I put down two books in a row last week. Both books looked promising; yes, I judge a book by its cover. And the book flap. So do you, so let's not quibble over this. One book was even by an author I have heard of, who has a good reputation.
That said, I got no more than three and a half chapters into the first book, and about ten pages into the second. Which in both cases turned out to be more than enough. The books were plagued by different problems on opposite ends of the storytelling spectrum, but both were deeply flawed enough for me to put them down.
Well, let me rephrase that. "Deeply" flawed makes them sound like the offspring of someone's high school project and a vanity press. These were not. Under different expectations, they might have been good, and I'll say honestly that had I picked up the second book when it was published in the late 80's, teen me would have probably loved it.
Teen me also liked hot pockets. Tastes change.
Both books were science fiction, which is the extent of what they have in common. The plot of either isn't really important to this post, save that, again, both looked interesting. The first book collapsed under the weight of it's own characters. Specifically, the first three chapters dealt with three separate characters, and sadly only one of them was interesting. There was also head-hopping, wherein the POV changed character mid-scene as two of the characters interacted. That's a particular pet peeve in my case, but it happens, and I've kept reading despite it before. (Michael Crichton was guilty of this at least once.) Only the first chapter, the one with the interesting character, dealt with the science fiction premise of the book. The other two could have been at home in books of any genre. They were stock, they were cliche, and because of it they were boring.
I got the impression the book was the first author's foray into science fiction, and that the author is primarily a writer of literary-esque fiction. There's probably an argument in there about sticking to what you're good at, but as the late Iain Banks demonstrated, you can be good at both. (The entire argument about genre vs literary could actually be buried under the weight of Banks alone, but I'm sure it'll pop up again zombie-like sometime soon.) That said, I think the author needed more practice. Or at least more time spent brainstorming a character sheet.
The second book had the opposite problem. There was too much science fiction all crammed into a small space. What I mean by this is that there was not a single piece of equipment used in the first set piece that the author hadn't renamed. The gun, the plane, the outfit, the ammunition. A few had been confusingly renamed, as it took me a bit to figure out he was referring to some sort of vision enhancement thing. I know there's a fine line in science fiction about these things. You are going to have to create terms, at least beyond "thing," which as I've just demonstrated only gets you so far. But there's a limit. You don't have to reinvent the wheel every time you design a car, after all.
Some of this I think is more a product of the book's time than anything else. I seem to recall a lot of science fiction doing this in the 80's, and perhaps if I went back and reread some of the books I loved then, they wouldn't make the cut, either. Yet I have a strong suspicion the really good ones would make it. (I am tempted to re-read Neuromancer just to see.)
Neither of these two problems, on opposite ends of the genre spectrum as they are, would be inherently fatal to the books for everyone, I am sure. It was for me, though, and I have enough other books waiting to be read without spending time on a bad book no matter how strong the premise.