Creating and maintaining a culture of innovation is important for any business. All companies express interest in innovation—who doesn't want to come up with a new product or process?—but few do much about it.

Many companies back away when they see what's actually involved in sustaining innovation. Then there are businesses that really want to make innovation work, but face an internal roadblock: A long-running, moderately successful Six Sigma quality effort led by fanatics.

Don't get me wrong. I'm a big fan of the Six Sigma philosophy of defect reduction. Using statistical quality control methods to shrink or eliminate unnecessary sources of variability is a great idea, and where it's been done well, it has revolutionized businesses.

But Six Sigma can conflict with innovation because it tries to eliminate different ways of doing things in favor of the "best" way. Innovation thrives on trying new approaches. So when Six Sigma starts to be used as a universal solution for every possible challenge, it stops being a useful tool and becomes a religion.

When you're making widgets or processing transactions or making routine decisions, reducing unnecessary sources of variability in process outcome is a good thing. Unnecessary variability costs you in unproductive resources or lower customer satisfaction scores, and that impacts business performance. But what happens when you only do something once or you actually want to achieve a different outcome? That's the essence of innovation in many cases: Take the same inputs but get a different—and better—outcome.

In the decade or so I've been working on innovation, I've never believed these two approaches should be in conflict; indeed, they can and should be complimentary. You should measure your innovation processes and look for ways to make them more effective. You should inspect and adapt as you go along. You should eliminate ideas that can't or won't work for you early in the process so you don't waste scarce investment resources. You should have an effective investment and development process to nurture the ideas you want to try. Doing this, your actions will be entirely consistent with the principles underpinning Six Sigma.

Where conflict occurs, it's often about the definition of unnecessary sources of variability. Many Six Sigma black belts (certified experts in the application of Six Sigma principles) tend to apply the principles as doctrine without regard to context or situational need. People should think outside the box and try different approaches to routine tasks, but the black belts see this as perverting their carefully optimized process by introducing unnecessary sources of variability. And it doesn't help to point out that repeating the process unchanged and expecting a different outcome is a good proxy definition for insanity.

It's tempting to isolate the innovation efforts from the core of the business—and hence from the Six Sigma doctrine—but this is almost always a mistake. It's tempting to force a showdown at the executive level—a sort of sponsor shoot-out—but that's often a mistake as well, because Six Sigma is a good idea and this shouldn't be an either/or outcome. Ultimately, if nothing else works, you may have to change the people who guard the roadblocks, to the benefit of the business's quality improvement and innovation efforts.

John Parkinson has been a business and information technology consultant for more than two decades. Please send questions and comments to editors@cioinsight-ziffdavis.com.