Thoughts on how one requirement can lead to another and then another, the temptation to "boil the ocean" in making changes, and drawing the line with diminishing returns.
Those are great questions to ask to assess diminishing returns.
There's also another way of looking at tasks in front of you.
For example, rather than asking, "is X worth doing?" My software development team likes to ask, "what is the next task that adds the most value to customers with the least cost?"
One problem with the former question is it pits egos against egos on a team. *Somebody* thinks it's worth doing, while another doesn't. No matter who has the final say, someone's going to be disappointed for sure.
The latter question reframes everything such that it's not that X isn't worth doing, but maybe it's worth doing later, when other deliverables are done that are more valuable to the *customer*. And in that situation, the customer becomes the decision maker rather than someone on the team, so internal politics shouldn't play a role in the decision making process.