Two decades ago, when I was just out of graduate school and working in the automotive industry, I got my first introduction to the statistical process-control chart. We used this laborious technique to make sure the machines employed in our manufacturing process did not drift out of control. Composed of three parallel horizontal lines, the “SPC” chart has long been an important tool in quality management. The center line represents the targeted value for the critical performance parameter of a product being manufactured. The lines above and below it represent the acceptable upper and lower control limits. If the product were, say, an axle, workers would plot the thickness of each piece they made on the chart. When I asked why there was typically a scatter of points around the target, my managers cited the randomness inherent in all processes.The “Quality Movement” of the 1980s and ’90s subsequently taught us that there isn’t randomness in processes. Every deviation of the actual value from the target has a cause. It appears to be random when we don’t know the cause. The Quality Movement developed methods for identifying those additional factors-and we discovered that if we could control or account for all of them, the result would be perfectly predictable, and there would be no need to inspect products as they emerged from manufacturing.
The management of innovation today is where the Quality Movement was 20 years ago, in that many believe the outcomes of innovation efforts are unpredictable. The raison d’tre of the venture capital industry is belief in the unpredictability of new businesses. A few ventures will succeed; most won’t, the VCs say. They therefore place a portfolio of bets, extracting premium prices for their capital in order to earn the high return required to compensate for the risk that unpredictability imposes. I believe, however, that innovation isn’t random. Every undesired outcome has a cause. Those outcomes appear to be random when we don’t understand all the factors that affect successful innovation. If we could understand and manage these variables, innovation wouldn’t be nearly as risky as it appears.
The good news is that recent years have seen considerable progress in identifying important variables that affect the probability of success in innovation. I’ve classified these variables into four sets: (1) taking root in disruption, (2) the necessary scope to succeed, (3) leveraging the right capabilities and (4) disrupting competitors, not customers.
Of course, building successful businesses is such a complicated process, involving subtle interdependencies among so many variables in dynamic systems, that we’re unlikely ever to make it perfectly predictable. But the more we can master these variables, the more we will be able to create new companies, products, processes and services that achieve what we hope to achieve.