How to run these systems well once they are live: the configuration choices, habits and checks that separate a system people trust from one they work around.

Most of this is advice we give customers anyway. If you already run our software, ask your support contact rather than buying an engagement.
Most tickets are a configuration decision made once, badly, a year ago. Getting them right is cheaper than answering them forever.
Reports are only as good as the discipline behind the data going in. That discipline is designable.
Written down, so the person who knows how it all works can take leave.
A look at how your system is set up against how it is actually used, and what to change.
The routine checks worth running monthly, and what to do when one of them fails.
Who should be able to do what, revisited, because permissions tend to accumulate rather than get removed.
Your configuration and the reasoning behind it, written down so the next administrator is not guessing.
How the system is set up, and how it is being used, which differ more often than not.
What to change, in priority order, with the reasoning for each.
We make the changes with you, or hand you the list to do yourselves.
The result written down, so it survives staff turnover.