There's a wonderful article in Vanity Fair this month by Kurt Eichenwald, "Microsoft's Lost Decade," which details the cannibalistic culture that took Microsoft from invulnerable to highly vulnerable. Ten years ago, we all were echoing that Microsoft had all the applications, all the operating system revenue, and all the talent, and so was an impregnable fortress. It would swat all competitors like so many mosquitoes. And we were dead wrong.

Fifteen years ago, we were saying the same thing about IBM. Its profits were larger than those of the next seven computer companies combined and its R&D budget was larger than anyone else's profits. How could we be so consistently incorrect? It's a talent.

Eichenwald puts the blame squarely on CEO Steve Ballmer and makes the case that his talents lay not in technology, but in sales/marketing and other assorted alleged skills. But the real reason is management. Microsoft, like every other large company, was the victim of its own attempts to bring order to chaos. Ten years ago, everyone at Microsoft was getting rich--the stock appreciation alone created 4 million (!) millionaires as the stock options became vested. Then the music stopped. And when it stopped, the only way to get ahead was to get promoted. You couldn't get rich on your stock options anymore because the stock wasn't moving up.

[ If IT organizations don't get their act together, the best and brightest will chart their own destinies. See Here Comes Corporate Brain Drain. ]

Those who had gotten rich either physically or mentally checked out. Then Microsoft subscribed to the forced curve, which said that only 20% of any team could be judged "excellent," 50% had to be called "acceptable," 20% had to be labeled "deficient," and the bottom 10% had to be made eligible for "reduction." Sounds like some MBA theory; in fact, it's a lift from Jack Welch's reign at GE.

So what would you do if you worked at Microsoft and wanted to be a survivor? Would you gravitate to the high-performing teams? No way! Even if you were quite accomplished, you might wind up in that bottom 30%, and you wouldn't be promoted within your lifetime. No, No, No--you would migrate to poorly performing teams, where you could be a superstar. In the land of the blind, the one-eyed man is king. In a high-performing team, you might be cannon fodder even if you met all of your yearly objectives, especially if your peers exceeded theirs.

The second behavior of the Microsofties was even more insidious. In any team, you would delicately and subtly sabotage your peers. Did someone on your team need data by Wednesday for a report due to a Big Boss on Friday? Then why not send it late, send it incomplete, send it by interoffice mail. You sent it, didn't you? No one could complain that you were non-compliant, except that your peer would miss a deadline, which meant that someone's year-end review wouldn't look quite so good.

The result of all of this insidiousness: products were late and decisions were postponed--even as Microsoft racked up record earnings and cash kept accumulating. This dysfunctional "gaming the system" was just the sort of behavior which would have driven Steve Jobs crazy--and which he never allowed.

There's a lag time between when a company gets into trouble and the world realizes it. For a while, everything is hunky dory--the company is cranking along, making some acquisitions, building derivative products. But there's trouble below the water line. In other words, its success is a predictor of its failure.

I remember a discussion I had once with Cisco CEO John Chambers. I was keynoting a Goldman Sachs conference evaluating the technology sector, and I was rating companies on growth, international sales, products, and the ability to handle adversity. I gave Cisco high grades on all but the last category, and Chambers was ripped that I would even suggest that he didn't face sufficient adversity.

"Howard! All of my key management is filthy rich, and they're all leaving because their stock options are golden--isn't that adversity?" Chambers said. I had to admit that I didn't think that when your company was worth half a trillion dollars you'd be crying the blues.

Every company needs to find some common enemy upon which to rally the troops. If it can't find such an enemy, it will find the enemy within. This is where the backbiting, the gaming of the system, the convoluting of the mission takes place.

Too many years ago, I rowed crew at the University of Pennsylvania. The coach, Joe Burk, had just installed the first ergometers right on the shell, where one to four lights would light up at the cox's position, depending on how hard you were pulling. The coach would follow the boat, looking over the cox's shoulder, and shout: "Anderson, pull harder! You're only lighting three bulbs!" And I would drive my legs harder... while he was looking. Then I (or my teammates) would slack off until the coach came by again. Needless to say, we didn't win many races.

This is what has happened at Microsoft (and IBM, AT&T, Yahoo, HP, Cisco, and SAP), and it will happen sooner or later at Google, Facebook, and Kayak. Maybe not at Oracle, as long as Larry Ellison remains the meanest SOB in the Valley and just won't put up with this BS.

Karl Marx said that capitalism breeds its own destruction, and I think he may have overstated his case. Sorry, Karl, but success can destroy a company much faster (examples: Wang Labs, Digital Equipment).

Groucho Marx (no relation to Karl) was once told that a solution was so obvious that "a 5-year-old child could see it." His answer: Send for a 5-year-old child.