I'm not saying anything about any particular person's skill. I am saying that a given song either rises to the bar or fails to.

A few years ago when my passion for this songwriting hobby ignited...I agreed with what you are saying.
I saw songs as progressing along a developmental continuum. And I believed that a good song at an early rough stage was better than a bland song finished with bells and whistles. A beautiful woman has to have good bone structure.

People would occasionally say that nobody is even going to give your song a listen unless it sounds finished and good, regardless of potential. I didnt fully believe that, but now I do. Now I think that for a song to be good, it has to first, sound good to the ear's of groups of random listeners.

Personally, I've expanded my hobby to distrokid. At this point, almost entirely alone, I am able to produce mp3s that are acceptable (not great examples of engineering, but acceptable) to the major online stores through distro kid. The progress of my songs is measured in pennies, but my goal is for a stranger to like what they hear well enough to stream my song. My "goal" is to get strangers to stream my songs more than once. I am absolutely convinced that if they dont like the sound that they immediately hear and continue to hear...it ends right there.

So for me...a song can only be good if it has enough appeal to cause someone to listen. And there is no hope of that if it is not truly finished and reasonably good sounding. And then it has to be appealing. And even then...any measurable success is a very very long shot.

But I enjoy trying.

https://soundcloud.com/sunset-poet-of-texas