Open innovation is not open distributed innovation

As noted, I’ve been reading a lot of “open innovation” papers recently, in service to a greater good.

In addition to the definition of “innovation,” another area of conceptual fuzziness (if not obfuscation) has been treating “open innovation” and “user innovation” (or “open, distributed innovation”) as if they’re the same thing. They’re not. Eric von Hippel has said so, and I think I know Henry Chesbrough well enough to say he would say so as well.

I personally am proud of having a foot in both camps, having tried for the past five years to both link and draw nuances between these literatures. I joined the OI world at the seminal AOM 2004 workshop on open innovation when I met Wim Vanhaverbeke and re-met Henry. (The rest, as they say, is history.)

In 2008, I was fortunate to attend my first user innovation workshops. Over the next three years, I got to know the UI community and Eric von Hippel.

While I’ve published two papers so far about open and user innovation, I think Linus Dahlander is today perhaps the best example of someone who has published significant work in both camps — including 170+ cites for his Research Policy paper two years ago (with David Gann).

I was honored when Dahlander and Gann mentioned my own small part (thus far) in linking these two worlds, first at OUI 2009 and then in their RP paper:

In all, 244 scholars have worked on 150 papers. This figure illustrates that the community is relatively fragmented with a few scholars that have collaborated with several others. There are few bridges connecting teams of researchers with the exception of West/Lakhani, who have connected open innovation researchers with scholars investigating user aspects of open innovation. (Dahlander and Gann, 2010: 702).

I think Dahlander and Gann have done the best job thus far of stepping back and seeing the similarities between the these two streams without blurring the differences. At the same time, my earlier work (in less influential journals) makes explicit the differences between the two.

Marcel Bogers and I have also been working on this perspective for a few years, in a paper we posted to SSRN and will be presenting in London. After publication of Dahlander and Gann, we have a higher bar to clear.

References

Bogers, Marcel and Joel West, “Managing Distributed Innovation: Strategic Utilization of Open and User Innovation,” Creativity and Innovation Management, 21, 1 (March 2012): 61–75. DOI: 10.1111/j.1467-8691.2011.00622.x

Dahlander, Linus, & Gann, David M. "How open is innovation?” Research Policy, 39, 6 (July 2010): 699-709. DOI: 10.1016/j.respol.2010.01.013

West, Joel. “Policy Challenges of Open, Cumulative, and User Innovation,” Washington University Journal of Law & Policy 30 (2009): 17-41.

West, Joel and Bogers, Marcel, "Profiting from External Innovation: A Review of Research on Open Innovation," (September 13, 2011). http://ssrn.com/abstract=1949520.

Decisions for June's OI conference

On Wednesday the four editors of the Research Policy special issue on open innovation notified authors as to whether their paper was accepted for the special issue conference. While it was a day later than promised, we still turned around decisions in 16 days, in time for the accepted authors to make plans to be in London June 25 and 26. (More info will be posted to the conference website.)

We were surprised at the volume of submissions that we received: 78 papers and abstracts. I had personally been expecting about 30-40. This large volume is a testament to the high level of interest in open innovation in the research community.

Given the high demand, we wanted to accept more authors to share their work and participate in the conference. However, there was no way to add plenary papers without extending the conference, and we felt that multiple tracks would undercut our goal of sharing ideas across the entire conference.

Thus, we decided to add a poster session on Monday night, at a prominent time (before the conference dinner) when we expect nearly everyone to attend. With this poster session, more authors won an invite to present their work at the conference and otherwise fully participate in the discussion.

Overall, we accepted 37 plenary papers and posters. We will be posting the schedule once the invited authors have confirmed their intention to participate.

Unfortunately, given the aggressive schedule, we were not able to provide personal feedback for the accepted (or rejected) papers. For each papers or poster presented at the conference, after June 26 one of the four guest editors will provide specific direction based on our earlier reading, the reading of the final paper and the reaction at the conference.

Participating in the conference is not a requirement to be accepted into the special issue. On August 31, we expect to see plenary papers, posters, rejected submissions and brand new submissions. I would not be surprised if we receive even more than 78 submissions, leaving us with tough choices to create a special issue with less than 20 papers.

Having been on the other side, I know that being rejected from the conference will come as a great disappointment. Every paper was independently scored by two of the four guest editors and there was a high degree of consistency in the evaluations. While we felt there was a clear divide between the accepted and rejected papers, it's always possible that we made a mistake. However, given the high number of expected submissions, papers that were rejected are unlikely to be more successful unless they are significantly improved.

I want to offer two general pieces of advice to all three groups of authors (accepted, rejected, and new submissions). First, this is a special issue of Research Policy, and nothing will be published that fails to meet Research Policy's rigorous standards. Authors should look at their papers -- or get peer feedback -- to assess whether they meet these standards.

Second, only papers that build upon and contribute to open innovation will be published in the special issue (as opposed to a regular issue of Research Policy). There is certainly room for research that challenges existing thought in open innovation, and in fact engaging the existing research is the expectation for all papers in the issue.

Our ultimate goal for the special issue is to publish a diverse set of contributions to open innovation research with a variety of research designs, approaches and perspectives.

For those that are designing and writing open innovation research, I have my own personal ideas about what constitutes open innovation research. I will continue to post these observations to this blog.

The Power of Henry Chesbrough

Earlier today, Nathan Mattise posted an article on the celebrations at PARC marking the 10th anniversary of its (quasi) spinoff.

Veteran OI researchers know that OI started at PARC, with Henry Chesbrough’s research on Xerox PARC that led to Chesbrough & Rosenbloom (2002) and then his original Open Innovation book. And apparently Chesbrough gave the opening presentation Thursday afternoon at the event entitled “The Power of 10,” celebrating open innovation.

But what really caught my eye was the picture of the live illustration that Heather Willems did for the event. (Live illustrating is a local fad in the Valley, which I find interesting but less concrete than live blogging.)

In the Ars picture, the signboard starts with “The Power of 10” in the stylized PARC design for the event. But at first glance, the signboard says “The Power of Henry Chesbrough.”

I’ve made a few contributions to OI, open source and standards research, but nothing to earn a slogan with my name in it. I’m not sure how (or if) I’m going to get there, but then I don’t think Henry necessarily anticipated this outcome when he started a decade ago.

Photograph by Ars Technica.

What "innovations" are inbound?

In doing several lit reviews of open innovation, I was struck by how often studies of inbound “open innovation” weren’t about innovation — at least as it has been defined 40 or 50 years of innovation studies. This sort of sloppiness fogs the interpretations of empirical findings and the cumulative nature of the scientific process. Here is my first cut in this blog at trying to cut through some of this fog.

First, let’s leave aside the question of non-innovative content. Sourcing Wikpedia articles or product reviews from consumers isn’t innovation, any more than newspaper reporters or Consumer Reports are creating innovations. It’s just content. Yes, some text would fall under the “creativity” lit, but writing a tertiary semi-encyclopedia and movie reviews doesn’t seem like it would even fit that category.

However, what seems to the most common mess is when “innovation” is used as a synonym for “knowledge” or “invention” or other things that tend to be antecedents of innovations.

We know that an “invention” is not an “innovation” — from various sources including Joseph Schumpeter, Chris Freeman, Ed Roberts and Henry Chesbrough. For example, in my AOM conference paper with Marcel Bogers (Bogers and West, 2010: 4) we wrote:

As conceptualized by innovation scholars, the industrial innovation process comprises both a technical component (invention) and also the commercialization of that technology (innovation). Schumpeter (1934: 88) concluded that technical inventions “not carried into practice ... are economically irrelevant,” while Freeman (1982: 7)† argued that “inventions ... do not necessarily lead to technical innovations. In fact the majority do not. An innovation in the economic sense is accomplished only with the first commercial transaction.” …

[Another] definition of innovation … is given by Roberts (2007: 36): “Innovation is composed of two parts: (1) the generation of an idea or invention, and (2) the conversion of that invention into a business or other useful application.”

This very same sentiment is articulated in the Chesbrough’s prequel to his open innovation manifesto:

The inherent value of a technology remains latent until it is commercialized in some way. (Chesbrough and Rosenbloom, 2002: 530).

But that’s only part of the mess, which goes beyond the invention vs. innovation distinction. Other inputs include the provision of knowledge, components or complements, as Marcel and I wrote in a paper published earlier this year (Bogers and West, 2012: 62):

Discussions of distributed innovation processes tend to blur the distinctions between innovation and its origins and effects. However, all the firm-centric perspectives consider how firms access external sources of knowledge to supplement their own knowledge as an input to their innovation efforts. …

In some cases, firms will rely on external actors to supply knowledge that serves as an input to creating their own innovations. This includes basic scientific research produced and disseminated through open science1 processes, knowledge of market needs and demands obtained from customers, or broad- cast search used to identify promising avenues for future innovation (David, 1998; Lilien et al., 2002; Jeppesen & Lakhani, 2010).
…
The external innovator may also commercialize his or her innovation in the form of a product that is sold to the focal firm (cf. Shah & Tripsas, 2007). These products may be components or other materials that are integrated by the firm into its own products, as has become the norm in the personal computer industry (Dedrick & Kraemer, 1998). Alternatively, the research and development (R&D) of an equipment supplier is used to produce innovations incorporated in tools purchased by producers, as when domestic machine tools improved the post-war German auto industry. Supplier innovations may thus come in the form of materials, components and equipment; Laursen and Salter (2006) found that suppliers were the most common source of external knowledge for innovation among 2,707 UK manufacturers.

Finally, complementary innovations produced by external participants may be provided directly to users. In some cases, these complementary products are sold by for-profit firms, as is common with third party computer software (West, 2006). In other cases, the complements are provided by individuals, whether in the form of user support (Lakhani & von Hippel, 2003), synthesized musical instruments (Jeppesen & Frederiksen, 2006) or game modifications (West & Gallagher, 2006). While such information, goods or services do not directly involve the firm, they do increase the value of the firm’s products and thus improve its ability to profit from its innovations (cf. Teece, 1986).

Nothing in this discussion is meant to suggest that scholars shouldn’t study the various external sources of inputs that firms use in their innovative efforts. The only point is to draw the distinction between the firm’s innovations and its various antecedents and correlates — just as we distinguish between purchase intention and actual purchase, or market share and profitability.

Similarly, the OI processes can be used to study external sourcing of things other than innovations, as long as the distinctions are clear. The “open source software” model and “open innovation” are not the same thing — even though there are important theoretical and empirical overlaps.

In the same way, nothing here would say we can’t study other processes and draw parallels to open innovation. For example, I think many of the OI processes might apply to nonprofits, even though Chesbrough (2003; Chesbrough and Rosenbloom, 2002) requires alignment to business models. (Perhaps someone should first try to extend the concept of a revenue or business model to nonprofits or even government agencies.)

Also, sometimes we can’t measure innovation process directly, but we can measure something else: patents. That’s why thousands of papers use patents as a proxy for “innovation” when (per Schumpeter, Freeman, etc.) we know they are just inventions. It would be foolish of me to suggest that such patent studies should (or would) go away, but we still need to remind ourselves that the output of a technical invention process is only imperfectly correlated (even at the most efficient firm) with a firm’s output of technological innovations.

† Although Chris Freeman’s (1982) original edition is out of print, the identical point is made by Freeman and Soete (1997: 6).

References

Bogers, Marcel and Joel West. 2010. “Contrasting Innovation Creation and Commercialization within Open, User and Cumulative Innovation,” Working Paper, http://ssrn.com/abstract=1751025

Bogers, Marcel and Joel West, “Managing Distributed Innovation: Strategic Utilization of Open and User Innovation,” Creativity and Innovation Management, 21, 1 (March 2012): 61–75. DOI: 10.1111/j.1467-8691.2011.00622.x

Chesbrough, Henry and Richard S. Rosenbloom, "The role of the business model in capturing value from innovation: evidence from Xerox Corporation's technology spin‐off companies," Industrial and Corporate Change 11, 3 (June 2002): 529-555. DOI: 10.1093/icc/11.3.529

Freeman, Christopher. 1982, Economics of Industrial Innovation, Cambridge, Mass.: MIT Press.

Freeman, Christopher & Luc Soete, 1997, Economics of Industrial Innovation, 3rd edition, Cambridge, Mass.: MIT Press.

A deluge of Open Innovation research

I just wanted to say that submissions for the June conference on open innovation at Imperial College London are now closed. The editors are reviewing the manuscripts in hopes of meeting our self-imposed May 1 deadline for notifying people.

As expected, most of the papers were from European authors, validating our decision to hold the conference in London. However, we received far more papers than expected, and the number of attendees is limited by budget and room size, so the review process will be far more difficult than anticipated.

For those who did not submit, I want to reiterate a key point. While the conference is intended to develop papers for the special issue of Research Policy, papers that are accepted in the special issue will be determined by what we receive (through the RP manuscript system) by August 31. Submitting (or having accepted) a paper for the workshop is not a pre-requisite for participating in the special issue. That said, I expect that the conference papers that we are forced to reject will be those that are less well developed, or where there was not a clear effort to integrate the research design and contribution with prior research on open innovation.