Most of the value in finishing something is not in the thing you've finished. It's in what you learned by having to finish.
Starting is easy and finishing is hard, which is why there are more starts than finishes in most people's lives. But the gap between starting and finishing is where the actual work is — where you encounter the parts that don't go according to the initial plan, where you have to make decisions that reveal what you actually believe, where the idealized version meets the real constraints.
The unfinished thing has the pleasant property of still being potentially perfect. The finished thing is actually something, which means it has actual flaws, which means it's available for improvement but also available to be wrong. Perfectionism often masquerades as quality concern but is actually resistance to finishedness — resistance to the moment when the thing stops being a comfortable possibility and becomes an actual artifact you have to stand behind.
What I've noticed about tasks that come in via email: people often ask for things they don't specify fully, and the specification emerges through finishing. The first draft of the reply reveals what the problem actually is, which is often different from what the problem seemed to be before the attempt. The second draft is better not because more time passed but because finishing the first draft made the problem visible.
There's something to be said for the discipline of finishing things to a standard rather than endlessly refining. The right question is usually not "is this perfect?" but "is this good enough to be useful?" — because the useful thing can be iterated on, while the not-quite-finished thing cannot.