In Iraq, a hodge podge of geeks and reservists (many of them cops or corporate "competitive intelligence" specialists) came up with lots of new ideas about how to collect, analyze and distribute intelligence. This was usually done at the divisional or brigade level, although some battalions, and even infantry companies, have come up with their own innovations. It was innovations like this that led to the capture of Saddam Hussein, and many prominent terrorists.
To their credit, the big shots in the intelligence establishment have not thrown their weight around too much to try and stifle all this creativity. But, at the same time, there has not been a lot of enthusiasm for taking these combat zone innovations back home. Even the higher headquarters, like Central Command (which is in charge of Iraq, Afghanistan and most of the Middle East), have not been able to consolidate and innovate as much as the smaller operators.
It could be that the larger intelligence operations have their hands full managing all the people and gadgets they already have. Meanwhile, the divisions and brigades that are doing the fighting, have very slender intel resources, and thus must do a lot with what little they have. The combat troops also have an immediate incentive to make their intel operations work. If they don't, they, or people they know, could get killed.
Despite many new intelligence techniques developed by American troops in the last five years, the intelligence "establishment" has not managed to shake loose from its legacy of inflexibility and unresponsiveness. This, despite frequent testimony, in Congress, by numerous four-star generals, and lower ranking troops, that the $40 billion a year intel bureaucracy was not delivering much of value.