Intel cells

Rifle-company Intel Cells:

Once again, the Marine Corps leads the way. The databasing/sharing concept is the most important one. Companies, like any other military unit, have seams with other units -- they control a given area, but the enemy moves in and out of that area. What happens here matters across the way, and vice versa. Better sharing is critical.

Too, analysis tends to get better with experience, but added responsibility and distance degrades the benefit. The G2 section normally is better at recognizing patterns and important details than the S2, because its members have more experience with it. They have a problem in being responsible for a lot more areas, so they can't focus down on any given area except periodically. They have an additional problem in being distant: instead of dealing with the area eyes-on, day in and out, they're often learning by reading reports. Even the best report leaves out a lot of information that you get by being there, and looking things over -- information that may not seem important now, but that might be later.

Adding extra eyes to your data can compensate for that to a certain degree. If you have two guys with less experience looking at the data, they may still both miss the pattern; but it's better than just one guy. And the closer you get to the grunts on the ground, the more you'll be drawing on patterns they've seen playing out. They've got all that 'extra' information that isn't captured in their reports, the stuff they saw or heard that didn't seem important or relevant at the time -- but that can come out when they get together and compare notes with the company across the way.

Better sharing across companies, as well as up-and-down between organizational levels, has the potential to substantially improve predictive analysis on the ground. The Corps is doing something smart here; but that's what we expect from the Marines.

No comments: