There's a great book waiting to be written about how to handle working in a setting where 'leapfrog' is rewarded, when you're the one being leaped.
Organizationally, leapfrog is when someone goes over her supervisor's head to get what he wants. When the organization makes a practice – even if only sporadically – of rewarding leapfrog, it will instantly turn too much of its energy inward on political maneuvering, and lose focus on dealing with the outside world.
(Before the inevitable flaming, I'll draw a distinction between leapfrogging and appealing. Appealing is open; the person whose decision is being appealed gets a turn to speak; the person making the appeal carries the burden of proof; the decision is based on the merits of the case. Leapfrogging is secretive, the person being leaped often doesn't know it happened until after the fact, the burden of proof is random or nonexistent, and the decision is based on heaven knows what. I have no problem with an appeal process, properly understood, but a huge problem with leapfrogging.)
Although there can be a short-term temptation to reward the occasional leapfrog, the long-term damage is terrible.
The temptation comes from a few factors. First, and most obviously, it makes an obviously-angry person go away for a little while. This shouldn't matter, but sometimes it does. Of course, it creates another angry person very quickly, and a whole extra bunch of angry people over time as folks discover that anger gets rewarded.
Second, sometimes the leaper makes a compelling (immediate) case. This is especially true when the person in the middle has some known shortcomings or blind spots that seem to be relevant in a given instance. Everybody is wrong sometimes, regardless of level. Heaven knows I've made a few decisions that, looking back, make me wince. It happens. A higher-up who isn't taking the long view, who is focusing only on what's right in front of him, will be tempted to 'fix' the immediate problem.
Finally, some people live in terror of the dreaded “he knew about it and didn't do anything!” By doing something, even if it's ill-advised and destructive, you can look decisive. George W. Bush is a terrible decider, but he's decisive. Gotta give him that. In the immortal line from Animal House, “There's a time for thinking and a time for acting. And this is no time for thinking!”
But if the leaper gets rewarded, the leaped (lept?) is immediately rendered irrelevant. And the irony is that what might, at first, look like a blow for the little guy – I beat my boss! -- actually works to concentrate power more centrally, in the office of an even bigger boss. Because a boss who makes decisions on a whim can change them on a whim. Replacing a process model – as slow and frustrating as it is – with a patron/client model only empowers the boss.
All of that said, it's still hard to know how to function as the leaped/lept. You can try to out-leap the leapers, but that diverts energy from what needs to be done and implicitly endorses what they're doing. You can try to ignore it, but when the higher-ups change the play, they change the play. You can try to talk the higher-ups out of it, but in my experience, self-awareness is a rare thing; very few people are capable of moving from abstract recognition to actual behavior change. Or you can try to pretend that whatever the final result was was what you intended all along, but that comes at the cost of both credibility and self-respect.
Is there an elegant strategy I'm missing? Other than the Hobson's choice of either rolling over or quitting, is there a better way to manage a situation in which leapfrogging is rewarded?