There is a lot written about how to write, and perhaps a bit less written about how not to write. Somewhere in the middle is perfection. The closer a manuscript hits that mark, the more likely it is to become published.
In my mind there are two types of writers: Those who plan every detail so that by the time they have written, rewritten, and edited each chapter before moving forward it is totally rigid and has no energy left; and those who write stream of consciously thinking every word is golden, which results in a whole lot of superfluous dribble and clever remarks to the reader.
Either style has to have an end point in mind. The story line has to stay on track and get from A to B with little distraction and wandering off to too many antidotes that do not advance the story. The writer has to keep asking, "What does this have to do with the story?" and stay true to the story line.
The elements of good writing that I look for when I consider a queried piece are these:
1) Does the story move along to a logical endpoint in an interesting way to keep the reader reading?
2) Are the characters developed with a heartbeat so the reader cares about them?
3) Is the dialogue realistic or forced? (Someone once told me they write the first draft, then go back and add the dialogue. I can't imagine how that would work.)
4) Has the ms been polished enough to show me the author cares about the craft?
5) The ms has to be marketable. Marketability has many facets. It has to appeal to a definable audience, and for me, it has to contribute to the Cactus Rain catalogue.
Before a book can reach the reader, the author has to first reach a literary agent or publisher and sell the story. It would be insane not to do multiple submissions, but each should be tailored to the recipient. Do the research. It makes no sense to spend years writing the perfect ms then sending it to the wrong person.