My recent NECRWA conference attendance had an element in common with every writer conference I've attended. That is, the most common question presented to editors and agents: "What type of book do you want?" Authors are always very frustrated by the standard response heard at every conference: "A really good book." They ask what makes a really good book, or makes an editor/agent think a book is really good. If it was that cut-and-dried, if we could just lay out a list of elements, then every book published would be a best-seller because that's all we'd buy. But it's not that simple. It's like the Supreme Court Justice's comment when asked to define pornography: "I know it when I see it."
A story has so many elements--writer voice, style, pacing, plot, characterization, conflict and resolution, satisfying ending...and on and on. And what grabs one editor/agent will be ho-hum to another--personal experience, marketing inclination, and taste do play a part.
Unfortunately, being "a really good book" does not always guarantee being a successful book in terms of sales. Current market trends and reader fads, the publishing and book retail climate, the author's reputation and fan base, and again on and on... So many things impact a book's sales potential, as much as the story itself. (Every editor/agent knows the frustration of having a book they feel is absolutely fantastic, and seeing it not get the attention it should or be a hit with enough readers.)
Frustrated in their attempt to pin down "what do you want" from the editors/agents, the authors then move on to asking what we don't want. (No, we don't answer, "We don't want a really bad book.") Specific genres or plot elements will be raised as contenders for "I've heard no one is buying any more ----." Recently the common comment is that publishers don't want any more vampire romances. Yep, it was asked again at the NECRWA conference, and one of the editors or agents (alas, I don't remember who spoke up for us on this) gave the standard and very true response. You can still sell an editor on a book in an over-populated genre, as long as it is not the same-old-same-old. It needs a new twist or unique take. Go read lots and lots of vampire romances--and then write something different, make an element of your vampire or your vampire's world a change-up on what everyone else has written. (Forget having your vampire sparkle in sunlight--it's been done.)
Right after returning from this last conference, I read a new book that I would indeed say is a really great story and a new twist on an overdone genre. I wish I'd read it before the conference so I could have used it as an example. Angels' Blood by Nalini Singh. It's the start of a series, and is about vampires, vampire hunters and angels. Vamps and their hunters are way common in paranormal romances nowadays; angels are newer but look to be heading toward overused territory. But Ms. Singh's book just...stunned me. Blew me away. I love it, it will easily rank on my top-ten list for the year. Her development on the character types is unique; her vampires were intriguing and her take on "angels" was fantastic and startling. Her world building is superb and incredible. This is the perfect example of "not the same-old-same-old". And it had the effect that editors and publishers want from readers--it sent me immediately to the author's website to find out what other books she had, when the sequel to this one would be out. (Aargh, not for a year! Torture! But two novellas in the same world will tide me over to the next novel.)
So, you want to know what types of books agents and editors want? A really good story that will play on current reader tastes but is different from all the rest. Is that enough information? Okay, get writing.