There’s a story, usually a series, or a long-running webcomic, or something else that’s long and spread out over time. It has a world—maybe not the best-detailed, but quirky. Interesting. Worth exploring. It takes our expectations and toys with them, it shows us that it’s actually affected by its metaphysics: in short, the world itself is a strength, possibly even a selling point.
At least, until somewhere about two thirds, maybe three quarters, of the way through. (I find the two thirds mark to be more common with book series.) There, something changes. Sometimes, it’s the plot picking up. Sometimes, it’s the creator running out of rocks to turn over, or the characters’ tunnel vision becoming so complete that there isn’t really time to look at the world around them. In all cases, though, the end result is that the ratios of new world detail to other plot elements fall off. I see this a lot, even on some of the authors I most respect.
Why is this a problem? It might not even bother us if the world itself hadn’t been so much fun earlier on. But that in itself is the problem; we’re used to the good world-building. We look for it. Heck, sometimes we come for the world-building, not for this character or that plot or whatever it is that’s going on. Then too, the longer the skewed ratios continue, the worse it gets.
One way to avoid it is to keep the areas where the ratio switches short and fast. It’s not so bad when it’s the end of the book; we’ve seen most of the important things, or at least most of the ones the character’s likely to be tripping over, and the characters and narrative voice alike are too busy worrying about what’s going to happen next and how to make sure that there is a thing happening next to take the time to drop an offhand comment on how this odd little change to the culture in the wake of that event affected the world and consequently them. If the pacing is right, we aren’t even going to notice. On the other hand, if you’re going an entire book, or a six-month comic stretch, and nobody’s seeing much of the background, let alone the quirk therein? That’s going to be an issue.
For the creator whose problem is running out of new things to show, it’s a bit more difficult. Are there any stones left unturned? Is there any place left untouched? For that sort of creator, I’d suggest looking at more minor details. Sometimes, even the character identifying a color with a particularly evocative, world-specific name can be enough to make it clear that you’re not neglecting the world, just spending less time in a position to build on it.
There may not be some golden ratio of world-building to everything else, but it’s still something people notice; unless you (or everyone else) can’t stand your own world, tapering off the shots of what makes it nifty over a long period of time isn’t going to make it any better for anyone.