Using Social Media to Sample Ideas : Lessons from a Slate-Twitter Contest

Crowd sourcing is one of the many ways in which firms can engage in open innovation. Firms looking for ideas from outside furthermore have to decide whether they want to perform a one-shot broadcast search or whether they want to adopt a more iterative process. While one-shot broad- cast search is relatively well covered in the literature, far less attention has been paid to iterative sampling. The latter mode of sampling, however, is likely to become more and more prevalent as firms move to adopt social media. An important feature of social media is that the crowd that is addressed is aware of itself given that the members of crowd use social media mainly to monitor their peers. Hence it is no longer save to assume that ideas are generated independently. This has several consequences. One-shot broad-cast search has been analyzed by Kornish and Ulrich (2011) on the basis of idea-contests they ran in classroom settings.In particular, they looked at the number of unique ideas that were generated in their experiments and tried to ascertain whether the most valued ideas were surrounded by many similar ideas or not. In this paper we look at data from an idea contest that was organized by Slate on Twitter in 2010. Applying the framework of analysis proposed by Kornish and Ulrich, it appears that the use of social media will limit redundancy but at the same time bias the search