu (a new entry to the manifests.xml), see pic at the end. Only users with enough rights would have access to the section. Because it would be a restricted area, it's meant for administrators or developers/implementers. As for the content/functionality of the section, here are a couple of examples:

1. currently logged in users, historical data of logged in users and their past activity
2. possibility to send a direct message to a user
3. current (not real time) and historical data of resource usage by Tridion components (service host, publisher, etc.)
4. centralized overview of log files (similar already exists in WorldServer)
5. possibility to export the log files into a zip (for example to easily send them to SDL Customer Support)

1. I believe it's important to have an overview of all the logged in users of the system (past and present). Since this is not a public system with (tens of)thousands+ of users I believe this to be quite possible to implement. If the need arises, you could check all the user activity, or filter it by categories (publishing, workflow steps, item manipulation (creation, localization, etc.), etc.), by user groups (editors, reviewers, etc.), etc. Using this method an issue could easily be backtracked and analyzed so it doesn't happen again. It would help organizations create and adjust their workflows (not just Tridion workflows but the "in place" workflows/rules of said organizations which they work by). It can also help the organizations to plan their activities according to some trends which they could extrapolate, this could be adding new users to certain roles, planning maintenance activities when the number of users is the least etc.

2. As an admin it would be useful to have the possibility to send a message directly to a user which would show up in the message center (if possibly only his). The reason for the message can be something that was noticed in 1) or just a simple notification for example.

3. This would help developers/server administrators see whether the current resources are enough, and in combination with the user activity overview help analyze whether the noticed is just a one time peak or a constant trend due to high user activity. Also can be useful for planning maintenance activities.

4. Very useful to have all log files in one place for easier troubleshooting. Again, some filtering capabilities would be present.

5. Self explanatory.

NOTE: The idea is 'in-progress', I will update it when I think of something additional.