Participants have 18 minutes to build a tower out of spaghetti sticks and marshmallows. This is important because the best way to learn is to try something quickly – “rapid prototyping” – and see what works, and see what needs to be improved

Group Size
  • Any Size
  • Marshmallows
  • Uncooked Spaghetti Sticks
  • String
  • Scissors
  • Measuring Tape


  1. Divide the room into teams of four people each. Every team should get one marshmallow, one pair of scissors, 20 spaghetti sticks, a meter of tape, and a meter of string.
  2. Tell everyone that they have 18 minutes to buld the tallest possible free-standing structure, using only the ingredients that have been given to them. There are just a few rules: The marshmallow has to be on the top of the structure. (In other words, they can’t put the marshmallow on the bottom, as the base of their creation.)They can’t tear up the marshmallow into smaller sizes. They have to use the full marshmallow. (They can however cut up/break the spaghetti sticks, the tape, or the string, if they wish.)They can’t tape the structure to the wall, or tie a string to hold it up. It must stand on its own at the end of 18 minutes.
  3. Once every team has their materials, you can begin the 18-minute countdown.
  4. You can announce, “5 minutes left” at the appropriate time. This heightens the pressure and gets people working more frantically.
  5. At the end of 18 minutes, call time. Everyone must cease working on their structures.
  6. You will go around and measure the height of each one that is still standing. (Often when we do this activity, none of the structures stand! They all collapse at the end of 18 minutes when we tell everyone to stop! That’s an important part of the debrief.) If there are any structures standing, you see which one is tallest, and you announce the winner!
  • It often helps to project a countdown timer onto a screen in front of the room. (You can do this easily with your computer. Just do an internet search for “countdown timer,” and you will find plenty of free resources that you can program to count down from 18 minutes.)
  • Even if you have the countdown timer, it’s always a good idea to announce milestones throughout the 18 minutes. For example, you can say at the 9-minute mark, “You have used half your time already!” This usually gets people working more frantically, especially because so many teams spend a lot of time just planning their structure, and haven’t start to construct anything yet.4.
Have a tip?

Submit your tip and we will review for some language about the process.


When you debrief this, you can begin by asking what went wrong (especially for groups who failed to get the structure to stand at all.) Why weren’t they able to solve this challenge? Some may mention the scarcity of time and resources, and the stress and pressure.

That’s when you can challenge them with a surprising fact: Ask them what group of people has consistently built the tallest structures. They may make a few guesses, but the correct answer is young children in kindergarten! They consistently outperform students at Harvard Business School, as well as students in other prestigious schools, and other intelligent adults in the working world. Why is this?

The participants can come up with their own answers. Here are some that you can contribute if nobody else comes up with them: It turns out that adults tend to spend much of their time planning and designing. This is normally helpful, but it backfires if some of the assumptions are wrong. For example, most adults assume that a marshmallow is fairly light, because it seems so fluffy and insubstantial. But when they place the marshmallow on top of a tower made out of spaghetti sticks at the end of 18 minutes, they see that the entire tower topples over!

Young children, by contrast, will just start building their structures – iterating right away. They might discover in the first minute that the marshmallow is heavy. Therefore, they will try a second attempt, and a third attempt, right away; they are constantly learning from their mistakes.

This is valuable in that it gets to the main idea of social innovation and entrepreneurship. The best way to see if your theory of change is correct is to test it out in the real world.

As Eric Ries, the author of the Lean Startup writes, you want to create a “minimal viable product” (MVP). You want people to create rapid prototypes: to figure out what works, and what needs to be improved.

This is the most effective way to learn, especially for social innovators and entrepreneurs. You will never know if your plans really will succeed if you spend all your time in planning and thinking. You need to test it out again and again, with what author Peter Sims calls “little bets.” These aren’t expensive tests where a failure will end in catastrophe. They are small attempts to test one part of your model. Sims gives the example of comedian Chris Rock. We see his comedy specials and think that he’s a genius, because every joke seems to be hilarious. But what we don’t see if all of Rock’s little bets; he goes to comedy clubs for months, testing hundreds of jokes and comedy routines to see which ones fall flat and which ones succeed. Then he takes only the best ones and puts them together in his final act.

Similarly all of the great entrepreneurial success stories have only come out of the wreckage of countless failures that we never see. For instance, Howard Schultz, the CEO of Starbucks, had many initial tests for his coffeehouse, which were disasters. Originally he had opera music in all the earliest pilots of Starbucks; he wanted it to be as Italian as possible. But customers didn’t like it at all, so he ditched it. Similarly, he also had waiters in tuxedos at the early coffeehouses. Again, this was a failed test that never made it. But he could never have been successful if he hadn’t tested out a lot of different variations to see what would work best. This is why in the “resilience” section of this manual, it’s so important to try so many attempts.


When we focus on the problem, it often paralyzes us. For example, Texas A & M professors David Jansson and Steven Smith had their engineering students tackle several design challenges. One of these was to design an inexpensive, spill-proof, disposable coffee cup. They showed one group of students an example of the problem that needed to be fixed: a coffee cup with a straw that not only spilled frequently, but also burned the users! The other randomly selected students were only told to design the ideal inexpensive, spill-proof, disposable coffee cup. The engineering students who saw the problem were 17 times as likely to fail as those who just focused on the ideal solution. They became fixated on the problem. They were trying to fix the straw for example. In the ideal solution, almost none of the students even used a straw! They came up with many other creative, far more effective designs.


Tom Wujec

Additional Readings

Jansson, David & Smith, Steven. (1991). Design fixation. Design Studies. 12: 3-11.)