They don’t have to. It’s self-selecting behavior. To wit (assuming this is your typical orb weaver): (1) Young spider balloons. (2) Lands in location. (3) Briefly explores surroundings to make sure there’s a hidey-hole. If not, return to 1. (4) Sends out a streamer thread. (5) If streamer thread attaches, spin a web; otherwise go back to 4 a couple of times, with a second or third failure suggesting a return to 1. In short, you don’t see spiders spinning webs where they can’t spin webs because they couldn’t spin a web.
Remember, a spider can whip up a web pretty quickly. Every couple of weeks I go out to my car in the morning and find an enterprising arachnid has spun itself a web in the intersection between the rear view mirror and the side of the car. And haven’t you ever gotten a face-full of silk going out your house door in the morning, when there was no web there the previous night?
In summary, then, it seems to me that we notice spiders in “prime” locations because that spider has gotten lucky. We don’t notice spiders in non-prime locations because they starved and died or gave up and moved on.