This is a wonderful project, which I completely support in concept.

I would comment, though, that where an existing relationship already exists, a new affiliation will likely detract from what is already happening.

I also try very hard to avoid rolling rocks uphill - I much prefer rolling them downhill, so existing affiliations would allow exactly this. Having said that, two tiny NGO's swapping emails would not constitute "pre-existing partnership" so, where would the threshold be, or are you deliberately avoiding this issue with your the random strategy?

The value of existing "pilot" regions, which may allow you to adopt their pre-existing work under your banner is pretty valuable in your strategy?

What you could do is have a "cut off" point, where after a certain date, random allocation takes place. Before that date, twinned areas can apply to match themselves.

I would also like to know in your concept exactly which type of people twin with whom? Is it governments, social entrepreneurs, NGOs?

I would be very interested in looking at ways to link Kaduna State, Nigeria with a random area, although some pre-existing linkages are already happening. What is your process and timescale?