Kickstarter has managed to avoid a widespread collapse of its ecosystem despite the fact that it's made up of complete strangers donating money to frequently quixotic projects.
In a new study, Ethan Mollick, an assistant professor at Penn's Wharton School of Business, explains why.
He took surveys from 47,188 project backers, representing 30,323 projects. He found that no one factor, like age or gender of the project creator, could predict which projects would fail, which he defined as a backer not receiving their reward.
On average, backers saw a failure rate on only about one in 10 projects.
"I was surprised — it seemed low," Mollick admitted to me by phone.
The only somewhat reliable predictor of project failure was the size and type of project being funded. Smaller projects and very large projects tended to have higher failure rates—the former because they faced less scrutiny, and the latter because of their complexity. Here's his chart:
That sweet-spot of $1,000 to $50,000—where about 60% of projects fall—is where the answer to Kickstarter's success lies.
"It's Linus' Law," Mollick said, referring to a maxim put forward by Linux creator Linus Torvalds that the more "eyeballs" on a project, the easier it is too ferret out problems. "If a project gets to $50,000 it will attract more attention, and raise alarm if something is going wrong, compared to a project raising under $1,000.
In the rare instance a project failed, backers got refunds just 13% of the time, implying that if you do plan to donate money, you still should not put in more than you're prepared to lose in the one-in-10 chance the project fails.
But, "the fact that failures seem to be distributed in non-predictable ways," and at such low rates, should offer comfort to potential backers, Mollick concludes.
So: Avoid donating to anything with too small, or too large, an ambition.
Rob covers business, economics and the environment for Fusion. He previously worked at Business Insider. He grew up in Chicago.