Bring Me The Head Of Shinseki The General

When I was a kid, I used to watch a lot of old WWII movies and B-grade science fiction on TV. There wasn’t a whole lot of difference between them. The WWII movies involved airplanes or submarines, while the science fiction involved space ships. Beyond that, the plot lines were remarkably similar. The bad guys always appeared absolutely overwhelming and were led by a seriously tough, supremely competent general who terrified everyone. He, and it was almost always he, would usually be introduced in scene that involved him killing off one of his subordinates for failing at something or another. The good guys always were slightly disorganized and Our Hero started the show in deep trouble: he was either being dressed down for some major screwup or the major screwup occurred early in the show. But, because these were the Good Guys, he would be given another chance. Naturally, because this was the nature of that type of movie, Our Hero would then turn out to be the one person who could save the day. It was very clear, even then, that if the good guys killed people off for failing, they would have been defeated. Indeed, this was the major difference between the good guys and the bad guys in a lot of those movies, a point emphasized in some movies where it would also be revealed that Our Hero’s earlier screwup was due to attempts by the bad guys to discredit him. Meanwhile, assuming he survived to this point, the bad guy general would kill himself or be killed for his failure.

The fact is, when a team fails it’s not uncommon to kill off the leader, albeit these days the death is more likely to be symbolic. As news of the problems at the Veteran’s Administration surfaced, General Eric Shinseki ended up resigning his post as head of the VA. Now that he’s gone, naturally all the problems at the VA will immediately disappear.

Well, maybe not.

Killing off the leader can be a very satisfying move, and certainly has a sense of poetic justice to it. Certainly, when the screwup is large enough, it’s more satisfying than killing off some junior flunky. However, as a means of producing effective organizational change it is not necessarily going to be all that effective; indeed, you just may be getting rid of someone you’ve spent a long time training. Instead, it helps to stop and look at the organizational system and understand the forces at play and what is actually taking place. Organizational systems can be very complex and unexpected interactions or badly constructed goals can have serious unintended consequences independent of any particular leader.

For example, at one time Sears Automotive famously gave all of its car mechanics a goal of generating some $200 dollars an hour of billable revenue. The problem, of course, is that they had no control over how many people came to them for auto service nor did they have any control over the particular problems those drivers were having. But the goal focused only on the result: a specific number. Failing to make that number meant failing to remain employed. As a result, the goal became all-consuming: mechanics focused on it to the exclusion of all else. Not surprisingly, they found a way to make their numbers: they invented problems out of thin air. This worked very well until Sears was caught. The wrong short-term goal can blind people to longer term consequences. Changing leaders only helps if the goals are changed as well.

In another situation, IBM in the early 1990s decided that it needed to do a better job of getting technology out of its scientific centers and to the market. They decided that the engineers in the scientific centers needed a stronger incentive. The incentive some senior VP came up with was to tie the performance evaluations of the engineers to how well their products did in the market. This produced a couple of significant problems:

First, the engineers had no control over the sales force. Salesmen had their own numbers to make, and tended to push only those products that they were most comfortable with. They had no particular desire to risk their bonuses! The net result was that it was pretty random which products were being actively marketed and which were not. This, as one might imagine, did not exactly thrill the engineers. The problem was further aggravated by the fact that the sales people were often in a different geographic location from the engineers.

Second, instead of collaborating and cooperating, engineers on different projects now had an incentive to compete with one another. Since they really had no idea how to make one product or another more attractive to the sales team, competition was, at least, mild. Mostly it served to waste energy and distract people. Each new leader who came in was caught up in “the way things were done,” and a lot of good people quit. Replacing the VPs didn’t change anything; it wasn’t until Lou Gerstner came in that anything actually changed. Changing leaders can help, but only if the new leader can also change the culture. Otherwise, you’re just replacing an experienced leader with a less experienced one, and telling the new one that he’d better not make any mistakes. That is not a recipe for success!

In a third situation, a manager was fired because customers were complaining that products were being released too slowly. The manager had been told several times to speed up the process. After the manager was fired, shipment speed dramatically increased. Unfortunately, so did two other things: customer complaints about defective products and, to the surprise of no one except senior management, product returns. I suppose one could argue that they fired the wrong manager in this case. The real culprit, though, was problems with team coordination across the company. Killing the various leaders was not the solution; training them properly was. When that happened, and the various managers were allowed to learn from their mistakes, things began to improve.

Particularly in high profile situations, killing the leader can feel very satisfying. It has a feeling of justice being served. However, quite often it does not actually solve the problem. It’s only when we stop to look at the system and understand what is really happening that we can take the actions that will actually make changes that we want.