Henry Chesbrough on open innovation, erosion factors, and multiple paths to market
Henry Chesbrough is the author of the new book, "Open Innovation: The New Imperative for Creating and Profiting from Technology." Beginning August 1, he is Executive Director of the Center for Technology Strategy and Management for the Institute of Management, Innovation & Organization at the University of California, Berkeley. He is an assistant professor at Harvard Business School and spent 10 years working in product planning and strategic marketing at Silicon Valley companies.
UBIQUITY: Tell us a little about your background.
CHESBROUGH: I'm an over-educated disk drive industry retread. I went to college at Yale, got an MBA from Stanford, and worked in the computer industry for several years. I went back to school and got a Ph.D. from Berkeley. Now I'm teaching and doing research.
UBIQUITY: You have a somewhat different view on Xerox's PARC (Palo Alto Research Center) experience. Tell us about it.
CHESBROUGH: The typical story about PARC is that these brilliant scientists in Palo Alto created the future computer industry and those blockheads back in corporate in Stanford, Connecticut just didn't get it. They "fumbled the future," to use the expression of a book that was written in 1988 on this phenomenon. A more recent book called "Dealers of Lightening" covers much of the same ground from a more sympathetic point of view but likewise concludes that the folks at Xerox corporate dropped the ball. So what do I contribute to this story? I remind people that the folks at Xerox corporate were struggling for their own survival at the time of many of the crucial events at PARC.
UBIQUITY: What was the nature of Xerox's struggle for survival?
CHESBROUGH: Xerox had just settled a consent decree and anti-trust suit with the Federal Trade Commission. In the consent decree they were forced to license their xerography patent portfolio. In the aftermath, several Japanese companies entered the copier business and went after the low end of the market. IBM and Kodak came in at the high end of the market. The "dumb idiots" who didn't understand the importance of PARC fought back effectively against both parties and accomplished a great deal in the copier business against some formidable competition. They had more talent and skill than the earlier accounts have acknowledged.
UBIQUTY: Let's give credit where it is due. List some of those accomplishments.
CHESBROUGH: In terms of technology, beginning in the 1960s, the Xerox technology base went from fully mechanical to electro-mechanical to fully electronic and even digital by the 1990s. Xerox was among the first companies in the world to incorporate semiconductor dialed lasers into their printers and copiers. So this was not a poorly managed bunch of idiots running a brilliant research lab badly. If we understand the context, the question becomes even more interesting, which is, how does a well-managed company with such good technology go wrong? How did this get away from them?
UBIQUITY: Well, what's your answer? Why did it go wrong?
CHESBROUGH: In my research, I concluded that they were managing the PARC laboratory according to the best research practices of the era. Where I think they went wrong was in the philosophy they used to manage the PARC, which was a philosophy of "closed innovation".
UBIQUITY: What is your definition of closed innovation?
CHESBROUGH: Xerox was managing its lab according to the best practices of many companies. It was a vertically integrated, internally oriented model in which everything from basic materials through design, development, and manufacturing, all the way through to financing and servicing -- all this was done within Xerox's four walls. That was the mistake for PARC. PARC failed to see how the individual pieces in its vertically integrated value chain could become the foundations of profitable businesses. What was needed to unlock the value in a facility like PARC was a different philosophy toward industrial R&D -- a philosophy that I call, "open innovation".
UBIQUITY: Explain your philosophy of open innovation.
CHESBROUGH: I think of open innovation as meaning that ideas can come from inside the firm or from outside the firm, and that they should be taken to market both within the firm and outside the firm.
UBIQUITY: Can you give examples of companies that are open innovators and others like PARC that would be closed innovators?
CHESBROUGH: AT&T and Bell Labs are another dramatic example of a closed innovation approach. It worked brilliantly for decades, not only as measured by Nobel Prizes, but also in terms of fundamental technology breakthroughs. The communications industry as we know it has its roots in the technologies out of Bell Labs. It was a great scientific and technological fountain of resources. But in the last 10 years or so, the model has come under enormous pressure. One could argue convincingly that the model today is broken. For example, Lucent can no longer sustain the investigation of basic scientific phenomenon in Bell Laboratories because it can't make use of it fast enough to warrant the high level of investment. The IBM mainframe business, where they did all the service, sales and support, is another highly integrated example.
UBIQUITY: Now give examples of companies that use open innovation.
CHESBROUGH: In my book I talk about the new IBM in the PC era. In today's world, IBM's global services integration business unit is a very open company by comparison [to the mainframe business]. IBM's transition from closed to open is fascinating. A second example is Intel. Even though Intel, in terms of sales and profits, is the largest company in the semi-conductor industry, it does remarkably little internal research for a company of its size and stature. Outside the technology sector, Proctor & Gamble has adopted new practices to try to inject external thinking into its product pipeline. At the same time, it has adopted initiatives to take internal ideas and put them into the open market to be licensed to other companies and even competitors to generate new revenues.
UBIQUITY: What were the triggering events that made the closed model less viable?
CHESBROUGH: I call them erosion factors. What I mean by that, both as a metaphor and as an explanation, is that these things didn't happen overnight but they combined to wear away and remove the foundation that supported the closed innovation model. The first thing I point to is the increased mobility of workers. When people jump from job to job, they take knowledge with them. That causes knowledge to spread much more widely. The companies where they learned that knowledge don't get any compensation when these folks leave.
A second one is a great success story for our society that you don't read about much anymore, and that is the higher education system. After World War II and the GI Bill, the number of people going to college expanded enormously. It greatly increased the pool of available trained and talented people that could go to work, and that benefited companies of all sizes.
A third erosion factor is the rise of venture capital. The risk capital and private equity that's now available to support people starting up companies is vastly bigger than it was, say, 30 years ago. When you have a talented pool of people, the diffusion of ideas out of companies, and the risk capital able to finance it, that creates a powerful combination to get the startup companies going. A fourth factor is the rise in quality and quantity of universities.
UBIQUITY: Where does university based research fit into this picture?
CHESBROUGH: The quality of university research has gotten much better relative to industrial research. There was a period, beginning before World War II and even in the '50s and '60s, that the best work in the technology space was being done in the central R&D labs of industry. The universities have closed that gap dramatically and today they're at the forefront of a number of areas in both the basic and applied fields. That again creates a bigger pool of ideas to be drawn on by companies of all sizes. The Bayh-Dole Act of 1980, which allows universities to patent and license technologies, created the incentive to move those ideas out of the universities into companies.
UBIQUITY: How does intellectual property play out in this story?
CHESBROUGH: Once you bring ideas in from outside, or move them to market outside the company, you're immediately in the world of managing intellectual property. It's a keystone of this new model, where people will be buying and selling ideas and technologies. There will be multiple paths to market for ideas and multiple revenue streams that companies can realize for their ideas.
UBIQUITY: What is your view of how corporations should think about the balance of basic research and applied research?
CHESBROUGH: As practical matter, I see fewer corporations investing in basic research. I think the days when AT&T had a monopoly position in telecommunications, or IBM had a hammerlock on the computer industry, are over. One of the costs of those days being over is that those companies can't and don't support as much basic research as they used to.
UBIQUITY: But isn't it also true that even universities have tended to focus much more effort on applied research?
CHESBROUGH: Recent studies show that with the ability now to patent and license research, there's been a big growth in university patents. Much of that growth has come from applied work. While basic research still is being done in universities, it is fair to say that more applied work is being done than before. The hard question is, how much basic research do you need? I think there's some reason to think that we're probably under-investing in basic research as a society. I would say that's not true in the life sciences, but in the physical sciences and information technology areas, I think there is reason to be concerned.
UBIQUITY: What makes you think we are under-investing in basic research?
CHESBROUGH: Well, I was at a conference last month where a gentleman from the defense department said that as recently as 1985, two-thirds of the submissions in the Physics Archives, which is a well-known online physics journal, were from folks in the US. In 2002, only one-third of the submissions were from folks in the US. On the one hand, it's great that fundamental knowledge of physics is becoming more global and there are many more minds working on problems in physics. But from the US perspective, we're probably not funding as much physics work as we should. Moreover, many of the people in the US doing that work aren't US citizens. Many folks in the master's and doctorate levels from other countries study here because we have wonderful higher education, but we're not training that many folks in our own school systems to join those ranks. The issue is not just corporate spending on basic research. I think it goes all the way back to science and technology education in the school systems.
UBIQUITY: How would you address those issues, or would you address them at all?
CHESBROUGH: From a company standpoint -- and my book is written primarily for company managers -- you can't fight those forces on your own. You have to align yourself with the realities that are there. These realities mean the companies have to look more outside their own four walls to keep their innovation systems healthy.
UBIQUITY: In your book you focused on a discussion that IBM's James McGroddy had with a Citicorp guy who forcefully told him that IBM's research interests had nothing directly to do with Citicorp's interests, such as applications and customers. Shortly after that time, IBM started shifting its research toward projects that would be of more direct use to its customers.
CHESBROUGH: That's correct. An event that preceded that conversation was a trip that Ralph Gomory made when he was running IBM research. He went to Japan to survey the industrial laboratories of companies such as Hitachi and Fujitsu and others who were competing with IBM. He came back with a sense that IBM's technology in the lab was still without peer in the computer industry, but the Japanese were getting their ideas into the market much faster than IBM was. In terms of what was actually shipping, IBM was increasingly under pressure to keep up, and so IBM's research needed to start moving to market much faster. That was Gomory's take. Then McGroddy took over and struggled with the problem. He had the conversation with the guy at Citicorp. This coincides with IBM's near-death experience at the last quarter of 1992 and the first quarter of 1993, when John Akers was replaced and Lou Gerstner was brought in. During this period, IBM recorded what was then the largest quarterly loss in US business history. Lou Gerstner writes in his new book that IBM was close to technical insolvency when he came on board. The situation was drastic and quite severe. In my own view, if it hadn't have been so bad, it would have been much harder for IBM to make the transition to an open innovation model.
UBIQUITY: Is there any danger that company research organizations will become more slaves to transient customer whims and self-perceived needs rather than to forging new directions, marching to their own drummers?
CHESBROUGH: To the extent that veteran R&D managers are voicing these questions, surely the perception is there. Alan Kay expressed it when he said, "the best way to predict the future is to invent it." He was taking a strong view that the research and technology folks can lead a company into a new future by dint of their hard work and foresight. I'm a bit more skeptical of that.
UBIQUITY: What is your take on the state of research and development?
CHESBROUGH: Most companies set up their research organizations as cost centers. They set up their development arms as profit-and-loss centers, or business units. This creates some interesting budget disconnects between R&D. The folks running research try to hit their budget because going over budget is a bad thing. They also don't want to go much under budget because that means there was more research that could have done. If they keep going under budget, then somebody will take money away from them. Subject to that, they want to run as many projects as they can, and learn as much new stuff as they can. That means that once they've worked on something for a while, developed an understanding of it conceptually, and maybe written papers and demonstrated it in a laboratory setting, they are ready to hand off the project.
UBIQUITY: The project then goes to development. What happens there? What's the disconnect?
CHESBROUGH: The folks in the business unit, or the development organization, want to make a profit. They want to minimize risks and losses. Much of the research organization's stuff looks quite risky and unproven to development. There's an awkward disconnect where the researchers think it's done and the development organization says no, it looks very promising but it needs more work. Things can literally get stuck on the shelf between R and D. At Procter & Gamble, for example, a head of global R&D was quoted in an article saying that approximately 10 percent of the stuff developed in their research organization is in use in their products. And that 90 percent is sitting around unused in the company. I applaud them for their honesty.
UBIQUITY: Do you think that is usual number, that a company would develop only 10 percent of its research?
CHESBROUGH: That figure is probably not atypical for most companies. A lot of stuff doesn't get used. This is where the idea of multiple paths to market enters the discussion. In the old model the only way that stuff got to market was through your own channels of distribution. In the new model that is perhaps a preferred path to market but it's by no means the exclusive path to market. Under many circumstances, you'll make stuff available for use and license by others, in part to create use and money from things that are being unused inside your own company.
UBIQUITY: We were just talking about the difficulties between the research and the development arms. But the troubles are by no means over once a product gets into the development organization. Comment on the nature of the relationship between R&D and marketing.
CHESBROUGH: One problem area is taking a technology to a new market versus taking a new technology to your current market. It's not easy to take new technologies to your current market but at least you know something about the customers. You have an idea of their needs, how they'll use the product and what it will replace. That gives clues to the marketing organization for how to prioritize the requirements. What are the must-haves versus the nice-to-haves? On the other hand, when you take a technology to a new market, it is difficult for the marketing organization to know what to specify and how to prioritize. That greatly complicates the targeting of technologies to new and unproven markets. My colleague, Clay Christensen, wrote a book called "The Innovator's Dilemma," about disruptive technologies. One of his major points is that companies tend to do a very good job of serving their current customers in their current markets. But those same people tend to create a myopia that causes companies to overlook or pay insufficient attention to new markets.
UBIQUITY: Did you have any personal observations of this behavior when you worked in industry?
CHESBROUGH: In the disk drive industry at Quantum Corporation, where I was occupied, we saw this with IBM all the time. IBM was enormously effective in the mainframe storage market. On the other hand, we found markets in personal computers, workstations and network servers that IBM didn't participate in. They didn't understand the opportunity in these markets, or else they didn't have the business processes to compete as an OEM selling to other computer manufacturers. So, even though IBM had better technology, we were able to serve these markets very profitably. I attribute at least some of that to difficulties between the marketing organization and the R&D folks.
UBIQUITY: You were at Quantum from 1983 to 1990, and worked there as a consultant for another five years. What did you do there?
CHESBROUGH: I was a product manager and later on became the vice president of marketing. I had responsibility for all the customer-facing side of things with the sales organization, and also worked with the engineers to manage the schedules and define the requirements for the products.
UBIQUITY: Now you are in education. What do you teach?
CHESBROUGH: I teach courses on managing technology and innovation. The most recent course I taught, with Clay Christensen, was called "Building a Sustainably Successful Enterprise." This fall, I'll be teaching a course called "Managing Innovation", at Berkeley's Haas School of Business.
UBIQUITY: You're involved in a new business concerning neurological disorders. Tell us about that.
CHESBROUGH: My daughter has epilepsy. In the course of learning about her situation, I ended up ransacking the Web for all the information I could find. While there was great medical research for physicians, I found little of value for patients and their families. In the process, I met other people who were also seeking information. That led to creating a Web site called Epilepsy.com. The idea of the Web site is to provide information to patients with the disease and their families, especially about how to live with the disorder and what are some of the options. After we did the Web site, we set up a research foundation called the Epilepsy Cure Project. It is a charitable foundation intended to act as a catalyst to fund promising new therapies or devices for epilepsy. Finally, we're raising a first-round venture capital fund for the profit-seeking opportunities in the epilepsy market space. The large pharmaceutical companies are abandoning the epilepsy market because it's not that big. They have to go after bigger markets to justify their development costs. A bunch of compounds that were in development have been "orphaned" by the big pharmaceutical companies. These compounds could be available for out-licensing and re-purposed to a smaller company for whom the epilepsy market might be a plenty big market. The reason we're doing that is, of course, to try to create more therapeutic choices for epilepsy patients and their families. But from the commercial side, there may be some profit-making opportunities there too, because these therapies will have some level of development already done and some of the risk will already be gone. So there may be a way to make money with products that address smaller markets that the big companies are walking away from. It is a very "open innovation" kind of concept.
Printer Friendly Version
Ubiquity welcomes the submissions of articles from everyone interested in the future of information technology. Everything published in Ubiquity is copyrighted ©2003 by the ACM and the individual authors.
A Ubiquity symposium is an organized debate around a proposition or point of view. It is a means to explore a complex issue from multiple perspectives. An early example of a symposium on teaching computer science appeared in Communications of the ACM (December 1989).
To organize a symposium, please read our guidelines.
Ubiquity Symposium: Big Data
- Big Data, Digitization, and Social Change (Opening Statement) by Jeffrey Johnson, Peter Denning, David Sousa-Rodrigues, Kemal A. Delic
- Big Data and the Attention Economy by Bernardo A. Huberman
- Big Data for Social Science Research by Mark Birkin
- Technology and Business Challenges of Big Data in the Digital Economy by Dave Penkler
- High Performance Synthetic Information Environments: An integrating architecture in the age of pervasive data and computing By Christopher L. Barrett, Jeffery Johnson, and Madhav Marathe
- Developing an Open Source "Big Data" Cognitive Computing Platform by Michael Kowolenko and Mladen Vouk
- When Good Machine Learning Leads to Bad Cyber Security by Tegjyot Singh Sethi and Mehmed Kantardzic
- Corporate Security is a Big Data Problem by Louisa Saunier and Kemal Delic
- Big Data: Business, technology, education, and science by Jeffrey Johnson, Luca Tesei, Marco Piangerelli, Emanuela Merelli, Riccardo Paci, Nenad Stojanovic, Paulo Leitão, José Barbosa, and Marco Amador
- Big Data or Big Brother? That is the question now (Closing Statement) by Jeffrey Johnson, Peter Denning, David Sousa-Rodrigues, Kemal A. Delic