While working at Westinghouse, Lemuel Tarshis had a vision that sent him to Stanford University for a doctorate in material science: solid state electronics and semiconductors were going to revolutionize the world. His educational goal was to better understand the coming change. “I saw what was happening,” he says, “and I wanted to be in the middle of it.”
During his career Tarshis has moved through a series of positions at large and small companies where he has been either a leader or member of a team of innovators who developed new products. In his 12-year tenure at General Electric, he helped to create the plastic flatiron, the consumer smoke alarm, and a number of electronic and electric flashbulb designs for different products. While at General Instrument Corporation (now Motorola Broadband Communications), he was principally responsible for the use of fiber optics in cable television.
Because of his innovation experience, Tarshis was invited to the Stevens Institute of Technology in 1991 to help found and build the Howe School Alliance for Technology Management (HSATM), a consortium of companies focused on innovation. Tarshis joined because of his desire to bring together the university and New Jersey companies around the subject of innovation.
Tarshis will serve as moderator at a conference on “Managing Risk and Uncertainty in Pursuit of Innovation,” on Wednesday, June 4, at 8:45 a.m. at the Babbio Center at Stevens in Hoboken. Cost: $300. Speakers will be Aaron Shenhar, professor of management and founder of the project management program at Stevens; Richard Westerfer, founder and president of R&R Consulting who, while at WorldGate Communications, led the development and manufacturing of the Ojo VideoPhone; Donald Stanford, an adjunct professor of computer science at Brown University; Jeff Krull, product development consultant and former vice president of product development for Shure; and Michael zur Muehlen, assistant professor of information systems at the Howe School of Technology. To register, contact Sharen Glennon at 201-216-5381 or SharenGlennon @stevens.edu.
Because innovation by definition creates something radically different from everything you already know, it brings with it a lot of uncertainty and risk. As managers try to balance the short and long- term needs of a company, the long timeline of innovation is likely to conflict with the short-term pressure to meet the current quarter’s numbers. As a result, managers often use risk as an excuse for not supporting innovation.
To effectively innovate companies need ways to manage risk. As Krull wrote in an abstract for the conference, “One of the great unspoken risks to innovation comes from internal factions that do not fully understand the intent and role of innovation within the organization.”
On the other hand, warns Tarshis, “There is no real silver bullet. You can’t go out and say, ‘I have the answer, we will wave a wand and say there will be no risk.’” Risk is inevitable, and any company that wants to innovate and move forward must learn to manage it. Tarshis shares a few ideas about how to do so:
Don’t use risk as a reason not to innovate — you may lose out to the competition. When big companies are so risk-averse that they avoid innovation, smaller companies will sideswipe them as they make product breakthroughs. Big companies that avoid risk are then likely to find their competitors taking over part of their market.
Back in the mid-1980s, General Instrument, where Tarshis used to work, was the number one company in the lottery business. Because the company had a big infrastructure that was set in its ways, it did not look at how it could innovate and improve its service. But GTEC in Providence did look to the future and decided to replace its mainframe computers with PCs. The result? The percentage market share of the two companies reversed. General Instrument had 80 percent of the market prior to GTEC’s innovation, and within five years, GTEC had 80 percent and General Instrument sold out that business.
Manage risk with an elegant project management system. One approach for risk management is to lay out potential risk in great detail and build in contingencies. The approach is to put together a parallel plan, specifying what will happen if an invention does not work. “There has to be a backup answer to a question,” says Tarshis. “It may not be as good or cost effective, but the product will do what it is supposed to do.”
This careful planning often uses PERT (program evaluation review technique) charts to create a timeline of what has to happen when and what particular tasks depend on other tasks in order to manages risk systematically. “If it gets to the point of unmanageable risk, you go in another direction that is already planned for,” he says.
A related approach is to analyze individual tasks in great detail to determine the qualitative and quantitative risks associated with them. Project development is an accumulation of hundreds of individual tasks. For each one, experienced techies get together to assess a task’s difficulty and then figure out a way around or through that difficulty.
Determine risk by developing and maintaining a database of all previous experience. Say a company prints lottery tickets and has developed software that picks numbers and prevents fraud in a failsafe way. As the company brings on racetrack clients, it has to integrate their existing systems with its proprietary software. Because each new system could be different from what the technical team had seen before, each new client has associated risks.
To mitigate these risks, the technical team develops detailed case studies for every client it has worked with and is able to apply past experience to future challenges. Both sides would solve problems and then sit down together to document failure modes — for example, where computer programs did not work or where interfaces failed — and how they overcame these difficulties.
3M developed a similar approach that has yielded amazing innovation. The company has what Tarshis calls “probably the most fabulous data retrieval system ever seen. They capture every idea, whether usable or not, and store it in a data bank that every employee has access to.”
Post-It notes, for example, were the result of a database entry for a glue that didn’t stick very well. The story goes that one day a secretary to the president said to an engineer, “Gee, I would love to have a way to post things up; I don’t have a place for pushpins on my wall.” The engineer looked in the data bank, swabbed this glue onto paper, and presto! — a Post-It note. “I attribute many of their innovations to the fact that they learn from the past, warehouse information, and make it accessible,” says Tarshis.
Tarshis grew up in Brooklyn, where his mother was a bookkeeper and his father a certified public accountant. He received bachelor’s and master’s degrees in metallurgical engineering from Brooklyn Polytech in 1962 and 1963.
For Tarshis, who says “I’ve been in innovation my whole life,” one of his most powerful experiences of innovation and risk was when he worked in GI’s cable television division — back when they used coaxial cable to distribute the signal. “I and a bunch of others asked, ‘Why don’t we use fiber optics?’”
The junior and senior engineers sat in groups to identify the technological risk by asking the question: Why doesn’t fiber optics work? Mostly, says Tarshis, it didn’t work because no one had looked closely at the problem and tried to solve it. When a light signal was sent down a tube to a television screen, it produced interference lines. By taking account of the fact that the light emitted by the laser in the glass wire was not linear, they were able to design circuitry that fooled the laser into believing the signal was linear. “That was the birth of fiber optics and cable TV,” says Tarshis. “We identified the problem, asked the question ‘Why?,’ did research on the question, found the answer, and solved the problem.”
But once they resolved the technological issue, another risk reared its head. “People in the coaxial cable business fought tooth and nail to avoid some other technology taking over,” says Tarshis. The solution was to show the naysayers they were wrong through creative marketing and management.
Their argument to the coaxial cable supporters was that fiber optics would be replacing coaxial cables, no matter how much they fought. But rather than take away business from the company, fiber optics would provide more business and would go for a higher price because it would save network users so much money.
And what was the lesson that General Instrument’s employees learned about risk? “We showed the employees that risk received a major reward,” Tarshis says.
— Michele Alperin