I was catching up on some podcasts recently, and happened upon a very interesting one from Boston radio station WBUR’s “Here and Now” on “crowdsourcing“. While the phenomenon itself is not completely new, I found the approach to it intriguing.

This issue of “What is…?” describes crowdsourcing, looks at the origins of the term, and briefly considers some of the pros and cons of the practice.

What is crowdsourcing?

The idea of crowdsourcing is a play on the well-known practice of “outsourcing”. Outsourcing is most commonly known as the practice of substituting cheap labor from other countries to cut costs at home. (It also has a more general meaning of simply using skills from outside a company where the company does not have the required expertise internally.)

“Crowdsourcing”, a term coined by Wired author Jeff Howe, refers to the increasingly prevalent practice of using the skills and time of unpaid or low paid amateurs to create content or solution for established businesses. The general idea is to make use of the talents of the “crowd”, particularly those people who are not necessarily employed in the industry with the problem to solve, but nonetheless have a talent that is valuable to that industry.

As defined by Wikipedia (itself a crowdsourcing project), crowdsourcing is used to:

…create content, solve problems, even do corporate R & D. Crowds targeted for crowdsourcing include data companies, such as Jigsaw, garage scientists, amateur videographers, freelancers, photo enthusiasts, smart mobs and the electronic herd. …

Crowdsourcing attempts to replace selectively hired, trained and managed workforces with mass volunteer participation and self-organization. While not a new idea, it is becoming mainstream. Open source projects are a form of crowdsourcing that has existed for years. People who may not know one another work together online to create complex software such as the Linux kernel, MySQL database, and the Firefox browser. In recent years Web 2.0 technology has evolved to allow non-technical people to participate in online projects. Just as important, crowdsourcing presumes that a large number of enthusiasts can outperform a small group of experienced professionals.

The term is new — possibly as new as Howe’s June 2006 article in Wired — but appears to have captured the public imagination very quickly. According to Howe in his crowdsourcing blog, a Google search for the term resulting in only three hits in mid-May, whereas on 27 may the same search returned 182,000 results, and searching today returned 203,000 results.

The examples of crowdsourcing described by Howe range from media industries to hard science. The quality common to these examples is the way that people have contributed their time beyond their everyday jobs, sometimes with startling results, and often with great benefit to the company using their efforts.

For example, stock photography (pre-existing images used for commercial purposes) company iStockphoto offers the work of amateur photographers for as little as $1 an image, in comparison to photos that might cost hundreds of dollars if purchased from an organisation like Getty Images, the world’s largest agency providing images. iStockphoto relies on contributions from individuals — mostly amateurs with good digital cameras and a copy of Photoshop — who tend to regard any extra income made from photo sales as a bonus (and thus do not need to be paid a great deal). Not surprisingly, iStockphoto has been very successful, so much so that Getty Images purchased iStockphoto earlier this year.

Another interesting example that Howe recounts is the use of crowdsourcing by large corporations eager to crack particular research and development problems. InnoCentive, which was founded by pharmaceutical giant Eli Lilly, provides a service to companies with problems to solve. Companies looking for solutions (“seekers”) post problems; individuals (“solvers”) provide those solutions. Seekers pay a fee to InnoCentive, and solvers receive between US$10,000 and US$100,000 if they solve the problem. About 30% of the problems posted by Seekers are solved by Solvers. And the companies that use this service are not insignificant; they include Boeing, DuPont, and Procter & Gamble. Who are the Solvers? Some are hobbyists. Others are highly-qualified — just not necessarily in the field in whch the problem is located.

Who benefits from crowdsourcing?

The benefits of crowdsourcing are significant to members of the “crowd”, particularly when they are paid for their contributions. Receiving $100 for a photograph (when otherwise you would have received nothing) can be a welcome windfall. While contributing a solution to an InnoCentive Seeker’s problem may involve a considerable investment of time, the potential payoff is also considerable, as noted above.

The benefit may be even greater for the companies using the talents and resources of the crowd. As Howe notes in his article, the R&D problems presented on InnoCentive may present small but significant issues encountered as part of a larger project, and being able to solve them outside normal channels can result in significant cost and time savings. This in turn results in a benefit to the ultimate customer, who has access to a product that might not exist otherwise. In the case of iStockphoto, images are made available to consumers at a much lower cost than is possible for companies like GettyImages.

Who loses?

So what can be wrong about crowdsourcing? The argument against crowdsourcing is a bit like that against outsourcing. The objection to foreign outsourcing is that jobs are taken away from domestic workers. The objection to crowdsourcing is that what would otherwise be a salaried position is taken away in favor of what are basically standalone contracts; instead of hiring someone to do R&D, and paying them benefits as well as a salary, the work goes to random individuals who are paid for individual pieces of work.

Is this really happening? According to Howe, there is no data to suggest that jobs are being taken away as a result of crowdsourcing. But then again, there isn’t much data on crowdsourcing at all yet.

The conclusion so far: either it’s too early to tell, or crowdsourcing results in a net gain for everyone.