Right now I’m only a single chapter ahead of my party. I think I’m going to keep it that way. This way I don’t spend time on something the party probably won’t experience. As we all know: there’s no telling what your PCs are going to do.
To date I’ve been crafted each chapter to end on a cliffhanger. I told them to imagine a big “To Be Continued…” floating in the sky. And, to date, this has been centered on their decision on which path to follow after the ‘big event’ of the chapter.
I haven’t formalized a chapter plot because I don’t want it to be predictable. However, so far, there’s been a ‘big event’ and then a decision point. This has worked well because of what I wanted to do, but I see this will probably not work in the subsequent chapters.
One thing I need to keep in mind: this is a quest campaign that is time-boxed. I need to start helping my PCs do the investigative work required. I need to start encouraging them, reminding them, they have something to do with a deadline. Chapter 3, the chapter I’m currently defining, will start laying this groundwork.
As for chapter length: I’ve not tried to fit a chapter into a single length of time. I’ve tried to focus on the story and where there might be a natural end-point. Now game sessions, I’m trying to keep those to around 4 hours. A chapter can have as many game sessions as is necessary.
And I’m trying to stay keenly aware of “over preparation”. As I understand it this is when every single detail is mapped out, typically at the chagrin of your PCs. However, I also feel this phrase is subjective. I do prepare a ton for my NPCs to make it easier for me to play them, like an actor to a role. However, not every NPC gets this treatment; some NPCs don’t really drive the story forward. As for encounters, sometimes I’ll only have a monster or two defined and set the PCs against them. Sometimes I’ll have more elaborate plans, not as a script but as a contingency. Again: you have no idea what your PCs are going to do, or not do.