Writing a large project in graduate school, whether it is a master's research paper or a PhD dissertation, is an endurance race. Plenty of people know and get that--that it is a marathon and not a spring. But the race is not just with oneself but with one's adviser. How so?
Supervision largely entails this cycle of: discuss the project with the student, read a draft, recommend revisions, read the revisions, recommend more revisions, read the revised revisions, etc. When does the supervisor tell the student that the work has been sufficiently revised and is ready for defense/approval? Perhaps when the work is perfect? Nope. Perfection is the enemy of the good enough. What does good enough look like? Well, it looks like one thing early on in the project and often looks like something else at the end.
Why? Attrition or exhaustion. The supervisor will often find that the marginal utility of reading another draft or the student revising another draft declines, sometimes quite sharply. It can decline because the student anticipates the likely problems and writes a terrific draft OR it can decline quite sharply if there is a realization that not much is going to change.
Students vary in how they play this race, and most are unaware of the fact that it is a race. Professors, on the other hand, are often too aware of this dynamic. Perhaps I am being too whiny since this month surprisingly became major research project reading month for me with multiple students giving me stuff to read all at the same time.