I received a mass email years ago from someone I was acquainted with who was heading off to volunteer at a hospital in Somalia. She was spreading the word that they were rebuilding this hospital, lot of details, and then asking for anyone who could devote a month of their life to come to Somalia and train staff or help with construction.
I can't train medical staff, and I doubt I'd be much use with construction, but I reached out to ask if they needed any help securing their facility. The person coordinating the volunteer efforts didn't know until she contacted the woman heading the whole project. She told my friend that everything was being stolen. Sheets, office supplies, mattresses, and the drugs, which would be a huge target, hadn't even arrived yet. They were in serious need of a comprehensive physical security solution and didn't have a plan.
That is a situation where I can provide potentially invaluable help, but despite knowing there was a problem, they didn't even know that there were people in the world who could help with that on a volunteer basis.
So, again, I don't have a solid idea, but if there were any way to help project creators think through and detail their pain points, I think that could be incredibly helpful. My example was extreme, but I'm sure there are many people muddling through archaic, complex problems that don't there are better ways. The Toyota logistics donation to NYC Food Bank springs to mind:
http://www.nytimes.com/2013/07/27/nyregion/in-lieu-of-money-...