As a general trope that's fine. But as far as judging the credibility of any individual assignment, we can't reasonably expect one size to fit all - to be equally "meaningful and not misleading" to all these different types of readers who will have different ideas of what that term means to them.
How can I expect to hit a credibility benchmark for a user that hasn't been identified by me, using criteria that was never disclosed to me? It can't be done. I just had an appraisal report that was prepared for one client/user that was reviewed by an incompetent at an AMC who apparently isn't even an appraiser (let alone a commercial appraiser) but who thought she could run her mouth about my SOW decision being inadequate. Despite acknowledging she had no access to or any way of knowing what my client's policies are.
(BTW, I explicitly named that user and referenced my intention of meeting all the requirements of their policy, so that disclosure probably slowed her roll considerably)
That's an IRL example of the same TYPE or user and the same TYPE of use, but they each had different criteria. Now expand the types from just the one activity into all other activities and it becomes impossible.