Blue Sky Research Comes Down To Earth

For your average General Electric Co. researcher accustomed to the quiet, contemplative life in Schenectady, N.Y., being a member of a "one-coffeepot" product-development team is no tea party. The teams unite GE researchers, manufacturers, and marketers in a single location, where they can sweep away obstacles to a new lightbulb or locomotive as they swill hot java.

Trouble is, this exciting opportunity to team up with far-flung comrades might just involve a few months in a Waukesha (Wis.) motel. Lonnie Edelheit, GE's senior vice-president for research and development, says most of the company's researchers have embraced the one-coffeepot concept. In any case, he says, there's no alternative: "Speed means change. Everybody's got to understand that."

That's research and development in the '90s. Every year the belt gets tighter and the pressure for results increases. True, the 900 companies on BUSINESS WEEK's annual R&D survey spent 4% more in 1994 than the same companies did the year before. But adjusted for inflation, their spending rose just 2%--the second year of anemic growth. While spending seems to be picking up again for 1995, no one is counting on a return to the boom years of the 1980s.

Exacerbating the slowdown in corporate R&D is the budget-cutting in Washington. Congress may reduce funding for basic research, in addition to taking aim at such popular items as the Commerce Dept.'s Advanced Technology Program and the Defense Dept.'s Technology Reinvestment Program--both of which are aimed at helping industry do fundamental research that they could not otherwise afford. As corporations began to focus on short-term R&D in the 1990s, they counted more than ever on federal science funding to lay the groundwork for future generations of products. That strategy won't work so well anymore, which worries people like Samuel H. Fuller, Digital Equipment Corp.'s vice-president for corporate research. "Computer and communications technologies could suffer," he says. "My hope is that they don't unnecessarily cut off future innovation."

"MORE JOY." The good news is that companies continue to make progress in getting more bang for the R&D buck through closer collaboration with business units, more openness to external sources of technology, and investments in labor-saving computers and measuring equipment. Pittiglio Rabin Todd & McGrath, a Weston (Mass.) consulting firm, studied seven major electronics companies and found that the share of product revenue coming from new products shot up from one-third in 1992 to 45% last year. And they did it without spending gobs on R&D: The firm says its index of R&D effectiveness for electronics companies has risen one-third since 1992.

After 1994's R&D freeze, there are even signs that companies are regaining interest in research that is aimed at revenue growth rather than just cost-saving. The resurgence of corporate profits has given them some breathing room. Indeed, the R&D trends forecast of the Washington-based Industrial Research Institute "indicated some optimism this year for the first time in about four years," says Executive Director Charles F. Larson. "There's a lot more joy in the ranks these days."

If not joy, at least acceptance. DuPont Co.'s senior vice-president for research, Joseph A. Miller Jr., isn't complaining about the 8% decline in 1994 in DuPont's R&D spending. "I don't expect those constraints to continue, but I think they've been good for us," Miller says. "They've forced us to focus more. We're pressed on relevance."

While wielding the carving knife against programs deemed irrelevant, DuPont has continued building up capabilities in core businesses. It's putting special emphasis on using advanced diagnostic and measurement gear. It's a co-sponsor of the Advanced Photon Source at the Energy Dept.'s Argonne National Laboratories in Argonne, Ill., which will be the source of the world's brightest X-rays when it starts operating next year. The extremely brief X-ray pulses should let DuPont researchers see for the first time how crystals form in a fiber, for example. Says Miller: "We can't wait for that baby to get started."

Not that it takes megadollar equipment to gain insight into a product. Lotus Development Corp. needed nothing more than a little space on some of its customers' hard disks. Some 150 buyers of Lotus' Ami Pro word-processor agreed in 1993 to have their every keystroke captured in a special file, where it could be analyzed later by Lotus programmers. That led to the discovery that customers wasted lots of time clicking "skip" every time their spelling checker hit a strange word. The new version of Ami Pro, named Word Pro, will highlight the suspect words on the screen so there's no need to "skip." The writer clicks only on the words that really might be misspelled.

CONSOLIDATION. Eastman Kodak Co. found its own way to economize--by consolidating 23 different designs for scanners into six. The proliferation was an unintended result of the independence given to Kodak business units. Now, units can still doll up the basic scanner designs for special customers, but they're all starting with the same main components. Says James W. Meyer, Kodak's chief corporate technical officer: "There's no need to go back to earth, air, fire, and water with each one of these."

Nor is there a need to come up with every new product internally. Rockwell International Corp. has more than 200 formal relationships with universities, government labs, and the like, and has gotten some of its best ideas from places like China and Russia. For instance, a Russian institute is helping Rockwell understand how to make a stronger, lighter metal by adding small amounts of lithium and copper to aluminum. Such a metal could go into Rockwell's entry in the competition for a successor to the space shuttle. Rockwell is also teaming up with Boeing Co. and NASA on a project with Tupolev, maker of the Russian supersonic transport. The partners have taken a TU-144 supersonic jet out of mothballs and are flying it to see how air flowing around the plane becomes turbulent at supersonic speeds--vital to the design of a future American supersonic plane.

The theme at every company is to tie R&D more closely to business needs. GE-style teams are one way: A Booz-Allen & Hamilton Inc. study found they dramatically reduce wasteful "churn" in product requirements. To understand what customers might want in an electric car, General Motors Corp. had its research engineers accompany Los Angeles commuters as they drove to work. Hercules Inc., the Wilmington (Del.)-based specialty chemicals company, pulled back on a plan to start a biotechnology business and decided instead to use biotech to improve production of existing products.

At central labs, meanwhile, blue-sky research has clouded over. Says GE's Edelheit: "We used to do a lot more long-range research, but one definition of `long range' is `never.' We're not worrying about, say, the size of the universe, because that's a `never."'

Ditto at Nynex Corp. The Baby Bell's researchers used to do world-class work on making computers recognize handwriting, but that's been canceled. Instead, Nynex Science & Technology President Casimir S. Skrzypczak prefers to brag about how company researchers helped the Nynex phone business reduced the number of phone lines that malfunction in early summer. By analyzing trouble statistics, Nynex researchers realized the malfunctions are most common in areas that have dense phone usage and lots of customers coming and going--so lines of marginal quality are often in service.

The researchers' analysis defied conventional wisdom and is helping Nynex to set priorities about where to upgrade its wiring. "This is not something that's going to win you a Nobel prize, but the impact on the company is sizable," Skrzypczak says. Indeed, Nynex spent about $350 million last year on solving line problems.

Companies still like to brag about how much they spend on R&D--in a way they don't brag about, say, spending on salaries or raw materials or grass seed for headquarters. That's because well-focused R&D generally pays off in long-term revenue growth and profits. On the other hand, more isn't always better.

Just take a look at the uneven record of companies that have led BUSINESS WEEK's rankings of R&D spending per employee since 1984 (table). It's not an issue of long-term vs. short-term research. The key is to make sure that R&D plays its proper role as the engine of a well-balanced corporate machine.


The complete BUSINESS WEEK R&D Scoreboard with information on 900 companies will be available via BUSINESS WEEK Online on the America Online computer service. To order printed copies, send $5 to BUSINESS WEEK R&D Scoreboard, 1221 Avenue of the Americas, 39th floor, New York, N.Y. 10020.

Before it's here, it's on the Bloomberg Terminal.