Is open innovation the future of academic research?
The not invented here syndome is commonplace around the world. It probably stems from Ronald Coase's treatise on the importance of transaction costs. Coase believed that the cost of finding and recruiting talent meant you were duty bound to have the best brains within your organisation. That was of course 75 years ago. Has the world changed?
Quite probably so. Organisations from a wide range of sectors are now opening up their innovation processes to people outside of their organisation. GE for instance have ecoMagination and their recent partnership with Quirky. Lego have their Cuusoo platform. Darpa are using the crowd to design their latest tank, and so on.
Arguably the most far reaching of these attempts at open innovation occured at Harvard however. Back in 2010 they began the Harvard Catalyst project, which hosted idea challenges in the medical field, and in particular on diabetes. The challenge saw 250,000 people produce 150 research ideas that were eventually narrowed down to 12 winners.
It's important because academia is rife with people that believe they're super smart, and as such believe they have all the answers you'll ever need. Being able to open up that process to outsiders is therefore a pretty big leap into the unknown.
First of all it requires a shift in approach. The traditional academic research model requires a scientific team to first of all determine the hypothesis they wish to explore. Panels then decide whether this is a suitable topic for research, and whether it meets funding criteria, before the research is then conducted by the team, and the results put out to peer review.
The shift in approach at Harvard was focused around the relatively narrow field of Type 1 diabetes. They used this relatively narrow topic to explore how openness could benefit the research process, whether this was in generating ideas, evaluating those idea or encouraging insight from diverse fields.
Using Einsteins quote about the importance of framing the problem, they first set about using open innovation to generate the right question. They used a prize-based approach via the InnoCentive platform to determine the direction the research should take. Participants in the challenge had to come up with a well defined problem or hypothesis to advance our knowledge about Type 1 diabetes. They weren't asked to define how they would solve their problem, merely to define the problem itself. The prize on offer was $30,000.
A total of 779 people participated, with 163 individuals from 17 countries submitting a total of 195 ideas. After a quality assurance process, 150 of those went through to evaluation. As hoped, many of the ideas submitted came from people with expertise outside the field of diabetes. Indeed, just 9% identified themselves as having a specialism in the area.
What's more, the ideas were found to be very different to the angles currently taken by the diabetes research community.
Picking the best ideas
The next stage of the process was to sift down the 150 ideas into a more select group that could then be explored. As with the idea generation stage, this process is typically dominated by a small group of supposed experts, each often bringing their own biases to the process. To combat this, a wide range of experts from various fields were invited to review the submissions. Of course, not only did this bring fresh perspective into the evaluation, it also made sifting through 150 submissions an easier task.
In total, 2,130 evaluations of the 150 submissions were made, with healthy debate about the best proposals. Indeed, there was only 4 submissions that were in the top 10% for all reviewers.
In true wisdom of crowds style, Harvard Catalyst chose to aggregate the responses from each reviewer, and provided funding to the 12 submissions with the highest average score. The winning submissions came from a wide range of people, including a HR professional, a retired dentist and a college senior.
Doing the research
The next stage is obviously then to conduct the research into the themes chosen. Harvard wanted to ensure that the research teams had a wide cross section of expertise and therefore reached out to people from outside of the diabetes specialism. The process was then put before the Leona Helmsley Trust to secure funding for the projects.
31 teams were created, each vying for $150,000 funding from the Helmsley Trust. If initial results proved positive, further funding would be secured. In the end, seven winners were announced, of which five had no significant history of diabetes research.
The moral of the story
Harvard Medical School has access to some of the brightest minds in the world, so on the surface they would seem an unlikely home for open innovation. That they nevertheless embraced the concept, and indeed achieved significant input from those outside of the domain being investigated should provide lessons to us all, regardless of the area we are looking to innovate in.
The success of the project required strong leadership to ensure that openness was maintained throughout the entire process. It would not have succeeded had submissions been sought from far and wide, only for them to be reviewed by the usual suspects or for researchers to be selected from a shallow pool. The temptation is always there to resort to the tried and tested, so it requires discipline to resist that urge.
An interesting side effect of the project is that Harvard found a strong latent desire to engage widely amongst their academics. Whilst traditionally their fields of interest have been very narrow, the project unearthed a desire to let their intellectual passions wander. There are clear lessons here for the commercial world, where many talents may lie untapped by managers.