First, I wanted to thank all of the Effective Altruism Global organizers and participants. I found it to be very valuable and overall well put together. There was obviously a ton of work put into it, most by conference organizers who I don't believe will get that much credit for it, and I very much commend their work.
That said, there's always a lot of room for new ideas, and I find I often get a bunch of ideas at and after these conferences. Because of the EAGx events, ideas described now may be able to be put into action somewhat soon and experimented with.
As may be expected, I recommend that people make all of their ideas be independent comments, then upvote the ideas that they think would be the most useful.
I agree that thinking explicitly about the goals of the conference would be good. Fwiw, though, my instinct is that trying to quantify it into an EV estimate would be a bit of a distraction from the main benefits.
In my experience, in the cases where there are EV calcs, few people will pay attention to them anyway, though the ones that do seem to find them interesting & useful. It could be a distraction to the people who make it, but I don't see it being harmful to more than 10 people (in the worst 10th percentile).