Thursday, December 11, 2008
Kill the Perfect
The pursuit of operational perfection should be killed for most innovation companies. Perfection has become a competitive position that everyone competes on thereby commoditizing perfection. When everyone is perfect, no one is. The Japanese have the concept of kaizen, continuous improvement. Tacit in kaizen is that perfection can never be reached. Our goals should not be unattainable. They should be framed in areas where we have a competitive position, ripe for continuous improvement. Often, that’s our ideas.
The problem with operational perfection in business lies in what I call the Stakeholder Value Paradox. The Paradox says that those most likely to be in a position to judge perfection have the smallest worldview on which to judge it. Those less likely to judge perfection have the largest worldview to truly understand what perfection looks like. The danger is that we as knowledge experts chase operational goals to satisfy operational stakeholders at the detriment of the strategic stakeholders. Certainly, poor execution won’t be tolerated. However, there is a fairly large range of acceptable values between above average and nearly perfect which will suffice. It’s the pursuit of the last 10% of perfection that gets us in trouble.
In fairness, the existence of the last 10% is a competitive opportunity for a production-oriented firm. But, most of us aren’t built from the ground up to be a production-oriented firm. Instead, we’re a marketing-oriented firm trying to compete under the playbook of firms with a genetically different makeup. In my experience in innovation companies, we’re usually looking to engage Moore’s Innovators, Early Adopters, and perhaps some of the Early Majority. These fellows have a degree of tolerance for less than perfection. They have no tolerance for lackluster ideas.
One of my professors always used to say “It ain’t strategy if you can buy it”. You can buy operational near-perfection. If your firm didn’t breed it, buy it. Spend your time thinking.