Marsmallows and Soap: Why failure is a great thing!

There’s a great team exercise I heard about years ago. Take teams of 4 people and give them the following items:

  • a regular marshmallow
  • 20 pieces of uncooked spaghetti
  • 36” of string
  • 36” of scotch tape

Now give them 18 minutes to build the tallest freestanding structure they can with the marshmallow being on the very top. At the end, measure the results and award the winning team!

What happens when you give this task to different groups of people is very interesting and teaches us a lot about business principles. For starters, the group of people that consistently perform the worst are recent business school graduates! It turns out they spend the first minutes acclimating themselves to the process, figuring out who is in charge and what role everyone will play on the team. They spend a good deal of time coming up with a plan and design and then, finally start to build. Usually they find that when they go to put the marshmallow on top, CRASH!  Now they are short on time and find themselves in a panic!

One of the best performing groups turns out to be Kindergarteners. Not only do they produce some of the tallest structures, but they have the most creative designs. Tom Wujec, who has run hundreds of these challenges, says that instead of the kids trying to figure out who gets to be CEO of Spaghetti INC, they jump right in and start building prototypes. By the time our business graduates have come up with a “solid plan” our 5 year olds have been through multiple designs and have a much better understanding of what it takes to succeed. They end up building structures nearly two and a half times larger than our business graduates simply through “trial and error” and revisions of their designs.

Years ago, Unilever was developing a process for making powdered soap. Spray liquid detergent through a nozzle at high pressure and it will dry into a powder than can be boxed and sold. The design of that nozzle ends up being very critical to the process so Unilever hired experts in fluid dynamics and mathematics to design the nozzle. After lots of work and calculations they figured out the magic formula and built the perfect nozzle, only it didn’t work!

The complexity of the nozzle was so high the experts were unable to come up with a design no matter how hard they tried. What Unilever did to solve their problem was to create 10 random nozzles and test them all. They took the best one and made 10 variations on that one, tested them, and then took 10 variations of the best of those. 45 generations later they had a nozzle design that worked perfectly and no one had the slightest idea why!

Unilever created a systematic process of trial and error. It turns out top companies do this all the time. Global sales companies experiment with the colors on their websites and measure total sales to find out what colors patterns are more likely to get you to buy online. Retail stores test different layouts to see which ones increase the chances of impulse buys. Marketing companies use different advertising language in different markets and measure success rates to craft the perfect campaign. These companies don’t know why these things work, only that they do.

A lot of emphasis is put into the planning stage of a business or new project but don’t overlook the importance of trying ideas and letting them fail. Too many “entrepreneurs” never get off the ground because they can’t get out of the planning stages. What if we can find ways to start smaller, test our ideas and refine our designs? Maybe we need to jump into things even before we feel “ready” and learn some lessons along the way. They key is to actually LEARN from your failures and make adjustments that offer improvement!

The Marshmallow Challenge –
Tim Hartman’s TED talk on the God Complex and the importance of trial and error –


Never miss an article! Receive regular weekly updates and unsubscribe at any time.
We hate spam. Your email address will not be sold or shared with anyone else.

Leave a Reply

Your email address will not be published. Required fields are marked *