By Jake Orlowitz
It’s easy to look at projects like Wikipedia, Reddit, Duolingo, StackOverflow, or Zooniverse and think, we just need to take problem X, add some “crowd” to it (like a magic spell or recipe), and then boom: breakaway success.
If only it worked like that.
Most crowdsourcing projects, like most human efforts, fail.
Crowds are phenomenal tools because they’re made up of people, and people are the most important resource in any initiative.
Your crowdsourcing effort will most likely fail if…
- your crowd is not diverse.
- your crowd all thinks alike.
- their task is not clear.
- their mission is not compelling.
- the technical platform is poorly designed or overly complicated.
- there are not continued areas for growth and engagement over time.
- the interface and the organizers are not responsive to change.
- the community lacks social moderation or healthy behavioral norms.
- it lacks mechanisms to address technical abuse and human harassment.
- you do not recognize or empower the core users of your platform.
- you lock it down and people have to jump through hoops to participate.
- potential users lack free time, skills, access or awareness to contribute.
- volunteers are hampered by legal restrictions or monetization attempts.
- another more interesting or better crafted opportunity comes along.
- you never attract enough people to have a crowd in the first place.
The above is not a guaranteed checklist. You can do them all and still fail. Or, if your project mysteriously gets popular really quickly, because it scratches an irresistible itch or fills an unmet need, you might be an exception in which despite gaps in the list, it mostly all comes together, and no one knows what the magic secret was, and you wind up with the philosophical tagline, “it only works in practice; in theory it could never work” (an actual aphorism about Wikipedia).
Your crowdsourcing effort will fail, most of the time, because most things fail. And because important things are hard.
Success as a project also doesn’t mean you are perfect by any means. Wikipedia and Reddit still struggle with serious harassment issues. In some cases, you can “get away with” a level of perfomance that is sufficient but far from ideal. It’s also possible that you can only get away with it for so long before it threatens the core, sustainability, or growth of the project.
Knowing all this, next time you have a problem and want to add some crowd to it, at least consider the people, ideology, task, mission, platform, journey, adaptations, mores, resiliency, motivators, barriers to entry, prerequisites, distractions, competitors, and core users.
Then, just maybe, you too can harness the power of the crowd!
Jake Orlowitz, Wikimedian
This post is inspired by the work and writings of Clay Shirky, Jimmy Wales, Alexis Ohanian, Luis von Ahn, Joel Spolsky, Chris Lintott, Joseph Reagle, Anasuya Sengupta, Siko Bouterse, Jonathan Morgan, Andrew G. West, Aaron Halfaker, and Sue Gardner.
The text is licensed CC-BY-SA 4.0. It can be shared or reposted without permission under the terms of this Creative Commons license, which requires only attribution and that re-users keep the same terms.