In a world of rapid digital change, the pressure to innovate has increased dramatically. Many firms have turned to crowdsourcing—obtaining information for a task or project from a large number of people, paid or unpaid, typically via the Internet—to identify innovative ideas. Yet many crowdsourcing efforts fall short of expectations or are abandoned. In a new study, researchers developed a framework to help managers understand the process of crowdsourcing so they can use it more effectively.

The study, by researchers at Carnegie Mellon University, Northeastern University, Babson College, and Boston College, is published in MIT Sloan Management Review.

"Crowdsourcing can be an invaluable source of innovation, but only if the right type of crowd is used to address the right type of problem," explains Anita Williams Woolley, Associate Professor of Organizational Behavior and Theory at Carnegie Mellon University's Tepper School of Business, who coauthored the study. "Our framework helps managers match the right type of crowd with a problem of appropriate scope and complexity."

One reason some crowdsourcing efforts fail is the misconception that there is just one approach to crowdsourcing. As a result, many crowdsourcing efforts ask people to provide ideas for a task or project for which they may not be well-suited. Recognizing these obstacles to successful crowdsourcing, the researchers developed a framework that identifies three distinct types of crowdsourcing:

• Search crowds are the most effective way to find solutions to well-defined problems with relatively small scope. In this approach, which is easy to set up and has been used by some of the most well-known crowdsourcing platforms, seekers are connected with solvers. Firms need to provide a clear description of the task and the right incentives to attract the people they need, make the problem accessible to a wide audience, and recruit diverse types of solvers.
• Wired crowds help firms find solutions to more complex problems that draw on many different areas of expertise. In this approach, which takes more time, individuals need to connect with others to share information, build knowledge, and be able to reuse solutions developed by others. Firms need to identify which ideas are useful and offer guidelines and feedback on the group's work.
• Crowd teams are best used to implement solutions such as prototypes or software, or to tackle larger and more complex problems that are difficult to break into smaller pieces. In this approach, holistic solutions to problems can be identified, drawing on multiple areas of knowledge. Firms need to encourage bursts of activity rather than continuous streams of messages and promote diverse discussion topics.

"Using crowdsourcing for innovation isn't always easy," acknowledges Christoph Riedl, Associate Professor of Information Systems at Northeastern University, who led the study. "Our framework helps managers understand how to use crowds most effectively so they can unlock the power of smart crowds for their firms."