Abstract
Building on recent advances in crowdsourcing research, we argue that, when using crowdsourcing, governments should accurately select the crowd they wish to engage with, depending on the problem to be solved. While targeting a large crowd may be common, it is not always the most appropriate: it can waste significant resources without necessarily producing satisfactory results. We contend that the nature of the problem should determine the crowdsourcing strategy. We propose a typology of problems based on their scope and technicity: specific nontechnical, specific technical, broad nontechnical, and broad technical. We introduce the main crowdsourcing strategies—fishing, hunting, and collective production—explain the rationale for each, and offer a practitioner’s perspective on their costs and benefits. We then discuss how each strategy is suited for solving diverse problems and propose guidelines for governments on using crowdsourcing more effectively.