Every Project Has Its Marshmallow

In yet another fabulously stimulating TED talk (about 7 min), Tom Wujec details a team building exercise where groups of people are challenged to build the tallest structures they can out of 20 sticks of spaghetti, tape, string, and a single marshmallow that has to go on top. He says that most teams spend time planning, jockeying for power in the group, building the structure, and then at the last moment they put the marshmallow on top, only to have the weight of the marshmallow cause their structure to collapse. 

This exercise demonstrates the hidden assumptions of any project, and all projects, he says, have their marshmallow. Interestingly, it is also the kindergartners who excel at this exercise more than most, and the MBAs who do the worst. The kindergartners do well because they use trial and error, including the marshmallow the entire time, rather than only designing one structure that creates a crisis when it fails at the last minute. 

In my own projects, I have noticed that I will plan it out, spend a lot of time designing and planning and breaking things down into steps, and then there will be that nagging sense of something else, something I am missing. Now I have a name for it: The Marshmallow. I have had more projects become frustrations if not failures or fiascos because of those damned Marshmallows. 

More importantly, though, I now have a plan for how to deal with the Marshmallow. Iteration. Do it, and then do it again, and keep doing it, all the way through completion, and learn from what does not work. At the end of 2008, I looked back on my year and dubbed it the Year of the Failures. (It was not crushing. In fact, realizing it helped me to understand my intense feelings of frustration at the time.) And if 2008 was about failure, then 2009 was the Year of the Fiascoes. (Forget failure, I was operating on a much grander scale!)

Now, with the Marshmallow Paradigm, these were not Failures and Fiascoes. They were Iterations. And the thing with Iterations is, the learning curve can be nice and steep and eventually lead somewhere. That's what I'm talking about.