The first big sea-change at EA occurred when Larry Probst succeeded Trip Hawkins as CEO. Trip’s an entrepreneur: brash, charismatic, driven, and visionary. He’s good at selling a dream to venture capitalists – first with EA, then 3DO, then Digital Chocolate. Larry, by contrast, is a born executive. He can be outspoken when necessary, but for the most part his authority speaks for itself. EA moved to a quieter, more mature way of doing business. Larry took the company from being one game publisher among many to being the 400-pound gorilla of the industry.

During Larry’s long tenure, from a shareholder’s perspective the company was extraordinarily well-run, especially for such an unpredictable and fast-moving business. The stock split several times. Revenues grew from $175 million a year to $3 billion. Most importantly, the company never made a mistake that really hurt it badly. There were a number of failures – the short-lived EA Kids division, the effort to get into arcade games, Majestic, the EA.com debacle at the height of the dot-com bubble. But the company never invested more than it could afford to lose, so it was never in real danger.

EA also got quite good about picking the right technologies and getting out of them when they had run their course. It backed the Amiga, an excellent computer that later died off, and made a lot of money out of it; it backed the 3DO, a definitely not-excellent console, and made money out of it too, before the end. EA didn’t touch the Jaguar or the Dreamcast, and in the long run it was just as well. Probably the worst machine it ever tried to support, from a development standpoint, was the Sega Saturn; but again, the company didn’t commit itself to the point of danger. By refusing to be tied to one piece of hardware, it always had a fall-back position if one platform failed.

Early on, EA wisely recognized that simply publishing games wasn’t enough. Intellectual property is nice to own, but games can’t be exploited in as many ways as, say, movies can. The few attempts to exploit game IP in other media didn’t generate much revenue. A better way to lock in a solid revenue stream is to gain control of the store shelves.

If you want the best, eye-level spots on the shelves, you need a distributor with the kind of muscle that can get them for you, and EA set out to be one, another of Probst’s innovations. On the strength of its relationships with retailers, EA’s distribution business remains the basis of much of its strength today.

So what about the EA Spouse debacle, and the over $30 million dollars in class-action lawsuits that EA had to settle for unpaid overtime? What about the allegations of crippling workloads and family-killing schedules? Well, when I was there, attitudes about such things varied widely from project to project, or more accurately, from producer to producer. Some were absolute slave-drivers, no question about it.

But I never got the feeling that it was a matter of corporate policy that managers should be slave-drivers; rather, the company let it go on without paying enough attention. Senior management’s sin was one of omission rather than commission – but failing to look after your people in a people-intensive business is still a sin.