In Oregon's Farmland, HP Seeks New Way To Milk Its Cash Cow

Investor's Business Daily
Monday July 3, 7:00 pm ET
Ken Spencer Brown

At first glance, Corvallis, Ore., seems an unlikely spot to stage a technology revolution.

With 53,000 people nestled in 14 square miles of valley farmland near Oregon State University, the town is known more for its covered bridges and annual "Red White & Blues" gala than anything high tech.

Don't let the Norman Rockwell scenery fool you. Corvallis has one of the nation's highest concentrations of brainpower, as measured by per-capita patents and Ph.D.s.

The area also is arguably Hewlett-Packard's (NYSE:HPQ - News) most important site outside Silicon Valley. Engineers in Corvallis developed such breakthroughs as the 12c scientific calculator, the palmtop HP 95LX and the consumer inkjet printer, now HP's most profitable business.

Reinvent The Printer Wheel

But household printers sales have declined recently. So, with Corvallis once again taking center stage, HP is preparing to reinvent its printing operations to let it expand markets and slash costs. (See related stories on page A4.)

"We need to look at this business very differently," said Vyomesh Joshi, head of HP's printer unit, who's putting the finishing touches on a top-to-bottom overhaul of the business.

Part technology switch, part business restructuring, the $1.4 billion revamp will impact a broad swath of HP -- from its corporate offices in Palo Alto, Calif., to production lines around the world.

Think of the 4,000-person Corvallis campus as ground zero. Deep inside a labyrinth of hallways, cubicles and labs, HP workers are retooling their business around what officials call "scalable printing."

The name reflects the notion that the underlying technology can scale up or down to fit a range of products. HP is applying the idea to everything from home photo printers to commercial publishing.

The hope is that scalable printing will usher in a new wave of growth for HP. Joshi quietly launched the effort four years ago on a hunch that the firm's double-digit inkjet sales growth wouldn't last forever.

"The hardest thing was realizing we had to do it," Joshi said. "It's very easy when you're in trouble and it's obvious to everybody that you need to transform. To do that when you're making 13% to 15% operating profit is a very hard thing."

Expand Markets, Cut Costs

The changeover aims to cut costs in HP's existing markets. At the same time, it should let HP launch new products in untapped regions and market segments.

Engineers in Corvallis call scalable printing the biggest breakthrough since the advent of inkjet printing 20 years ago.

In its simplest terms, it's a process that forges the print head out of a single piece of silicon -- rather than as separate pieces to be aligned and glued together later.

Scalable printing replaces mechanical processes with a lithographic one, similar to the way Intel (NasdaqNM:INTC - News) makes computer chips.

That lets HP fit more nozzles on the same-size print head. The result: sharper pictures, faster prints and more reliable printers.

Scalable printing's other big advantage is what it does for HP's business. For the first time, the company will be able to make a wide range of printers on the same manufacturing line.

Case in point: HP's new retail photo printing kiosks, the firm's first stab at the retail photofinishing market, use the same technology as HP's just-launched $129 Photosmart D6160 home printer.

"Yeah, it is a big deal," said Jim Forrest, an analyst with Lyra Research. The technology should tighten HP's grip on its core home inkjet market and broaden its reach into new segments, he says.

Cindy Shaw, an analyst with Moors & Cabot, agreed in a recent report. "We expect (scalable printing) to hold an impressive speed lead over other inkjet technology for years to come," she wrote.

Though the idea makes sense on paper, it wasn't easy getting everyone in the company to agree on details, says Val Bach, a fluidics engineer who oversees several of HP's silicon-component factories.

HP's decentralized culture left the printer unit divided among product fiefdoms. Each had separate development efforts, product recipes and manufacturing processes.

With marching orders from Palo Alto, Bach and others figured that ink was a natural focal point. Ink particles must match the tiny pathways in a printer's print heads for good performance.

Instead of having each product manager list a set of requirements for the fluidics group, Bach got everyone to agree on one basic design for the printer nozzles.

It was the first time some engineers in Corvallis spoke directly with their counterparts in San Diego and other HP hubs.

She opened the session with sobering news. "I said, 'Hey, we're at the bottom of the food chain now. We're not going to get any more money. We're not going to get any more resources,'" Bach recalled.

In other words, looming cost cuts meant the group would have to eliminate redundant efforts.

That created a natural tension. The resulting compromise wasn't ideal for every product line. But everyone knew it would be better for the overall printer unit, Bach says.

Today the printer business resembles a hub-and-spoke network: diverse products connected by common technologies and processes.

The change started on the far edge of the Corvallis campus, in the print head manufacturing plant. A big part of the $1.4 billion investment paid for the chipmaking gear inside, which will make HP's new all-silicon print heads.

Manufacturing the print heads will now be like building a chip. Only instead of electrical pathways, the equipment will etch out the heating element -- tiny pathways and thousands of separate nozzles that together make up the most important part of the printer.

Though scalable printing will involve a more complex lithographic process, HP's nozzles aren't yet at the tiny scale of today's most advanced chips. So the company can use manufacturing gear several generations old. It'll be cheaper and more reliable than the newest gear used by most chipmakers.

New nozzles necessitate new ink. HP engineers had to whip up a liquid to fit the new nozzles, flow through plastic tubing without clogging and dry fast on a range of media.

Finally, HP had to redesign the printer itself: the plastic shell that holds print heads, ink tanks, paper feeder and everything in between.

These changes pose some challenges. Most of HP's printer profits come from selling ink cartridges, not the printers themselves. But the new designs will make it much easier for third parties to sell knock-off cartridges. (See sidebar this page).

On Corvallis' sprawling factory floor, HP builds prototype and early-run products to work out manufacturing kinks. It then hands off large-scale production to its plants in Singapore, Ireland and Puerto Rico.

As part of the overhaul, engineering project manager Jim Sizemore is retooling the facility for a faster product ramp-up. Similar efforts are under way elsewhere at HP.

Anchoring the effort is $20 million worth of new factory equipment. The machines, which can be quickly reprogrammed for different sets of products, are built for smaller, more flexible manufacturing runs.

Sizemore eschewed fully enclosed clean rooms in favor of new setups that create self-contained mini clean rooms. They blow a stream of air down and under a hanging plastic covering, like those used in some grocery stores to keep refrigerated shelves cool. Thiskeeps particles from tainting the gear.

In this setup, workers don't have to don the "bunny suit" used in most clean rooms. This saves time and makes the factory flexible.

With more HP products based on a common underlying technology, Sizemore expects to crank out product lines mere months after they're conceived. That's much faster than the three years or so it can take now.

"The core technology is what takes forever to develop," he said. "That's changing." It has to, if Joshi hopes to reform the economics of the printer business.

In most industries, the real innovation happens in high-end products. Those new ideas then trickle down into cheaper products over time.

Conventional wisdom says some customers are willing to pay more for the latest-and-greatest gadgetry. That premium helps cover the research and development costs.

Joshi -- an avid reader of management theorists such as Geoffrey Moore and Peter Drucker -- wants to turn that concept on its head.

He says the biggest innovations should be on mass-produced items. There, HP can perfect the technology and get a handle on the manufacturing process."A lot of firms talk about this approach, but nobody has done it," he said. "Until now."