Just Do It(avc.com)
avc.com
Just Do It
http://www.avc.com/a_vc/2013/01/just-do-it.html
3 comments
You pretty much described the way jobs are structured in the U.S. military officer corps. You typically have fewer than 2-3 years in any given assignment. There are some interesting pros and cons.
On the one hand, it really forces people to learn to learn and adapt quickly -- even in very different spheres (especially because people are usually rotated between operational jobs and policy/admin/planning staff jobs). That is paired with the "up or out" selection mechanism to try to weed out people who are bad at that. Through the arc of your career of many different jobs, the organization's goal is to give you a broad understanding of its whole portfolio of missions and business processes while making them an expert in one or two areas. (Similar to the industry term of being a "T-shaped" employee.) It is also an attempt to shake things up periodically so that people and processes don't get too entrenched. Performance reports lean heavily on what you have accomplished in a short period of time.
On the other hand, you lose quite a bit of corporate knowledge when 25%-30% of any given group leaves every year. As soon as you start to feel like an expert, you are reassigned and have to start all over again. As for the attempt not to get too entrenched... this works well for smaller units, but is markedly less effective at large, staff-based ones (i.e. the Pentagon).
On the one hand, it really forces people to learn to learn and adapt quickly -- even in very different spheres (especially because people are usually rotated between operational jobs and policy/admin/planning staff jobs). That is paired with the "up or out" selection mechanism to try to weed out people who are bad at that. Through the arc of your career of many different jobs, the organization's goal is to give you a broad understanding of its whole portfolio of missions and business processes while making them an expert in one or two areas. (Similar to the industry term of being a "T-shaped" employee.) It is also an attempt to shake things up periodically so that people and processes don't get too entrenched. Performance reports lean heavily on what you have accomplished in a short period of time.
On the other hand, you lose quite a bit of corporate knowledge when 25%-30% of any given group leaves every year. As soon as you start to feel like an expert, you are reassigned and have to start all over again. As for the attempt not to get too entrenched... this works well for smaller units, but is markedly less effective at large, staff-based ones (i.e. the Pentagon).
Yeah -- it's pretty common inside management consultancies, too (which is the model Meg Whitman tried to impose at eBay. I don't think it helped them.)
The military at least has less rotation in the Warrant Officer system (you can be a pilot for ~almost your whole career, or a comms expert, etc.), and has decent programs for technical expertise development for enlisted (until you get past E-8, but E-8 is a great terminal rank). There's a difference being a Blackhawk pilot with a combat units vs. medevac vs. 160th, but it's not that big a difference vs. running an infantry company to a staff officer position in an occupation force to running an acquisitions program in DC.
The officer corps really does seem to sacrifice a lot in order to create O-6 and in particular general officers. If the goal were just to produce the best possible company and field grades, you'd probably do something more similar to how warrants are handled.
One of the biggest problem the military has (in jobs like occupying a country) is gratuitous individual rotation. I spent more time in Iraq than ~any military officer I ever met, because I actually lived there. There were KBR cafeteria workers who spent the entire conflict, and truck drivers who knew more about the situation (by virtue of driving on the roads for 5 years pretty much straight) than the military responsible for them.
The solution to that, IMO, would be to assign higher level units to tasks, and then rotate their subordinate units through on a repeating basis. e.g. 1CAV should "own" Southern Iraq for the duration, and HQ elements should remain in place the whole time. Individual BDE could rotate through. This was essentially the British Empire model. For other capabilities, at least rotate a unit back and forth to the same area -- 6mo rotations would be fine if it were half on, half off, to exactly the same thing; sort of similar to how the navy blue/gold crews worked.
It is way too easy to win on OERs by checking boxes vs. actually doing anything. You end up with people making meaningless changes when something is already fine (sometimes, cycles every other commander!) just to make their mark and get something on the OER.
I personally have respect for JSOC and for combat arms during wartime as a good model for leadership (all the way from E-4 through O-6; never talked to people much beyond that), but I don't think the military model in the US works during peacetime or in roles outside combat arms, medical, aviation, or SOF.
Another interesting model for all of this is how the Soviet officer structure worked -- they'd put their best available candidate into a role, but wouldn't promote him to the official rank right away. You could end up with a Colonel commanding a Corps if he was the best guy available. There were maximum ranks for various positions, but not minimum ranks.
The military at least has less rotation in the Warrant Officer system (you can be a pilot for ~almost your whole career, or a comms expert, etc.), and has decent programs for technical expertise development for enlisted (until you get past E-8, but E-8 is a great terminal rank). There's a difference being a Blackhawk pilot with a combat units vs. medevac vs. 160th, but it's not that big a difference vs. running an infantry company to a staff officer position in an occupation force to running an acquisitions program in DC.
The officer corps really does seem to sacrifice a lot in order to create O-6 and in particular general officers. If the goal were just to produce the best possible company and field grades, you'd probably do something more similar to how warrants are handled.
One of the biggest problem the military has (in jobs like occupying a country) is gratuitous individual rotation. I spent more time in Iraq than ~any military officer I ever met, because I actually lived there. There were KBR cafeteria workers who spent the entire conflict, and truck drivers who knew more about the situation (by virtue of driving on the roads for 5 years pretty much straight) than the military responsible for them.
The solution to that, IMO, would be to assign higher level units to tasks, and then rotate their subordinate units through on a repeating basis. e.g. 1CAV should "own" Southern Iraq for the duration, and HQ elements should remain in place the whole time. Individual BDE could rotate through. This was essentially the British Empire model. For other capabilities, at least rotate a unit back and forth to the same area -- 6mo rotations would be fine if it were half on, half off, to exactly the same thing; sort of similar to how the navy blue/gold crews worked.
It is way too easy to win on OERs by checking boxes vs. actually doing anything. You end up with people making meaningless changes when something is already fine (sometimes, cycles every other commander!) just to make their mark and get something on the OER.
I personally have respect for JSOC and for combat arms during wartime as a good model for leadership (all the way from E-4 through O-6; never talked to people much beyond that), but I don't think the military model in the US works during peacetime or in roles outside combat arms, medical, aviation, or SOF.
Another interesting model for all of this is how the Soviet officer structure worked -- they'd put their best available candidate into a role, but wouldn't promote him to the official rank right away. You could end up with a Colonel commanding a Corps if he was the best guy available. There were maximum ranks for various positions, but not minimum ranks.
By nature this will work better--or at least be most available--working at companies with established headcount needs and turnover. Think of "analyst" programs at investment banks.
I agree though. It'd be cool if I could be sure that the period had value.
I agree though. It'd be cool if I could be sure that the period had value.
Similar to Mark Suster's[1]: What Makes an Entrepreneur? Four Letters: JFDI
http://www.bothsidesofthetable.com/2009/11/19/what-makes-an-...
http://www.bothsidesofthetable.com/2009/11/19/what-makes-an-...
I really like the @pmarca quote: "Marc Andreessen says you either will be the person who tells the computer what to do or the person that the computer tells what to do."
The Silicon Valley (and more so outside) default seems to be "we're hiring you for life", even though the compensation structure (vesting) is set up to encourage no more than 4 years, and in reality more like 1-2 years.
There are a lot of jobs I'd love to take if there were an explicit "accomplish this in 18 months -- if that is successful, you've been successful" -- but where psychologically the idea of "work here forever" would be horrible (even if I intended to quit). VC (EIR focused on security, or some associate type role, ...), politics or activism, turnaround/tech strategy for an existing business in distress, etc.