A man spinning in a large metal ring gradually transforms--before your very eyes!--into a werewolf. How cool is that?

Circus-act creators loved the monstrous metamorphosis. Audiences agreed with them. Circus managers were less impressed. The lesson for leaders outside the big top: When choosing among innovations to pursue, seek help from your  engineers, your designers, your R&D teams.

"Creators have all this untapped wisdom built up about their peers' ideas that is being underutilized," says Stanford professor Justin Berg, who studies what makes people good at "creative forecasting"--predicting the success of new ideas. The circus arts provided an ideal context for Berg's research "because the entire domain is about novelty," he says. "But people do have expectations of what they want to see in circus shows. So it's really difficult to know what to put on stage."

Business thinkers like to argue the relative importance of innovation and execution to a company's success. The red-haired stepchild in this debate is idea selection. The more ideas generated by a business, the better its chances are of producing a winner. But resources are more constrained than creativity. So at some point the business has to place a bet.

That's tough to do given that diamonds in the rough look a lot like quartz. Many hugely successful products (the Harry Potter books, digital cameras, the bagless vacuum cleaner) traveled rocky roads. Meanwhile market debacles (the DeLorean, those fat-free chips that caused diarrhea) were hailed as breakthroughs.

"Our companies, communities, and countries don't necessarily suffer from a shortage of novel ideas," writes Wharton professor Adam Grant in his new book, Originals: How Non-Conformists Move the World. "They're constrained by a shortage of people who excel at choosing the right novel ideas."

In Berg's study, a set of "creators"--more than 150 acrobats, jugglers, tightrope walkers, and the like--submitted videos of their acts, which were viewed by managers and fellow creators, also in the industry (339 participants in total). The managers and creators were asked to predict how successful the videos would be with a general audience; creators evaluated both their own acts and those of their peers. Berg then assessed the accuracy of these predictions by inviting more than 13,000 people to view the videos, and tracking how much each video was liked, shared, and financially supported by the audience. (Berg collaborated with James Tanabe, a former artistic director at Cirque du Soleil; and Lena Gutschank, a veteran circus artist.)

Berg found:

  • Creators are more accurate than managers at judging other creators' ideas. They are less likely to pass over the circus version of Harry Potter. By contrast, managers assume, on average, that conventional ideas are better than novel ideas.
  • Creators are less accurate than managers when judging their own ideas. Over-confidence rules.
  • Creators who have been successful with their own ideas in the past are less accurate when judging other's ideas. If they succeeded despite the poor quality of their ideas, then their accuracy is even lower.

That last one "is my favorite finding," says Berg. Some videos went viral and garnered thousands of clicks, even though when audience members intentionally compared them to other videos they fared poorly. "They think that their act is great but it is actually not, so they have gotten misleading feedback," says Berg. "Then they look for aspects of their ideas in other people's acts. So it messes with their taste. And it messes with their evaluative accuracy."

Berg offered advice to entrepreneurs trying to distinguish between Seinfelds (the show barely made it on the air) and Segways (the Blart-mobile was once touted as the next big thing):

Never stop being a creator. Business demands will drag you toward a managerial role--but don't lose touch with the innovator within.

Enlist creators in selection decisions. Have them vote or establish another feedback mechanism.

Keep managers engaged in creative work.  It will sharpen their novelty-assessment skills.

Know the difference between luck and skill. Don't evaluate every idea based on its similarity to ideas you personally succeeded with. Some of that success may have been serendipitous.

Don't fall in love with novelty. There are bad novel ideas and great conventional ideas. Prefer the latter.