The Problem with Innovation Contests

In 1714, the British government offered a cash prize to anyone who could devise a practical method of determining longitude at sea. In more recent times, open innovation contests have tackled challenges ranging from the silly (Frito-Lay’s Crash the Super Bowl campaigns invited consumers to shoot a funny Doritos commercial) to the substantial (the NFL and Duke University have offered up to $2 million for football helmet designs that better prevent brain injuries). Such contests are designed to get the competitive juices flowing, and for that reason participants can often view others’ submissions while working on their own. But new research shows that if not managed carefully, that same competitive element can kill creativity. Across two real-world studies and three lab experiments, participants who were shown lots of competing ideas or primed to think about competition before beginning a creativity task performed worse than others. In one of the studies, conducted on the open innovation platform HYVE Crowd, 540 participants were tasked with generating ideas for pillows and retail marketing campaigns. They were shown varying numbers of sample ideas, presented as if they were submissions from other contestants (among them: “hair dryer pillows” to prevent people from sleeping with wet hair and “headphone pillows” to let people listen to music while falling asleep). Contestants in one group saw 10 ideas; the others saw just two. Experts judged the submissions according to their originality and usefulness. The participants who saw two ideas earned scores that were 4.4% higher, on average, than the scores of those who saw 10. That might seem like a small effect, but even a minor improvement can make a substantial difference in a large innovation investment, the researchers say. In the lab experiments, the researchers asked participants to come up with creative everyday uses for bricks—as doorstops or to measure right angles, say—and showed participants differing numbers of competing submissions. Independent coders scored the results. Here, too, the more ideas people saw, the lower the creativity of their own submissions, with the drop-off especially dramatic when 50 ideas were displayed. A follow-up study found that the performance of expert innovators deteriorated even more than that of novices. When participants were told that the ideas displayed were simply examples, however, and not actual submissions, they suffered no decrease in performance. “We looked at the number of ideas generated, at their innovativeness, and at other dimensions of creativity—and all were worse when competition was thrown into the mix,” says Reto Hofstetter, a marketing professor at the University of Lucerne, who led the research. “The competition at the heart of contests seems to undermine performance.” This surprised Hofstetter and his team. It runs counter to the well-established “recombinant growth theory” of creativity, which holds that seeing others’ ideas stimulates innovation because people can build on pieces of others’ work or put them together in new ways. For example, drugs such as insulin and penicillin improved over the years as innovators created new and better technologies to produce them. “If the First Entries You Show Are Poor, You’ll Probably Get Lots of Poor Entries” As CEO of the innovation consultancy HYVE, Johann Füller initiated the HYVE Crowd platform and has overseen more than 200 open innovation contests for corporate clients. He recently spoke with HBR about what the firm has learned about how competition affects contestants. Edited excerpts follow. Thomas Dashuber Do you agree that competition can inhibit participants? It’s always a question: How much competition do you want in a given contest? You can set the tone a bit through the incentives you offer. For example, many contests have only one winner. But if you pay out just for first place, you may demotivate the other participants. If one entry is so good that people feel they can’t compete, they have no motivation to engage. So we award prizes to the top solutions. We often offer a smaller, “most valuable” prize to the participant who provides the most assists by asking challenging questions or making helpful comments on other people’s ideas. How do you ensure an appropriate level of incentives? If the prizes are very high, the level of competition may be extraordinarily intense—not necessarily what we’re looking for. At a certain point, you get diminishing returns, and if the monetary aspect is too dominant, it may decrease other sources of motivation. Our users don’t participate just to win prizes; they participate out of curiosity, to showcase their talent, or to interact with like-minded people. Of course, there is competition, but there is also collaboration. You need a balance. If you make the prizes too high, contestants will stop collaborating and may even try to cheat. What is your rule of thumb for compensation? The prize should be in line with what a firm would pay a high-end innovation consultant. For a “grand challenge”—say, designing a faster sports car—people might spend years on the solution, so the prize should be substantial. For most of the contests we run, coming up with an entry takes a day or two. So we offer a few thousand dollars for first prize—roughly what a consultant would charge for a few days’ work. Do you limit the number of ideas participants can see? Sometimes. And sometimes we don’t show any ideas, because clients don’t want us to. But in my experience, if you don’t inspire some level of competition, people won’t be very motivated. The first couple of ideas that participants see are important in setting the stage. If the first entries you show are poor, you’ll probably get lots of poor entries. Ideally, people looking at the early submissions will see that if they want to win, they’ll have to contribute at a high level. What’s the biggest challenge in running these contests? Companies should be prepared to receive a lot of zany ideas. This can lead to fatigue—they throw all the ideas into a drawer and forget about them. I’ve seen things get overlooked only to be later adopted by someone else. We ran a contest in 2008 for a lighting company, and one of the participants suggested putting lights on pets. Crazy! Except now it’s common, especially for walking dogs at night. True innovation always seems outlandish at first. Several factors may explain why a heightened sense of competition can override the recombinant effect. First, intense competition can increase participants’ stress, which often impedes cognitive performance. Second, thinking about others who are vying for a prize diminishes participants’ belief that they can win it themselves, diluting their incentive. Third and most significant, contestants tend to view competitors’ submissions not as a source of inspiration but as a constraint on their own thinking. “The creative process requires disinhibition and a free-flowing stream of ideas,” Hofstetter says. “Seeing other contestants’ ideas jams up this process because people focus on differentiating their ideas from preceding ones rather than on using elements from them.” Companies can take several steps to avoid these pitfalls and raise the overall level of creativity that their innovation contests produce, the researchers say. They can: Limit the visible competition. Across all the studies, displaying only a few submissions rather than showing all of them kept participants highly motivated and reduced their sense of constraint. Show only the most original submissions. Paradoxically, the study participants were less constrained after seeing highly creative entries than after seeing more-mundane ones. Extremely original ideas are less likely to overlap with one another, the researchers say, so competitors exposed to them are less prone to feel that their own thoughts have already been explored. Group entries together. The competitive aspect can be further tempered by grouping submissions thematically rather than presenting them one by one (the common default). When 100 submissions were divided into five buckets, for instance, participants didn’t suffer much loss of creativity, because they perceived just five constraints, not 100. This tactic, along with the advice to show only the most original ideas, would add to a contest’s costs, the researchers acknowledge; sorting entries by quality and category is highly labor-intensive. But they predict that advances in machine learning, computer vision, and natural language processing could soon make it easy for computers to do that work. Gloss over the odds. Contest organizers can tone down the competitive nature of the language they use. Phrases such as “only the best idea wins” and “you compete against others” were particularly damaging to creativity among the contestants studied. Downplay submitters’ identities. Finally, an unpublished study by the researchers showed that drawing attention to other entrants by displaying their usernames or avatars alongside their submissions increased contestants’ anxiety—so organizers would be wise to omit such details, the researchers say. “Contests can be an effective way of generating fresh ideas,” Hofstetter concludes. “But our research shows that, counterintuitively, companies should turn down the competitive element if they want to generate the best results.” About the research: “ Constraining Ideas: How Seeing Ideas of Others Harms Creativity in Open Innovation ,” by Reto Hofstetter, Darren W. Dahl, Suleiman Aryobsei, and Andreas Herrmann (Journal of Marketing Research, 2021)

The Problem with Innovation Contests

Partner Center