We hold these truths to be self-evident: We recruit people who are quick on the uptake, people whom we like and need. We encourage coworkers to get along. We make decisions based on experience. Now flip those assumptions upside down. Hire slow learners, people whom you dislike and don't need. Encourage them to defy and fight with their managers and peers. Think of ridiculous things to do, and do them.

As management practices go, these sound wrong -- even weird. According to Robert Sutton, a professor of management science and engineering at Stanford Engineering School, they're some of the most powerful practices for generating and capitalizing on new ideas. "What's weird," says Sutton, "is that people say that they want innovation, yet they can't depart from their deeply ingrained beliefs and practices about how to treat people, make decisions, and structure work."

During an illustrious career as a teacher and consultant, Sutton has distilled a kind of antiwisdom about innovation. He is codirector of Stanford's Center for Work, Technology and Organization, and he has worked with a range of companies, including Clorox, Hewlett-Packard, IBM, IDEO, and McKinsey & Co. In his provocative and relentlessly useful new book, Weird Ideas That Work: 11 1/2 Practices for Promoting, Managing, and Sustaining Innovation (Free Press, 2001), Sutton presents a slew of tactics and strategies for bridging the innovation gap. "A weird idea works because it trips discomfort," says Sutton. "The idea is to flip from autopilot to mindful creation."

In conversations with Fast Company, Sutton discussed the ideas behind his weird ideas and described how innovative people and teams bring fresh eyes to bear on every challenge.

Don't Get Too Weird Too Fast

Before you start getting weird, there's a major caveat. During the past several years, we got so frenzied about innovation that we overstated the notion that innovation is fun and desirable and that routine work is boring and not as valuable. A few management gurus who shall remain nameless did a real disservice to the business world with their exhortations that everyone should innovate all the time! That's my idea of hell.

The mantra that we have all lived with for the past five years is, "Innovate or die!" But it's just as accurate to say, "Innovate and die." All the excitement about all things new obscured the fact that most new ideas are bad and most old ideas are good. It's a Darwinian principle: The death rate of new products and companies is dramatically higher than that of old ones. Dozens of new cereals fail every year, while Cheerios and Wheaties persist. Even wildly popular new products such as Beanie Babies fade, while Play-Doh remains on the scene. Still, the world does change, new technologies are developed, business models do mutate, and customer demands do migrate. So the question becomes, Which horn would you rather be gored by? That's the innovator's dilemma. You can't choose between innovative work and routine work. That's like asking, What's more important: your heart or your brain?

The best companies organize themselves around two useful fictions. Routine work is guided by the assumption that everything is a permanent condition. Decisions are made and work is structured as if the future will be a perfect imitation of the past. That's how Intel dominates the chip industry. When Intel managers agree that something is a good idea, they use their "Copy Exactly" technique to implement it identically in every Intel factory around the world.