I had the luck (or the talent!) to get an agent with my third submission; I've not sold a novel yet but have written non-fiction. My first novel had three editors want it, but each time their sales teams decided my writing was too bleak. I'm now writing a second novel which I hope will be a little more cheery (although I do find that bleakness easy to drop into). I recently changed agents and was offered representation by my first choice. So I can't pretend to understand the misery of years of submissions.
However, I've seen submissions from the other side, and I wonder if that's got something to do with it. The huge majority of submissions I had to deal with as an editor was completely wrong for the company I worked for.
1) The company I worked for was a book packager which is very different to a publisher: we put books together for various publishers, so usually worked to their commissions, and found appropriate writers and artists to complete those commissions rather than finding a project we liked and then trying to sell it to a publisher. This alone made it highly unlikely that we would ever be able to publish something from the slush pile.
2) We dealt in "esoteric adult non-fiction". So produced books about meditation, religion, prediction, mythology, retreat, tantric sex, all that sort of thing. And yet I would daily receive proposals for childrens' books, novels, books about cars, weaponry, computing: can you see the problem? No matter how good the proposals were (and most of them weren't), the books simply didn't fit our remit.
3) We were looking for writers who could write quickly, write competently, and work well with our editors. The vast majority of the work submitted (somthing like 95% of submissions, I think) was simply not good enough to be published, regardless of the genre it dealt with. Or it was centred around a weak premise, or the submission was peppered with spelling mistakes or errors of grammar. No matter how interesting the ideas were, all proposals with more than a couple of mistakes in it would be rejected.
When I went looking for an agent I spent the better part of a year researching. I looked for books written by people like me, which dealt with similar subject matters (difficult, that one!) and in similar styles. I then found out who their agents were, and considered who else those agents represented. In the end I submitted to a short list of three, two of whom offered me representation.
Since then, my agent (a wonderful woman!) has told me about her slush-pile. Which is, like mine was, full of submissions which simply don't fit her, or submissions which were from novice writers who were not yet good enough for publication. A waste of time and money for everyone involved.
Now, I don't mean to upset or offend anyone here with this, and it isn't aimed at anyone here: but one very effective way to reduce slush piles would be for writers to take more care over their submissions. To not send work off before they are competent writers; to target their submissions more accurately; and to error-check them more closely before they send them off. If these steps were taken then slush-piles would reduce dramatically and editors and agents would have more time to work with the writers who DID have a good chance of publication. I can't see that happening, though. Can you?
if you will, but it's true. Publishers have discovered that if a writer is agented, then the agent felt they could earn money from the writing. It's the early filter.