GUEST EDITORIAL

The Path to World Class

 

Professor Wayne Wolf, Princeton University, IEEE Fellow

Lots of companies want to be world class. Lots of people want to work at a world class company. Being part of a world class team makes you feel good. Being good at what you do brings pride of accomplishment. But what does ¡°world class¡± mean?  And how do you get there?

First, I think it is important to remember that ¡°world class¡± truly means being one of the best in the world at what you do.  Many people in the U.S. and elsewhere like to use the phrase ¡°world class¡± but they never compare themselves to anyone else.  Being world class means understanding your competition. You have to know who else does what you do. You have to understand what makes them excellent. And you have to understand how you stack up against your competition.

If you want to be world class, then you want others---your customers, your competitors---to compare your work to the other leaders in the field. You want them to put your work side-by-side with the best people in the world and have them decide that your results deserve their place alongside that global competition.

So ruthless self-criticism is an essential characteristic of people who strive toward excellence. As Andy Grove, the former chairman of Intel often said, ¡°Only the paranoid survive.¡± That is, people who are complacent and who don¡¯t continue to compare themselves to their competition will eventually be surpassed by their competition. A common Silicon Valley phrase puts the same thought another way: ¡°If you don¡¯t eat your lunch, one of your competitors will eat it for you.¡±

Clearly, world class status requires obsessive focus on quality and a relentless drive to improve the quality of your products.  But what defines quality? Ultimately, your customer defines what quality means in any marketplace. The customer defines the features that the product must have and what attributes of the product are important. For example, cell phone customers have decided that the added functions provided by digital transmission are important elements of quality. A few years ago, when the cell phone industry moved from analog to digital, Motorola didn¡¯t pay enough attention to its customers. They thought that the small size of their analog phones was what customers wanted. Unfortunately for them, they were wrong and they lost market share as a result. They had to scramble to catch up to their competitors and provide their customers with what they wanted.  As the saying goes, ¡°The customer is always right.¡±

Giving customers what they want requires more than just listening. Even after you know what is important to your customers, you have to figure out how to give those important attributes to them in the products you create. Just dreaming that you make great products doesn¡¯t mean that excellent products pop off the end of your assembly line. It takes sweat and careful planning to turn a customer¡¯s desires into objects of desire for your customers.

It¡¯s also important to learn one market well and stick to it.  It takes time to understand what customers want. It takes even long to figure out how to give them products that satisfy their desires. If you constantly move from one market to another, trying for a quick win, then you will never learn enough about any one market to understand what quality means or to develop a high-quality solution to your customer¡¯s needs.

One of the keys to turning concepts into high-quality products is methodology. Design methodologies are critical in both hardware and software design. A methodology is a process that an organization uses to create a certain type of system. Different types of systems may require somewhat different methodologies, but all methodologies share certain characteristics. First, a methodology describes the design procedure in terms of small, well understood, repeatable steps.  Just saying ¡°create the system architecture¡± isn¡¯t a step. You have to break that goal down into many smaller steps that can be performed by team members. Second, they rely on measurement and analysis as much as possible. For example, if your chip has to run at a certain speed, you don¡¯t just hope that your design will be fast enough, you measure its performance.  Third, the methodology should define a set of decision points based upon that data. If you measure the performance of your chip and find it needs to be faster, then the methodology should tell you what to do. 

The methodology probably won¡¯t be fully automatic---it can be hard to automate some important steps in design work.  But the methodology should be documented and followed by everyone. If a tool can perform a certain step, then that¡¯s great.  The make utility from Unix is one of the most powerful tools in any designer¡¯s arsenal. Make allows you to write down a procedure to make something as a sequence of steps.  The tool makes sure that all the steps are performed to create a complete working version of the artifact.

As you develop your methodology, you will create unique capabilities that will help you create innovative products. For example, you may learn how to design a particular component particularly well. You could design a high-quality user interface that makes your product easier to use. Or you could develop innovative ways to save power and improve battery life. Those unique capabilities won¡¯t come easy---as Thomas Edison said, ¡°Genius is 1 percent inspiration and 99 percent perspiration.¡± But innovative features can help make your customers loyal to your product. And a well-designed methodology will help you get the most mileage from any innovations you create. You can often use an innovation in several different products or several members of a product family. Not only does reuse help you leverage your investment in innovation, it helps identify your products with particular attributes of quality.

In order to innovate yourself, you have to be willing to learn from others.  You must learn the basic techniques of your field and study how other people have solved hard problems in your area.  As Isaac Newton said, ¡°If I see farther it is because I stand on the shoulders of giants.¡± Reading books is one good way to learn about your field. Magazines and journals, like those from the IEEE and ACM, provide even more up-to-date information. Just as medical doctors read research articles to keep up with the latest treatments, hardware and software designers must also constantly read in order to know about new design techniques. Going to conferences is also a great way to learn. Conference talks are usually the most up-to-the-minute information that you can get in a technical subject. You could read the conference proceedings, but that isn¡¯t the same as hearing the person speak and asking your own questions.  International conferences that attract the world¡¯s leading researchers and practitioners are must-attend events for expert designers.

Of course, learning isn¡¯t the same as stealing. Merely copying someone else¡¯s work doesn¡¯t really teach you anything. Equally important, copying your competitor puts you behind your competitor. As you copy what they are selling now, they are busy working on something better.  Learning and innovating is ultimately the only way to survive.

And finally, working for the best boss possible is key to being a world leader in your field. Good bosses can recognize good work and bad work; they also know how to get the best work out of their people.  I have had the good fortune to work for some of the best people in the world---the inventor of process simulation, the inventor of the CCD, a Nobel laureate.  They supported me and taught me how to do better work.

Surprisingly, one of the most important markers of an excellent boss is that he wants you to tell him when he is doing something dumb.  Bosses who don¡¯t know what they are doing often don¡¯t want to be overshadowed by their subordinates, so they don¡¯t want to hear bad news.  A strong boss, on the other hand, wants to know when things are going poorly as well as when they are going well. In Silicon Valley, telling your boss that he is wrong is a time-honored tradition.  Great bosses want as much feedback as possible to be sure that they stay on the right road.

A great boss who leads a great team of technical innovators is bound to do well in the marketplace. Skill, knowledge, and patience are all important attributes of people who do good work. And people who consistently do excellent work will be rewarded by their customers and their competitors with the recognition of world class status.