The concept of crowdsourcing is not new, although current media reports may suggest otherwise. Long before modern crowdsourcing systems were developed, a number of projects took place that tapped the general population to help solve complex problems. The French government, for example, proposed several types of these competitions dating back to the 1700s. The Alkali Prize was awarded to a Frenchman in 1783 who won a competition in which the public was challenged with figuring out how to separate salt from alkali. In 1714, the British government established the Longitude Prize for anyone who could figure out a simple and practical method for the precise determination of a ship’s longitude — a problem that came into sharp focus as people began making transoceanic voyages.
But it wasn’t until recently that the Internet and social media tools made crowdsourcing more efficient and effective. A White House endorsement helped as well. In a memo issued March 8, 2010, the White House urged federal agencies to use challenges and prizes to crowdsource innovative approaches to government initiatives and programs. Today the use of crowdsourcing is on the rise in both the commercial and public sectors. According to Crowdsourcing.org, a research firm that tracks the industry, revenues of business-focused crowdsourcing firms grew 74 percent between 2010 and 2011.
While crowdsourcing is gaining momentum and legitimacy, the question arises: Is there value in its use for federal government agencies? The evidence appears to suggest there is, provided that a few guiding principles are followed.
InnoCentive has been capitalizing on the crowdsourcing concept since 2001. Based in Waltham, Mass., the company was spun out of an experiment at pharmaceutical company Eli Lilly. Instead of adding more chemists with similar backgrounds to solve a problem, a group of Eli Lilly employees asked, “Why not look outside to the greater public?” Realizing that out-of-the-box thinking may be the answer, the group began employing crowdsourcing to find solutions to significant problems while mitigating the chance of failure. The concept eventually launched InnoCentive, a company dedicated to using crowdsourcing to help solve problems.
InnoCentive has been working in the federal space since 2005. During that time, the company has conducted challenges for the Air Force Research Laboratory (AFRL), NASA, In-Q-Tel, the Oil Spill Recovery Institute and others. “We’ve seen a lot of interest growing from the government market,” said Jon Fredrickson, InnoCentive’s vice president of government. “The Strategy for American Innovation Act helped stimulate a great deal of interest, but beyond that, I think the timing is just right,” he said. “Using crowdsourcing, an agency can get a great number of potential solutions to a problem in a very short time frame. They then judge the submissions and pay a reward for the one they choose. This pay-for-performance model works very well in the federal space.”
InnoCentive essentially acts as a broker. Anyone can sign up as a problem solver. A client then relates its particular challenge and sets up the time frame, the reward and any other necessary parameters.
The AFRL is currently working with InnoCentive to crowdsource innovative solutions for a variety of problems. Alok Das, a senior scientist at the AFRL stationed at Wright-Patterson Air Force Base in Ohio, said they are tapping the crowd to help solve 20 challenges that have been defined. For example, one problem seeks innovative designs for a transportation device that can be carried into a combat site by a single rescuer, quickly unpacked to load an injured or deceased person, and used to safely evacuate the injured or deceased person away from the combat zone and into a helicopter. The current design is a collapsible medical stretcher that uses straps to secure the patient, but straps are too slow to use in the field. The AFRL is hoping that crowdsourcing will help with identifying new designs.
Five Ways to Transform the Public Sphere
Turning to large groups of people to solve problems, make decisions and generate ideas in a decentralized way is not a new concept. However, advances in technology and societal changes have made the process of tapping the wisdom of the crowd easier and more direct. Before jumping on the crowdsourcing bandwagon, however, government organizations must first understand the five basic models and when to use the activities associated with each.
Crowd competition: For a problem that has a defined solution but requires creative problem-solving, a contest or prize provides incentives for participants to generate actionable solutions. The contest might yield unrelated innovations in the process.
Want a quick $1.5 million? Figure out a way to store enough solar or thermal energy for a moon rover to operate during 14 days of lunar darkness, and NASA will cut you a check — it’s that simple. The Night Rover Challenge is one of many NASA competitions opened to the public in recent years. Through these contests, the space agency has access to the talent and ideas of highly skilled individuals from around the world at a relatively low cost.
Crowd collaboration: Some organizations face problems not only of creativity, but also of complexity. When citizens combine their ideas and observations, they can scour the details of a problem and build on one another’s insights with a degree of specificity that most bureaucrats only dream of.
In the depths of the recession, the small coastal city of Santa Cruz, Calif., faced a large budget shortfall and the scrutiny of its increasingly agitated citizenry. As pressure mounted, the city looked to these same citizens for help. The general population
analyzed city financial records, volunteered new ideas, and rated and contributed to one another’s submissions through an online platform. With the help of nearly 10 percent of its citizens,
Santa Cruz implemented publicly generated solutions to close a $9.2 million shortfall while raising immeasurable amounts of social capital that still benefit the city and its programs.
Crowd collaboration is ideal for building and sharing knowledge, coordinating emergency response efforts and developing citizen-driven policy.
Crowd voting: Sometimes before decision-making, an
organization needs to harness on-the-ground knowledge from the people who know a problem intimately. In this situation, voting is a means of aggregating the judgments of
the crowd to rank ideas. Crowd voting gives citizens a sense of
buy-in to government decisions, and it gives government valuable insights from people who are, obviously, experts in their own lives.
The dilapidated building next to your apartment building
that has sat empty for years is for sale, and flocks of developers, speculators and city officials now are deciding its fate. Will it be a community center or a charter school? Whatever the final decision is, they’re not asking you, and maybe that’s a problem. Popularise is a platform, currently covering Seattle and Washington, D.C., that invites the crowd to vote on ideas for new uses of old or empty buildings. By allowing the public to vote on proposed ideas, developers and site planners benefit from knowing what developments the local population likely would support.
Crowd voting is particularly good for making simple decisions and ranking options, but not well suited for strategic-level decisions that require organizational buy-in.
Crowd labor: Some tasks are merely tedious. When a project requires little creativity or coordination, just hours of work, crowd labor breaks it into thousands of small, simple tasks.
Geneva has thousands of years of rich history — and nearly 200 years’ worth of unorganized public property records. The city’s government has taken on the task of digitizing and organizing these records by breaking up this monumental effort into thousands of small tasks for the public’s enjoyment through an online video-game-like platform.
Crowd labor activities like microtasking suit efforts such as data validation, translation, data entry and digital archiving.
Crowd funding: Crowd funding is perhaps the most specific and simple way to engage the crowd.
In the midst of the 2010 Haitian earthquake crisis,
millions of dollars poured into aid organizations around the world — not from checks and deposits but from small payments made by thousands of individuals via their mobile phones. The impressive feat of channeling the good intentions of the masses into substantial funds in a short time period was possible because technology has bypassed checks and stamps — the simple act of sending a four-letter text message to a phone number was all that was asked of people.
The potential of crowd funding activities surpasses obvious applications, like disaster-relief efforts, and can include funding startups and individual programs within large organizations.
The successes of public-sector crowdsourcing initiatives and advances in technology have caused the public to expect to engage with government on a personal level. Once an organization creates a platform — like turning Geneva’s land records into a game or making charitable donations as simple as an SMS — it can harness the crowd’s incredible potential. Understanding the different models and uses of crowdsourcing is a key first step for any government agency that wishes to unleash the power of the crowd.
-- William D. Eggers and Rob Hamill
William D. Eggers is the director of public-sector research at Deloitte.
Rob Hamill is a GovLab fellow and consultant in Deloitte Consulting’s
So far, nine of the 20 challenges the AFRL crafted have been completed, with a success rate of about 50 percent. “That’s actually very high, because if you don’t like what you get, you don’t pay for it, yet the intellectual property that’s generated is yours,” Das said. “Overall our experience has been positive. This open innovation is a very efficient way of problem-solving.”
Das said one of the AFRL’s challenges was solved by a mechanical engineer in Lima, Peru. “What are the chances we would have ever found that person if we hadn’t used the crowdsourcing approach?” he asked. “Very slim.”
Fredrickson said occasionally the insights a client gets in response to a challenge can actually be more valuable than a solution. “Sometimes a client will find out there simply is no good solution,” he said. “That can save a company or agency from spending millions of dollars by going down the wrong path.”
The Chicago-based company crowdSPRING focuses on using crowdsourcing to meet graphic design, industrial design and writing challenges for both business and government. Founded in 2008, crowdSPRING now has a community of more than 112,000 designers and writers from nearly every country, according to co-founder Ross Kimbarovsky, who first became interested in the crowdsourcing idea when he was a partner at a law firm. During that time, he routinely chose from the regular group of local vendors to work on the firm’s website, but was increasingly dissatisfied with the results. “I started to rethink the way design and writing products are bought and sold,” he said. “Rather than picking one company and being stuck with whatever they bring you, the crowdsourcing model allows you to pick among actual designs from a huge number of creative designers.”
The company didn’t start out chasing government projects, but over the last couple of years, it has taken on several. Most recently, crowdSPRING hosted a logo-redesign challenge with the U.S. Department of the Interior. According to Kimbarovsky, the department had a well liked and well recognized logo. But it contained too many colors, so it became very expensive to print.
The Department of the Interior wanted something similar that was simpler and cheaper to print. It worked with crowdSPRING to set up a contest offering four awards totaling more than $17,000.
“In the two weeks the project was open, they received 617 designs from 169 designers,” said Kimbarovsky. “They then picked their four favorites. They were very happy with the work and invited all of their employees to participate in the final decision-making, which was also empowering for the employees.”
CrowdSPRING did a website redesign project with the House of Representatives Committee on Ways and Means, as well as a website redesign project for U.S. Rep. Mike Honda.
“It’s a powerful way for companies and government to do business,” said Kimbarovsky. “It exposes them to a much wider group of creative people. It’s also affordable because as a buyer you set your own price and your own schedule. Then it’s up to the public if they want to participate. It’s a very easy, effective and quick process to get really good design and writing solutions.”
In March 2010, the federal Office of Management and Budget (OMB) issued a memorandum on the use of challenges to improve government and encourage innovation. The memo also promised that the administration would make a Web-based platform for prizes and challenges available.
The OMB then commissioned the U.S. General Services Administration (GSA) to select an online challenge platform, and the GSA in turn issued a request for information. Eight organizations responded, and the GSA evaluated the offers and selected ChallengePost as the platform for holding the challenge. “Doing business only via RFPs and through contractors is very inefficient,” said Brandon Kessler, founder and CEO of ChallengePost. “It’s clear that everyone has to do more with less in this economy. There is nothing, in my view, more efficient than harnessing the public’s ideas and solutions. The government can’t be good at doing everything — why not let some people who have different skills and different ideas play a part?”
ChallengePost is available at no cost to federal agencies. Kessler said ChallengePost has more than 200,000 users and has run more than 100 challenges so far, including New York City’s BigApps and Apps for Healthy Kids with the U.S. Department of Agriculture and first lady Michelle Obama. In addition to hosting challenges, the platform also helps put challenges together and promote them.
Kessler believes that there are three parties that win with crowdsourcing. “The government wins because they get unique solutions, concepts and inventions without having to contract and pay for each one. The citizens win because they get the use of these tools and a more efficient government in return, and the entrants win because they get the exposure and recognition.”
Kessler said ChallengePost has been a huge success thus far. “This is a totally new paradigm for the federal government and has already saved them tens of thousands of dollars and created hundreds of solutions.”
Employing crowdsourcing effectively in the federal government requires more than just throwing a problem together and posting it on the Internet. Federal agencies that have had the most success have followed similar protocols, including:
- Pose the right question and word it precisely. “The ability to ask the right question reduces the signal-to-noise ratio,” said InnoCentive’s Fredrickson. “Getting the correct answer to the wrong question will take you further off the path. Asking the right question gives the agency the best chance of success.” Kessler agreed. “The challenge has to be crafted in a clear, compelling way,” he said.
- Have an open mind. Often, the people posting the challenge have a vision of what they want and are too specific in what they want. “It’s important to keep an open mind when looking at solutions,” said crowdSPRING’s Kimbarovsky. “Maybe someone out there has a better idea than you do.”
- Spread the word. Make sure your challenge reaches a large network of people interested in solving similar types of problems so you aren’t speaking to an empty room. Marketing is also important. Make sure people are aware of your challenge.
- Learn from other projects. Many of the challenges posted on crowdsourcing sites are open to the public, so you can see the creative briefs that were submitted and the winning solutions. Reviewing them can give an agency a good flavor for the timing, the award offered and the quality of responses.
- Have an internal champion, and don’t drop the ball. “Getting a great solution is just the first step,” said AFRL’s Das. “It’s important to have an internal champion to run the course of the project. You may receive 100 solutions to go through. Your internal team should be prepared to manage a fair amount of work in choosing a solution, and then they’ll also have to be ready to carry it forward. You need internal buy-in and someone willing to be involved for the entire duration, not just at the beginning.”
While crowdsourcing appears to offer many potential advantages to federal government agencies looking to solve complex problems, additional value may be found in the less-obvious benefits. “There are huge advantages to getting the community involved — to empower people and employees,” Kimbarovsky said. “Government is beginning to recognize this and leverage it in a number of ways, and I think this trend will continue. Crowdsourcing is a way to start conversations that couldn’t take place before because there wasn’t a way to do it.”
Kessler said another less-obvious benefit to crowdsourcing in general has to do with getting a glimpse of what truly motivates people. “Crowdsourcing proves that people aren’t only interested in money,” he said. “People are motivated by helping others and by status and recognition, in this case, by intellectual stimulation and doing something meaningful with their lives. I think that’s a huge driver of crowdsourcing’s success.”