What is microtasking?

The term “microtasking” is often associated with crowdsourcing, which involves both the use of technology and a large number of people. But there’s a difference, according to Alix Dunn, executive director at The Engine Room, a U.S.-registered charity helping organizations navigate the use of data and technology. Crowdsourcing is more about data collection — for example, when citizens take pictures of potholes across the city and send that to the government agency in charge of keeping roads safe for motorists and pedestrians. A number of nonprofit organizations use the concept in various ways. When it launched in 2007, for example, Ushahidi used crowdsourced information to keep track of and map election-related violence in Kenya. Microtasking, meanwhile, is a much more tightly managed process, Dunn said. Unlike crowdsourcing, organizations conducting microtasking already have the data — large amounts of it — but will need the assistance of a large number of people to process it. Think of a librarian needing volunteers to digitally archive thousands of literary works from classics to historical memoirs, or a group of scientists needing help in sifting through thousands of images of breast cancer tumors to look at best treatment options for those suffering from the same disease. Still confused? That’s understandable. The line between the two isn’t well defined. Patrick Meier — an expert and consultant on humanitarian technology and innovation, and the brain behind the iRevolutions blog — lireferred to microtasking as “smart crowdsourcing” in one of his blog posts. Simply put, he said it’s the “process of taking a large task and breaking it down into a series of smaller tasks.” Are development organizations already using it? It’s unclear where the concept originated, but over the past decade, new tools and ways for data collection have become more available to development and humanitarian actors. While the data revolution has presented new possibilities for organizations to assess and understand the work they do and what impact they’re making, this data explosion has also presented new challenges: What will they do with all these data and how do they process it without getting overwhelmed with the flood of information they are getting? This is where microtasking came into the picture, and apart from helping organizations process chunks of gathered information, it has the potential to help them save time. The nonprofit sector has only started looking into the concept in the past five years, one reason the literature is thin on examples of microtasking projects in the sector. However, in the few examples identified by The Engine Room in its research published in 2015 — as part of an Amnesty International microtasking project now called “Decoders” — a number of the organizations featured are using microtasking for data assessment. Greenpeace, for example, has asked volunteers to identify where illegal logging is taking place in Salta province in northern Argentina by comparing satellite images of forest areas from 2008 to present. Likewise, Global Forest Watch harnessed the power of volunteers to map fires across Indonesia’s forest via the use of satellite imagery. Amnesty International asks volunteers to classify whether a particular human rights case the organization has recorded in the past has been resolved or if the person is still in need of support or assistance based on Amnesty’s urgent action report in scanned format. How valuable is it for my organization? Microtasking helps process huge amounts of data, and by tapping a wider set of individuals, an organization gets to engage a broader community and introduce them to its mission. As mentioned above, there are already a number of areas where microtasking is being used by nonprofit organizations, but like any other trend or innovation, it’s no panacea, warned Dunn. “It requires planning [and] a lot of strategic understanding of what it’s useful for. It requires funding to be able to scope what should the ask be, how do we test to see if the public understands the ask in small pilots before we launch a big campaign to get people to participate in something that might not actually be so clear to most people,” she said. This is why organizations need to carefully understand the concept to be able to decide whether it’s right for them to use, and if so, how they should go about it. So how would I know if it’s right for my organization? As it is relatively new, most organizations would likely be in the dark on what they need to do if they are interested in pursuing microtasking for their own purposes. Start with these questions, Dunn said. 1. Are you collecting or are interested in processing large amounts of information that have common characteristics, such as using satellite imagery? 2. Are there processes you’re doing over and over again? 3. Are you collecting data but are having trouble in analyzing them as they were collected haphazardly or compiled in a way that makes it difficult for you to analyze? 4. Do you have the internal capacity to manage communities effectively? If your answer to most of these questions is yes, then there’s a chance that microtasking may be beneficial for your organization. However, if it only sounds like an interesting idea but won’t necessarily support the work you’re currently doing, you may want to think twice, as it may prove to be more of a distraction than a helpful tool, said Dunn. Once bent on using the concept, there are other technical issues you need to consider, such as how do you plan to engage volunteers, will you require user accounts and log-ins for better communication, how do you envision your platform to look like, does your site have enough bandwidth to accommodate the huge numbers of people who are likely to flock to your site for the job? For those who want to learn more about microtasking and others users’ experience, The Engine Room has converted its research in a more digestible format under its “Library.” Those with more specific questions could also reach out to the team. Any caveats I need to know? Yes. One is you have to be clear about what you’re asking from volunteers. If what you’re asking is too wide or is subject to interpretations, there’s a tendency for people to commit mistakes. As Dunn said, microtasking “isn’t a science and people aren’t computers.” Second, you need to put in place a system to verify and confirm information. That means not just relying on one volunteer’s input. Greenpeace requires multiple users to confirm a data input by one user before considering the information to fuel its advocacy campaign. Third, beware of actors who may have an interest in the project you’re working on. In human rights work for instance, there may be actors who would want to skew the data you’re processing. For example, a politician involved in human rights violations may use his or her resources to hire volunteers who would enter false information in a task that evaluates which individuals require support for human rights abuses. In such cases, Dunn suggested setting up red flags that would help alert you when something like that is happening so you can deploy the necessary actions to stop it. Fourth, be mindful of volunteers’ interest. Based on some organizations’ experience, some volunteers may be eager to contribute for periods of a few weeks, while others no more than 5 to 10 minutes a week. In The Engine Room’s research for example, public finance watchdog Fairplay Alliance commented that most of its volunteers would be active between four and six weeks. Adding an element of competition through game play could keep them on for a few more days or weeks.