According to ReadWriteWeb it’s known as the marshmallow challenge. Small teams are given 18 minutes to build a free-standing structure made of dry spaghetti, one yard of string, one yard of tape and a marshmallow, which must be placed on top. The team wins by creating the tallest structure of all the groups participating.

What Tom Wujec discovered is that this simple game revealed some fascinating insights into how groups collaborate.

Wujec has conducted this experiment with over 70 groups of “students and designers and architects, even the CTOs of the Fortune 50,” he says. Most teams quickly break into roles and plan their structure, and then spend the remaining time building it before quickly and gingerly placing the marshmallow on top as time expires. More often than not, the structure pitifully fails as the marshmallow is added, leaving the team with a pile of spaghetti and no time to try again.

Wujec says that business school grads are taught to seek out and execute the one correct solution their challenge, while kindergartners practice the iterative prototype and refine process, much like the methods of lean startups. The kids would build, test and repeat until they found a structure that worked, and most times, he says, they built the tallest and most interesting structures.

What startups can take away from the marshmallow challenge is that bigger teams and higher incentives are no substitute for having the right skills and the right process in place. Wujec found that larger teams performed increasing worse than smaller teams, and incentivizing them with a reward did not make up for the fact that they were not using the right process.

Photo by Egilshay.