How KLA-Tencor yields innovation
reed-electronics.com
Like any other good CEO, Ken Schroeder of KLA-Tencor spends a lot of time thinking about his employees. But unlike most other CEOs, he spends a lot of time thinking about what they know and what they still need to learn.
"We've been recognized as No. 1 in Silicon Valley for all kinds of training. We put a lot of time and attention into it. We emphasize unique training that makes us different," says Schroeder, 59, who has led the semiconductor process control and yield management company since 1999.
The emphasis on learning apparently works. For one thing, it contributes to employee turnover of less than 5 percent on a global basis. And the learned staff's outpouring of innovative products has kept KLA-Tencor on the leading edge for more than two decades. The 28-year-old company has more than a 50 percent share of the process control market, with 70 percent in some segments. Its closest rival, Applied Materials, has about 10 percent. KLA-Tencor sells more than two dozen products for wafer inspection, mask inspection and metrology. It is No. 1 in most segments and No. 2 in the others. More important, analysts say, it is the only vendor that offers a portfolio of integrated tools that analyze defects data and helps users make corrective decisions. The arcane technology improves fab yields and allows chip makers to ramp up quickly in each new process as they follow Moore's Law.
KLA-Tencor, which had net income of $244 million on revenue of $1.5 billion in fiscal 2004, routinely has the highest margins among semiconductor equipment companies (see "KLA-Tencor At-a-Glance," below). Since 1992 it had just one losing quarter—when it took a restructuring charge in 1998. Except for an occasional complaint about its ultraconservative guidance regarding quarterly orders, stock analysts are enthusiastic. They say that, compared to rivals, it gets more bang for its R&D buck (15 to 20 percent of revenue), has sounder fiscal management and is better at anticipating downturns.
"KLA-Tencor is one of the best-runsemiconductor equipment companies in the world," says Jay Deahna, an equity analyst at JPMorgan. "Its top-line growth story and competitive positioning are better than most."
What's behind the success? Central to its strategy is a collaborative, creative environment that fosters innovation and emphasizes learning. KLA-Tencor ranked fifth overall and first in Silicon Valley on Training magazine's 2004 Top 100 list. (Intel was second in Silicon Valley, at No. 17.) Schroeder and his executives expend large amounts of personal energy and corporate resources on teaching and mentoring employees in management, technology and innovation practices. "Training is one way to get consistent methodology throughout our 17 divisions," Schroeder explains.
Whether the training is in technology, innovation or management, traits that set KLA-Tencor apart are extensive advance preparation, context-specific learning and rigorous follow-up, says Dev Ogle, senior consulting partner at Ken Blanchard Co., who has worked with the company for years. "At every level in the organization, you can see people applying what they learn in the classroom back in the work environment with the help of coaching, mentoring and feedback," he observes. "As a result, the company gets consistent models for leadership at the supervisory, midmanagement and senior levels."
There's a message for others, Ogle argues. "You often hear the right words about training from CEOs but no passion for it. These guys at KLA-Tencor have the passion for it." Schroeder agrees: "There's a lot of lip service about training in Silicon Valley. But I don't think most CEOs understand the power of training and the impact it can have on their companies."
From eyeballs to expert systems
In the 1970s, when integrated circuits were still measured in microns and wafer defects were eyeballed through a microscope, Ken Levy had a passion for improving manufacturing yields. Levy, now 61 and chairman, founded KLA Instruments in 1976, and in 1978 the company offered the first automated inspection system, which reduced the time required to inspect photomasks from eight hours to 15 minutes. A rival, Tencor Instruments, was also founded in 1976. The two became the leaders in process control and merged in 1997, in what analysts still call one of the electronics industry's most successful mergers.
Rick Wallace, executive vice president at KLA-Tencor, was working at Cypress Semiconductor in the early 1980s when it bought one of the first KLA inspection tools. Wallace, who began his career in paper products at Procter & Gamble, recalls, "Compared to what the paper industry was doing, there was no sense of process control automation in semiconductor manufacturing. Every other industry had demonstrated that process control was necessary for that industry to mature."
Cypress was an early adopter, but not every chip maker could fathom why it would want an inspection machine itself or what to do with it, Wallace recalls. But before long, chips were measured in submicrons and eyeball and microscope could no longer do the job. As geometries shrank, automated process control grew in importance and inspection equipment became a bigger portion of a chip maker's total equipment expenditure.
"KLA-Tencor is one of the best-run semiconductor equipment companies in the world."
—Jay Deahna, JPMorgan
The market didn't spring up overnight, says Risto Puhakka, vice president of VLSI Research, a market research firm. "It took years for the industry to understand the importance of yield. Process diagnostics was a very small market all through the 1980s, just an afterthought to lithography."
As geometries grew smaller in the 1990s and KLA-Tencor stepped up efforts to educate the industry (its trade show seminars are well known), the market began to grow. Once it had a family of inspection machines, KLA-Tencor began to develop software to help users analyze yields and take corrective action to improve them, says Bob Johnson, a market analyst at Gartner Dataquest. "Once they had the inspection capabilities, they figured out what to do with them. That's how yield management evolved."
Puhakka believes that software is the key to KLA-Tencor's dominance. "There have been many efforts to take on KLA-Tencor in certain areas. Companies can offer equally capable inspection systems, but the differentiation for KLA-Tencor is the software. That's where you turn data into action," he says. "Software has never been a classic strength of any other company in the equipment business." About 60 percent of KLA-Tencor engineers work in software, and a large amount of its IP is software, according to Wallace.
Brett Hoddess, an equity analyst at Merrill Lynch, attributes KLA-Tencor's success to two factors: the chip industry's need for more, not less, yield management as geometries get smaller and the fact that KLA-Tencor offers the broadest portfolio of products. "It has integrated the data from products into expert systems. Its software ties all the data together from various steps and makes sense of it for the user. No one else comes even close to having a complete product set like this."
A culture of innovation
The products, including a services unit and the expert systems, and the need to stay two steps ahead of the chip industry's march toward smaller geometries (KLA-Tencor currently is developing products for 65 nanometers, generating ideas for 45 nm and researching 32 nm) are the innovative hallmarks of the company. But the culture that produces this innovation is where the management lessons are found.
What the company manages better than most is innovation. It starts with hiring—especially from the half dozen universities it targets—but doesn't end there. "You can't make a noninnovative person innovative," Schroeder asserts. "But you can take an innovative person and make that person more so." Not innovative for innovation's sake but for the sake of better products.
Ben Tsai, the chief technology officer, says a big part of his job is to teach, manage and nurture innovation. But it starts at the top, he says, where Schroeder, Levy and other executives "define what innovation means for our company." He says most articles and books on innovation are too ambiguous or general. "We define what innovation means for us. We want to solve problems for our customers in a better way."
Then with the help of the corporate learning center, the top executives spread this understanding of KLA-specific innovation throughout the organization. With a staff of 17—and also tapping experts across the company—the learning center has been an integral part of the innovation effort since it launched, in 1996. "A culture of learning leads to a culture of innovation," says Brent Bloom, senior director of corporate learning. "Ken Schroeder drills that idea home over and over again."
KLA-Tencor views the learning center as being so strategic that executives are reluctant to divulge many details about it. They speak in general terms and give only a high-level, broad sketch of its mission. "We consider a lot of the details proprietary," admits Tsai.
Even so, they offer some insight into how it works. The learning center, located at the San Jose headquarters, focuses on product innovation, teamwork and management. A separate facility in nearby Livermore, Calif., has a staff of about 50 that teaches pure technical skills to employees and customers. The company finds that even nontechnical learning is best accomplished in the context in which learners must apply the lessons, so innovation and management are often team-taught by one person skilled in the management process and one skilled in the technical area in which the process is to be applied. Even process experts come from the rank and file. Bloom started as a manufacturing engineer and worked in field service and technical support before spending the last three years in corporate learning.
Revolution and evolution
According to Tsai, KLA-Tencor promotes innovation on two levels: organizationally, through systems and procedures, and individually, through tools and learning episodes for product teams. Over the years, it has established a consistent yet flexible approach to improving old products and generating new ones. All this innovation is rooted in a rigorous understanding of customer needs and problems—especially problems customers may not even know they have (see the sidebar, "Innovation Is Where You Find It," below).
Tsai recalls Levy telling him years ago, "Marketing is too important to be left to marketing people," by which he meant that product development engineers must be involved in defining new markets. Yet as the company grew, it became less feasible for each engineer to meet customers. Now twice a year for each product line—quarterly in some cases—teams of key engineers and marketing staff members visit customers, collect information and bring it back to share and interpret with the entire product team.
KLA-Tencor also has procedures, which it doesn't divulge, for monitoring competitors. "Organizationally, we have created special processes to capture key learning from major customers and competitors and make them available for each product line," says Tsai. The learning center administers a database and disseminates information. "In partnership with corporate learning, we want to drive this sense of addressing customers' needs down to the rank-and-file engineers," says Tsai.
"You can't make a noninnovative person innovative. But you can take an innovative person and make that person more so."
—Ken Schroeder, CEO, KLA-Tencor
"We're leaders because we create markets," says Wallace, "and we create markets because we recognize customer needs." When it comes to market strategy, KLA-Tencor is too far out on the leading edge to be able to ask customers what products they need. "You can't survey for innovation," Wallace argues. "You can't ask customers what they need. You can ask what their problems are." And then you can spend time with customers at their fabs to better understand their problems.
To help individual engineers, learning center staff members track commercial software and methodologies that can assist innovation. "We select a small set of them we believe is most pertinent to our needs," says Tsai. The engineers incorporate the ones they select into their own methodologies. "Depending on your situation, your innovative approach has to be tailored," says Bloom. "We don't want to lock engineers into one specific method." He says some tools support evolutionary innovation from existing products and others foment revolutionary innovation—products that are entirely new and different.
Control through training
For revolutionary innovation, KLA-Tencor has adapted creative problem-solving techniques developed by Ideo, a design firm. Ideo is widely known for customer-centered problem solving, human-friendly design, rapid prototyping and brainstorming (see "Innovation Quotient," December 2000). KLA-Tencor does its own variation of Ideo brainstorming, Bloom says.
Engineers ask the learning center for help at the beginning of a product development project or when creativity stalls and they can't find a way forward. Executives—especially Tsai and Schroeder—keep close tabs on product development and often urge a team to seek the learning center's help, Bloom says. Once a team describes its needs, the learning center designs an appropriate course. "We try as much as possible to train team members together," says Bloom, "and to train them on the team's specific problem." By doing so, the team often generates ideas during training, he adds. "We always try to tie our training to a real situation."
A training session can last anywhere from a couple of hours to a couple of days, with regular feedback and follow-up afterward, says Bloom. "An average company spends about five days per person per year on training," he says. "We did that much just in the first few months of our mini-MBA program."
He's referring to KLA-Tencor's internal executive development program. It used to send managers to summer school at MBA programs run by various Silicon Valley universities, says Tsai. "We found they were not as effective as they could be, because they didn't teach about our unique culture and the way we do things." So, a few years ago—using its own innovation techniques—a group led by Tsai developed an executive training course that uses case studies from the company's own history and builds much of the learning around current strategic planning issues.
Is there any downside to all the learning activities? Analyst Hoddess says he sometimes senses a "not invented here" mentality and points out that the company tends to grow organically rather than through acquisition, even when appropriate acquisitions have been available.
Countering that assertion is KLA-Tencor's recent acquisition of the wafer inspection systems business of Inspex, a U.S. subsidiary of the Japanese firm Hamamatsu. And Schroeder points out that KLA-Tencor has a $50 million VC fund that invests in startups of strategic interest, including some in life sciences, which he believes is a potential growth area. "Within a year, you'll see us getting into some new market outside our current semiconductor space," he predicts.
Without the emphasis on learning, which is shared by Chairman Levy, Schroeder can't imagine how he would do his job. He views education efforts as the way to extend his influence. "As CEO, I want to control how decisions are made throughout the company," he says. "The only way I know how to do that is to take control of training. There is no way to manage a company of this size without common processes and methodologies. And the best way of spreading those throughout the company is through training." |
