Get It Right, Not Just "Right Enough"
Jan 24, 2019
The people who participate in wildly successful initiatives search for answers and learn incessantly until they break the code. Remember when your primary school teacher gave you credit in arithmetic for following the right steps, even if you arrived at the wrong answer? That practice still exists in organizations today. If we follow all the written and unwritten rules, check all the boxes, and tag all the bases, we get credit for a good effort, even when the results are not what we had hoped for.
I'm not making this up. Just look at ISO-9000. This very important certification does not demand that an organization operate effectively or produce good results. It requires only that the organization have a specified process and actually follow that process—regardless of how well it works. Wildly successful initiatives don't make this mistake.
Champions of wildly successful initiatives insist on finding the right answers to the hard questions. They place priorities where they belong, recognizing that implementing the right solution is ten times more valuable than a solution that is only "good enough."
Consider the example of a successful executive in one of the world's most renowned consumer goods companies. Let's call it Unigamble. This executive believed that his company could gain a great deal of benefit from improving manufacturing reliability—the percentage of time that the equipment was operating as it should. At the time, despite successful initiatives to measure reliability consistently and improve performance, many plants were stalled at 65 to 75%. He asked his organization to achieve at least 85%, and he required all new operations to reach that level in the second month. He set this extremely aggressive goal, then he gave his people the latitude to go meet it.
A pair of engineers in the diaper business took the challenge personally. They were well aware of the existing research on reliability, but the accepted theories did not jibe with their everyday experience on the manufacturing lines. Something was wrong. Operations had hit a plateau of improvement. No matter how hard the engineers tried, they could only get the manufacturing lines to be up three-quarters of the time at most.
Working from the middle of a very large company, the engineers had two problems. First, the predictive simulations that reliability experts provided just weren't working for them but for the engineers to continue operating intuitively wasn't going to get them to the next plateau. Second, in the absence of a good model, the company was awash in competing ideas about the right approach. Since there were no hard numbers on which to base these proposals, politics took over. In the engineers' view, a negotiated compromise was an unacceptable substitute for a valid, mathematical proof.
The engineers knew that their theories were incomplete, but one thing they did have was a wealth of data on machine reliability. Every stoppage was recorded, with detailed coding to indicate why it had taken place. They had thousands of recorded incidents, but no effective framework for analyzing them.
The engineers set out to try to make sense of the extensive data they had gathered in the plant. Their explorations ultimately led them to Los Alamos National Laboratories, where scientists were working on the predictability of nuclear weapons systems for the U.S. military. Lacking a real-world operations environment in which to test their mathematical model, however, they had never demonstrated its efficacy.
Through what one engineer calls a "courageous process," the two organizations joined forces and pooled their intellectual property—the data from one side and the mathematical simulation from the other. Over the following year, the team developed a categorically new reliability simulation—one that actually predicted failure in a manufacturing environment. This model found that machine stoppages interact. Even the smallest hiccup could start a chain reaction that would breed more stoppages. The model predicted that eliminating every small stoppage, instead of concentrating on eliminating the catastrophic failures, could substantially improve performance.
Why was getting it right so important? For the first time, the engineers were able to make accurate predictions of the value of particular improvements so that operating executives could make good investment decisions. This opened the door to reliability improvements that dropped more than a billion dollars to the company's bottom line.
One engineer remarked, "The first thing we did was to clarify the business impact of running the manufacturing process well. It sounds easy, but sometimes the truth gets lost in the accounting data." It seems that the practice of management has ramped up its emphasis on process over the past three decades, and this tidal wave of focus has almost swamped the other fundamental foundation—getting the right answer.
Getting it right doesn't stop at the intellectual solution. If the right answer can't be communicated to the people who must implement it, it has little value. As a result, the word simple finds its way into every story of working wonders. Unigamble's engineers used a homey little analogy to make their new approach easy for everyone, from their management colleagues to the machine operators in the factory, to understand. They talked about the family vacation. They gave operators a new view of manufacturing downtime by likening it to all those little stops that families make on a road trip.
One engineer explained, "The reliability of repairable production systems is like taking a cross-country trip with your family and predicting when you will get to Grandma's. People want to know the reliability of your car, the probability of an engine failure or a flat tire, and the chance you will run into construction. When you think about it, the number one failure mode with a car full of kids is the bathroom and gas stops. Conventional reliability science treats all those bathroom stops as independent events. They aren't. When we stop to get gas, if we all go to the bathroom, we don't have to stop again in 20 minutes. That way, you take advantage of the stop to be resetting as good as new. It turns out that's radically different from resetting a piece of equipment as good as old."
In another example, Intel's future-of-technology research organization—which is staffed with anthropologists and social scientists, not engineers—recently had an epiphany. A team that was nominally studying trends in home music usage found itself pondering a very different concern: how families were going to take care of aging parents. The team concluded that technologies which were already in the home, such as televisions and telephones, would form the foundation of a personal care network. If Mr. Johnson needs to be reminded to take his medicine, that message should come to him through the telephone, not through a personal computer or PDA. Why? Because Mr. Johnson already knows how to use the telephone. With this simple interface, he will actually get the message and take his medicine. As a result, the more complex technology underneath will become useful.
Leaders of wildly successful initiatives recognize that simple is the doorway to usable. And if a "solution" is not usable, it just isn't right.
Adapted with permission of the publisher from target=_blank>Spiral Up by Jane C. Linder. Copyright 2007, Jane C. Linder. Published by AMACOM, a division of American Management Association. For information about other AMACOM books, visit target=_blank>www.amanet.org/books.