The Rise of CrowdsourcingCompanies that want to outsource work to low-paid, highly productive workers often don’t need to look overseas to India and China. Instead, the new source of cheap labor can be found almost anywhere, from a teen in a basement to a stay-at-home mom at her kitchen table to a professional moonlighting as an amateur in another field.
Because of the convergence of cheap computing, powerful software, high-speed Internet access, and the networked relationships among Internet users, suddenly the best solution to many problems is to ask for help from millions of people who just might have what you need.
Consider the humdrum but highly lucrative world of stock photography. When a publisher or a graphic designer is putting together a brochure or an annual report and needs a photograph of the Eiffel Tower, a board meeting, or a cat, the usual practice is to order the pictures from a photographer or from a Web site, where prices can run to several hundred dollars per image.
But now, as Jeff Howe recently explained in a Wired magazine article,1 designers are discovering a much cheaper alternative: a Web site called iStockphoto, where standardpictures cost between $1 and $5. For clients, the savings typically amount to more than 99 percent. For professional photographers used to raking in fees from stock photos, it’s a disaster.
How can iStockphoto price its pictures so low? It gets its content from “the crowd” ? specifically, the 22,000 amateur photographers who do something else for a living but contribute their pictures to iStockphoto. Currently, about 1 million photos are available on the site.
For the millions of businesses and professionals that use stock photography, iStockphoto is a perfect opportunity to take advantage of crowdsourcing. Crowdsourcing, according to Wikipedia,2 “relies upon unpaid or low-paid amateurs who use their spare time to create content, solve problems, [and] even do corporate R&D.”
The phenomenal success of Wikipedia itself is the result of crowdsourcing. The online encyclopedia’s slogan is “the free encyclopedia that anyone can edit.” Though it was launched only in 2001, it already contains more than 4 million articles in 200 languages, and more than 1,000 new articles are added every day.
But with about 50,000 registered users who can write and edit articles on any topic, regardless of their educational background or expertise, how reliable is the content? That question gets at the heart of the controversy over crowdsourcing: By entrusting the work to amateurs, crowdsourcing might not produce the same level of quality that results from paying top rates to professionals.
To see if the skeptics are right, the prestigious scientific journal Nature3 compared entries on scientific topics in Wikipedia and the Encyclopedia Britannica, which is considered the gold standard in exhaustive research and accuracy. Nature used the rigorous peer-review process, assigning experts in each topic to study entries in both encyclopedias, without being told which publication provided the article.
Of the 42 entries they compared, Nature’s experts found very little difference in accuracy: The average science article in Britannica contained three errors, while Wikipedia contained four.
Considering that anyone can contribute to a topic on Wikipedia, that level of accuracy might seem amazing. However, because the vast community of users continuously updates the entries, the content is constantly being reviewed, and errors can be corrected instantly.
The crowdsourcing movement also gives companies an advantage they can’t get from their in-house experts: the broad range of perspectives and the wealth of creative ideas that emerge when thousands of brains, not just a few, focus on new ideas.
A case in point is InnoCentive. The pharmaceutical company Eli Lilly created the Web site as a “virtual talent pool” in 2001, and it has become a place where corporations turn to outsiders for innovative ideas. Among the 30 leading corporations that haveregistered as “seekers” of ideas on InnoCentive are Procter & Gamble, Boeing, and DuPont.
When they have a problem their in-house staffs can’t solve, seekers post it on InnoCentive’s Web site, where more than 90,000 scientists in 175 countries can compete to solve the problem for awards that can be worth as much as $100,000.
Alpheus Bingham, a scientist and the president and CEO of InnoCentive, told the New York Times,4 “We are talking about the democratization of science. What happens when you open your company to thousands and thousands of minds, each of them with a totally different set of life experiences?”
One answer is that you get solutions that your own scientists might have missed. InnoCentive’s Web site lists more than 100 award-winning solutions. Jill Panetta, the chief scientific officer at InnoCentive, told Wired5 that 30 percent of the seekers’ problems are solved, “which is 30 percent more than would have been solved using a traditional, in-house approach.”
Another answer is that you save money. According to BusinessWeek,6 “The success rate has been far higher than in-house performance, at around one-sixth of the cost of doing it all in-house.”
The strength of InnoCentive is that it attracts ideas from people who aren’t working side by side with people from the same backgrounds in the same research labs in the same cities. For example, Procter & Gamble’s staff of 9,000 scientists and researchers is likely to approach a problem in the same way. But a truly novel idea might come from someone with a completely different perspective ? a person in a different company, a different industry, or a different country.
According to Wired, when a team from MIT studied problems at InnoCentive, they found that “the odds of a solver’s success increased in fields in which they had no formal expertise.” That makes sense, because the seeking company’s in-house chemists could probably find a solution to a chemistry problem just as well as an outside chemist. However, a physicist or engineer can bring an entirely different background and frame of reference to the problem.
That’s what happened when Edward Melcarek spotted a problem posted on InnoCentive by Colgate-Palmolive. The company couldn’t figure out how to pump fluoride powder into a toothpaste tube without losing most of the particles as they sprayed through the air. Melcarek, a 50-year-old Canadian with a Master’s Degree in Engineering
Science who says he “designs painting robots to pay the rent,” lists his technical expertise on InnoCentive’s site as encompassing aerospace, electronics, remote sensing, particle physics, and machinery.
What Melcarek doesn’t have is an education in chemistry or biology, the two types of problems that seekers bring to InnoCentive. He simply applies what he knows from particle physics to each challenge, and tries to come up with a solution within 30 minutes of brainstorming.
Using this approach, Melcarek instantly recognized the solution that had eluded Colgate’s top minds. All they had to do was ground the toothpaste tube and then positively charge the fluoride particles. The powder would then be drawn to the tube like metal to a magnet, without getting lost along the way.Melcarek won the $25,000 award for the idea, and Colgate saved hundreds of thousands of dollars in research costs.
Based on the crowdsourcing trend, we offer the following three forecasts:
First, companies will increasingly turn to crowdsourcing as an alternative to in-house R&D ? and they will find huge savings. Procter & Gamble’s vice president of innovation and knowledge, Larry Huston, points out, “Every year, research budgets increase at a faster rate than sales. The current R&D model is broken.” P&G’s goal is to increase the number of innovations that originate beyond the company, from 15 percent to 50 percent. It now draws on the expertise of 1.5 million researchers through networks like InnoCentive. Using that approach, its R&D productivity is up 60 percent.
Second, crowdsourcing will boost the economy by boosting productivity. Just as previous trends like outsourcing and advances in information technology accelerated the growth of the economy over the past several years, crowdsourcing will make a positive impact on GDP. Costs will fall as more companies enjoy the benefits of lowering the costs of work that was once done only by highly paid in-house professionals, and productivity will increase as firms solve crucial problems that once stood in the way of new innovation.
Third, expect to see established companies strive to acquire upstart competitors that threaten their business models with crowdsourcing. For example, Getty Images, a leading professional stock photo company, recently realized it couldn’t compete with the miniscule rates charged by iStockphoto ? so Getty bought the company earlier this year for $50 million. While it is unlikely that cash-strapped Encyclopedia Britannica can afford to buy the upstartWikipedia, it would not be a surprise if Microsoft made a bid for it to minimize Wikipedia’s impact on its free Encarta on-line encyclopedia and its $50 Encarta Premium CD set. References List : 1. Wired, June 2006, “The Rise of Crowdsourcing,” by Jeff Howe. ⓒ Copyright 2006 by The Conde Nast Publications, Inc. All rights reserved. 2. To access general information about ‘crowdsourcing’, visit the Wikipedia website at: en.wikipedia.org/wiki/Crowdsourcing 3. Nature, December 15, 2005, “Internet Encyclopaedias Go Head to Head,” by Jim Giles. ⓒ Copyright 2005 by Nature Publishing Group. All rights reserved. 4. New York Times, March 26, 2006, “Here’s an Idea: Let Everyone Have Ideas,” by William C. Taylor. ⓒ Copyright 2006 by The New York Times Company. All rights reserved. 5. Wired, June 2006, “The Rise of Crowdsourcing,” by Jeff Howe. ⓒ Copyright2006 by The Conde Nast Publications, Inc. All rights reserved. 6. BusinessWeek.com, January 9, 2006, “Business Must Learn to Let Go,” by Linda Sanford. ⓒ Copyright 2006 by The McGraw-Hill Companies, Inc. All rights reserved.
