Intel's New CEO Faces Broad Challenges

When Intel holds its stockholder's meeting on Wednesday in Santa Clara, Calif., the company will host a changing of the guard, as Craig Barrett passes his CEO baton over to president and chief operating officer Paul Otellini. More than a ceremonial shift in command, the handover of day-to-day control to Otellini—Barrett will move up to the post of chairman—will mark the first time in its 36-year history that Intel will be run by someone who's not a techie.

Otellini, 54, holds a bachelor's in economics from the University of San Francisco and an MBA from the University of California at Berkeley. Despite that background, and the fact that he began his career at Intel in 1974 in finance, he sees himself as no less attuned to the company's technologies than his legendary predecessors Robert Noyce, Gordon Moore, Andy Grove, and Barrett.

"I consider myself a product guy," Otellini said in an interview. "I've had my fingerprints on products for 20 years. I ran the microprocessor division for a decade, during the 486 [and] Pentium days. I get into products. I like them, I use them. Sometimes, not having a thorough understanding of an electron structure allows you to ask the really hard questions."

Indeed, Otellini seems to have turned his status as a non-engineer who's intensely interested in technology to his advantage. "This is the first technically savvy, but marketing-trained, set of ears to take over the world's largest semiconductor company," said Rick Doherty, research director of The Envisioneering Group, a technology consulting firm in Seaford, N.Y.

Otellini has also cultivated a reputation within the company as a straight shooter. "Otellini was one of those people who was willing to say when he thought," said a longtime Intel engineer, who requested anonymity. "I've heard him speak up in meetings, where it took real guts."

Most immediately, Otellini may be looking to put some daylight between Intel and the debacles of 2004, when the semiconductor giant faced the glare of negative publicity for a string of glitches. Theses included the discovery of defects in its Grantsdale chipset, the scrapping of the planned Tejas processor because it ran too hot, and delays to its Dothan Pentium M mobile processor. Such troubles prompted Barrett to collectively take Intel's engineering staff to the woodshed, via a companywide memo which chastised employees for the spate of glitches and exhorted them to fix the problems.

The memo, according to the longtime Intel engineer, was widely resented by inside the company by engineers, who felt they were publicly hung out to dry. Yet those engineers are now embracing Otellini, who has gone out of this way to build bridges to them, conducting weekly one-on-one briefings where technical personnel can freely speak their minds without a manager in the room. "I've done over 200 over the last couple of years," Otellini said. "You learn a lot of things. We keep moving the [product-development] horizon out, so now we're looking at things that are three or four years away, as opposed to one or two years."

Such a future-directed focus will likely become a hallmark of the Otellini regime, as he formulates a strategy for building up some of Intel's non-processor businesses—particularly, communications—that haven't been achieved their full potential.

Most immediately, Otellini is focused on building up a head of stem for Intel's shift to dual-core processors. That will enable it to level the playing field with competitor AMD, which is widely seen as having beaten Intel to the punch on the previous round of innovation, in 64-bit processors, with its Opteron and Athlon 64 devices.

Otellini has championed Intel's move to dual-core CPUs, famously hailing it as a "hard right-hand turn" in the company's processor roadmap, when he announced Intel's plan to move toward the technology in May 2004. He's ramped the effort up to full bore in less than a year. This April, Otellini pegged the total number of dual-core efforts currently underway at Intel at 17. Two dual-core desktop Pentium processors have already begun shipping. Fifteen other processors, encompassing a mix of desktop, server, and mobile CPUs, are on the way.

Over at AMD, which has several dual-core versions of Opteron and Athlon already shipping and more in the pipeline, company chairman Hector Ruiz said he's taking Intel's efforts seriously. "It's a company with a lot of bench strength, but it's not [tied to] any one person," AMD chairman Hector Ruiz said in an interview. "We will continue to respect them."

But Otellini appears poised to throw an additional wrinkle into the battle between the two chip vendors. Namely, he wants to change the way Intel markets its processors. Rather than trumpet a processor's internal characteristics such as clock speed or its number of cores, Otellini has spent the past year putting in place a strategy called "platformization." Under this approach, Intel has sought to emphasize non-processor technologies, such as faster memories and I/O interconnects. The idea is that such features will drive overall increases in system performance beyond what can be accomplished simply by boosting clock speed. "Intel's probably going to go in a platform-centric direction in both communications and consumer," analyst Doherty predicted. "Not just processors, but platforms, meaning more than one chip, because there's a stiff headwind of competition in both sector, from companies like Broadcom and Texas Instruments, respectively."

To drive its platformization plan forward, Intel in January implemented a broad corporate reorganization, realigning the company into five divisions: the Mobility Group, which will develop notebook PCs and handhelds; the Digital Enterprise Group, which will develop computing and communications platforms; the Digital Home Group, focused on entertainment and media computing; a channel group, to serve resellers and systems builders, and the Digital Health Group, which will explore opportunities in health-care.

"What Otellini has to do is to figure out how to drive those businesses forward," said the Intel engineer, who requested anonymity. "He needs things other than CPUs."

Intel is also hanging a big part of its future on its ability to grow in emerging countries, which promise more robust rates of growth than the technologically mature domestic arena. After Intel's third year of double-digit revenue growth, Otellini says his biggest commitment remains the delivery of bottom-line results. "It's all about growth," Otellini said. "We want to continue to grow the business and deliver new capabilities and exciting new technologies."

As for outgoing CEO Barrett, he won't ride off completely into the sunset, although he does have some leisure plans. "In two weeks and one day I will be standing in a Montana river with a fly rod in my hand," Barrett said on May 5 at Intel's annual meeting for financial analysts in New York City. When he returns, he will oversee Intel's board of directors, in his role as chairman. He'll also focus on his pet project, Intel's "Teach to the Future" initiative, which seeks to promote the use of technology in the classroom.