Heterogeneous Submission Behavior and its Implications for Success in Innovation Contests with Public Submissions

Innovation contests are increasingly adopting a format where submissions are viewable by all contestants and the information structure changes during the contest. In such an ‘unblind’ format, contestants must weigh the costs of revealing their submissions against the benefits of improving their submissions through emerging information. We take a closer look at how contestants solve problems in innovation contests with public submission of solutions–that is, unblind contests, by examining the implications of their submission behavior for contest outcomes. We analyze the submission behavior in terms of three dimensions: the position of first submission by the contestant, the number of submissions the contestant makes, and the length of active participation by the contestant. The econometric analysis of a large dataset of unblind innovation contests and participating contestants indicates that, despite the potential for free riding and intellectual property loss from disclosure of submissions, contestants who have a lower position of first submission are more likely to succeed in the contest. Further, we find some evidence of a curvilinear relationship between a contestant's number of submissions and her likelihood of success, indicating a potential ‘quality–quantity’ trade‐off in unblind innovation contests. Finally, our findings indicate that increasing the length of participation in a contest has a positive effect on a contestant's likelihood of success. Departing from prior studies on innovation contests, where a contestant's success is assumed to be a function of her prior experience and problem‐solving skills, our study provides new empirical evidence that, in innovation contests with public submissions, the submission behavior of a contestant also plays an explanatory role in a contestant's success.