How do we measure the success of RSA networks?

Comments can get buried on a blog, so I think it is worth pulling to the top some points from the thread here. Justin Kirby has been throwing in some useful provocations ... Elizabeth said she had got great benefit from Fellowship and networking ... but has now removed her comment. Strange.

I said I thought that more help was needed for Fellows, particularly outside London, to connect ... and that a key issue was, as always, what's the purpose?

The issue as always is who benefits? Is RSA for staff, Fellows .. or making some beneficial change in the world? That's where leadership is needed, I think. Matthew has likened the RSA to the RAC of civic activism .... but I'm not sure that's how most Fellows see it.
I guess problems solved, rescues made would give some measure though.

Justin followed up with:

What I really really wanna know is how they plan to measure success. If Matthew T truly sees the RSA as being the RAC of civic activism then they are going to have to nail some kind of Social Impact Assessment/Evaluation Criteria to their mast or else its just more useless sloganeering.

In fact, I'd love to see what evaluation criteria was used to justify the funding of this initiative and where they are at as far as achieving their goals.

I'd also like to know whether some kind of Social Impact Assessment was part of their original funding proposal, because the likes of NING, LinkedIn, Facebook, etc, would have been far more cost effect ways of facilitating a member meets member programme as would a decent bar/cafe at their HQ.

Any ideas on how we do measure success? Is it about helping Fellows network ... and/or making a difference in the world? NESTA commissioned an evaluation, and the interim report is here.
However, the final report was turned into lessons learned, so I don't think we have any final evaluation against funding criteria. Can anyone from RSA or NESTA help on that?
Update: just to make clear there is a full report, which you can download here and decide how far it is an evaluation.