About The Gazette Search Back Issues Contact Us    
The newspaper of The Johns Hopkins University July 6, 2004 | Vol. 33 No. 39
Thinking Out Loud

William R. Brody

By William R. Brody

"Perfect is the enemy of the good"

Having trained as an engineer, I have to admit that I am susceptible to the engineer's credo: "If it ain't broke, fix it!" An engineer never resists the temptation to tinker with something that works until it ceases to function entirely. CEOs of high-tech companies will tell you that without strict deadlines, products would never get developed. Why? Well, engineers are never completely satisfied and want to continue to rework the design until it is perfect. The trouble is, perfect is the enemy of the good.

And it's a trouble not just confined to engineering departments within technology companies, either. The concept, unfortunately, seems universally applicable to large organizations. While it's in part driven by the desire for perfection, I think the dominant reason people in large organizations keep reworking designs, ideas, processes, procedures, memos and so on is fear of failure. Bureaucracies typically do not reward success and certainly do not value risk taking. Rather, they tend to penalize failure and emphasize playing it safe.

Why does it take the FDA so long to approve drugs, even for diseases with 100 percent mortality? Because a single mistake by the FDA appears on the front page of The New York Times, while rapid approval of important drugs is only likely to be noted — if at all — deep inside the pages of The Wall Street Journal.

HopkinsOne is a massive organizational undertaking to reinvent our business processes. We need to engage the entire Hopkins community in developing more efficient and effective ways to administer our enterprise. One of the important aspects of this endeavor will be engaging our employees in rapid cycle process redesign. In other words, "try it, fix it" and then "try it again, fix it again." Progress comes through using multiple iterations, each one an improvement on the last.

Contrast this to a methodology that uses elaborate and time-consuming planning followed by a single implementation. First of all, we will never reach perfection, no matter how much planning we do. You will always encounter the unexpected when you redesign processes, regardless of how much planning you do. So rather than shooting for a single rollout of a new process, when we use rapid cycle change implementation, we get many opportunities to fix things that don't work. And with short cycle times, it turns out that the people who work on the process redesign feel empowered and valued. Because we expect that the initial implementations will have imperfections, there is no penalty assigned to the group when the process has "bugs."

In the late 1870s Thomas Edison announced he was going to devise a safe, reliable, affordable electric alternative to gaslights. Even though inventors had been trying to work out the bugs in electric lights for 50 years, Edison's reputation was such that a syndicate led by the Vanderbilts, J.P. Morgan and others quickly advanced the money to found the Edison Electric Light Co. Edison and his team set to work, and soon found out why so many others had failed to find the right way to make a durable electric light. But they kept working. Reputedly, at one point the inventor was asked how he could continue to try to invent the light bulb when he had failed over 1,000 times. Edison is said to have replied: "I have not failed 1,000 times. I have successfully discovered 1,000 ways that do not work, and I do not need to try them again."

So, if you are engaged in the Hopkins-One design and implementation teams, develop a sense of impatience and urgency, don't be afraid to fail and remember: Perfect is the enemy of the good.


William R. Brody is president of The Johns Hopkins University.


The Gazette | The Johns Hopkins University | Suite 540 | 901 S. Bond St. | Baltimore, MD 21231 | 443-287-9900 |