I think a key aspect of the new website should be clearly describing it as a framework. The biggest problem of the old website was that it implied that things built with rapidSMS were things RapidSMS itself did.
RapidSMS allows for the quick development of applications.
I think another problem with the old website was that it was hard to balance the two primary audiences: 1) devs 2) programs people . The site does need to balance both needs. for a programs person they want to know why choose RapidSMS, how do I use it, who do I hired to build stuff for me, etc For a dev they want to know about the plugable backends.
Anyway, just a few thoughts. last time we built the site it suffered from a lack of a team focusing on it. This time it sounds like you guys will shepherd the project along!