Customers and users are a valuable source of new solutions, better products and services, novel concepts and insights of the future, both in B2C and B2B. Firms want to innovate better, faster, cheaper. In this triangle, crowdsourcing – assigning tasks to a large mass of contributors – offers an enticing mechanism to gain direct access to innovative input of external networks.
But is it that easy?
My personal mission as a knowledge-management expert, researcher and spare-time enthusiast is to catalyze online collaboration – the actual doing-together, not just posting-to-others (as I am yet doing right now!) or general chit-chatting (which has tremendous value for networking and building relationships, but does not equal concrete solutions).
There are already myriad studies and practical advice on what motivates users to contribute their personal knowledge online, starting from the early days of open-source software projects. There are also perfect examples of how online innovation contests or brand communities have provided firms new ideas and solutions, which have eventually turned out better than internal ideas. But the focus of this debate is not in doing-together – indeed, it is in the above mentioned posting-to-others.
As a result, what we coin collective intelligence is far too often individual intelligence, hiding behind the fashionable mask of communication technologies and online services.
In their insightful study, Afuah and Tucci (2012) did consider the general conditions under which crowdsourcing might be an optimal strategy compared to internal problem-solving or using designated suppliers. The starting point for any crowdsourcing initiative is to understand the nature of the task. How easy it is to delineate and transmit? How modular is the problem? In knowledge-management terms: the more tacit knowledge a problem involves, the more difficult it is to transfer such task to an external crowd. Tacit knowledge is sticky, it is based on long-term experience, and can only be ”transferred” by doing things together.
Complexity, in turn, implies that problems cannot be easily articulated due to cognitive limitations. Complexity also implies the problem-solving context involves multiple relations between actors – which are not straightforward to define. For instance, certain experts within a group may share the same professional or organizational background, whereas others are affiliated with competing firms, and others are voluntary enthusiasts who participate just to demonstrate their expertise. To collaborate, such group needs frequent interactions both 1) around themselves, the shared objective, and the means to achieve it 2) the actual problem and its features. And such interactions are costly (see e.g. von Hippel, 1994).
As a result, most crowdsourcing efforts represent the first two modes described below: either the crowdsourcing sponsor announces an open call to voluntary contributors, or pre-selects an ”elite circle” to which the actual problem is then assigned. Both result in individual-driven solutions which compete with each other. Naturally, challenges and contests may also involve an option for teams to participate, such as in this example of Konecranes’ chain wear indicator. But how much do crowdsourcing and collaboration eventually have in common? Research has demonstrated they should have much more: Hutter et al. (2011) found that the best solutions came from users simultaneously collaborating and competing with each other, labelled as communititors.
Why is the ”true” community-mode uncommon in crowdsourcing? Simply put, it requires sharing experience-based knowledge and insight, not only objective information – and this is a challenge for online collaborators. As early as in 2001, Catherine Durnell Cramton identified five reasons of why dispersed groups are not easily able to establish mutual knowledge: the knowledge that is common to each one within the group, and which they are all aware of having. These difficulties relate to contextual information, unevenly distributed information, salience of information, speed of access to information, and interpreting the meaning of silence.
No matter of their degree of novelty, the same challenges still apply with online innovation platforms and crowdsourcing services.
As a final conclusion: if we want to advance
- collective problem solving
- relying on tacit knowledge and experience – not only common knowledge,
- respectively demonstrating the greatest innovation potential,
- within dispersed online groups, skill swarms or communities,
we do not need any new platforms, tools or metrics focusing on individuals. What we need is advancing online sociability, allowing time and space for informal get-togethers and conversations, building trust, actively developing online-collaboration skills, setting common targets and working around them. Without such community-oriented focus, the potential of crowdsourcing for innovation may become compromised.
Why pick the lowest hanging fruit, when there are new trees just waiting to be invented?