TED2010
Tom Wujec: Build a tower, build a team
Filmed:
Readability: 4.1
6,113,394 views
Tom Wujec presents some surprisingly deep research into the "marshmallow problem" -- a simple team-building exercise that involves dry spaghetti, one yard of tape and a marshmallow. Who can build the tallest tower with these ingredients? And why does a surprising group always beat the average?
Tom Wujec - Designer
Tom Wujec studies how we share and absorb information. He's an innovative practitioner of business visualization -- using design and technology to help groups solve problems and understand ideas. He is a Fellow at Autodesk. Full bio
Tom Wujec studies how we share and absorb information. He's an innovative practitioner of business visualization -- using design and technology to help groups solve problems and understand ideas. He is a Fellow at Autodesk. Full bio
Double-click the English transcript below to play the video.
00:16
Several years ago here at TED, Peter Skillman
0
1000
2000
00:18
introduced a design challenge
1
3000
2000
00:20
called the marshmallow challenge.
2
5000
2000
00:22
And the idea's pretty simple:
3
7000
2000
00:24
Teams of four have to build the tallest free-standing structure
4
9000
2000
00:26
out of 20 sticks of spaghetti,
5
11000
2000
00:28
one yard of tape, one yard of string
6
13000
2000
00:30
and a marshmallow.
7
15000
2000
00:32
The marshmallow has to be on top.
8
17000
2000
00:34
And, though it seems really simple, it's actually pretty hard
9
19000
3000
00:37
because it forces people
10
22000
2000
00:39
to collaborate very quickly.
11
24000
2000
00:41
And so, I thought this was an interesting idea,
12
26000
2000
00:43
and I incorporated it into a design workshop.
13
28000
3000
00:46
And it was a huge success.
14
31000
2000
00:48
And since then, I've conducted
15
33000
2000
00:50
about 70 design workshops across the world
16
35000
2000
00:52
with students and designers and architects,
17
37000
2000
00:54
even the CTOs of the Fortune 50,
18
39000
3000
00:57
and there's something about this exercise
19
42000
2000
00:59
that reveals very deep lessons
20
44000
2000
01:01
about the nature of collaboration,
21
46000
2000
01:03
and I'd like to share some of them with you.
22
48000
2000
01:05
So, normally, most people begin
23
50000
3000
01:08
by orienting themselves to the task.
24
53000
3000
01:11
They talk about it, they figure out what it's going to look like,
25
56000
3000
01:14
they jockey for power.
26
59000
2000
01:16
Then they spend some time planning, organizing,
27
61000
2000
01:18
they sketch and they lay out spaghetti.
28
63000
3000
01:21
They spend the majority of their time
29
66000
2000
01:23
assembling the sticks into ever-growing structures.
30
68000
3000
01:26
And then finally, just as they're running out of time,
31
71000
3000
01:29
someone takes out the marshmallow,
32
74000
2000
01:31
and then they gingerly put it on top,
33
76000
3000
01:34
and then they stand back, and -- ta-da! --
34
79000
3000
01:37
they admire their work.
35
82000
2000
01:39
But what really happens, most of the time,
36
84000
2000
01:41
is that the "ta-da" turns into an "uh-oh,"
37
86000
3000
01:44
because the weight of the marshmallow causes the entire structure
38
89000
2000
01:46
to buckle and to collapse.
39
91000
2000
01:48
So there are a number of people
40
93000
2000
01:50
who have a lot more "uh-oh" moments than others,
41
95000
3000
01:53
and among the worst are recent graduates of business school.
42
98000
3000
01:56
(Laughter)
43
101000
3000
01:59
They lie, they cheat, they get distracted
44
104000
3000
02:02
and they produce really lame structures.
45
107000
2000
02:04
And of course there are teams
46
109000
2000
02:06
that have a lot more "ta-da" structures,
47
111000
2000
02:08
and among the best are recent graduates of kindergarten.
48
113000
3000
02:11
(Laughter)
49
116000
2000
02:13
And it's pretty amazing.
50
118000
2000
02:15
As Peter tells us,
51
120000
2000
02:17
not only do they produce the tallest structures,
52
122000
2000
02:19
but they're the most interesting structures of them all.
53
124000
3000
02:22
So the question you want to ask is:
54
127000
2000
02:24
How come? Why? What is it about them?
55
129000
2000
02:26
And Peter likes to say that
56
131000
2000
02:28
none of the kids spend any time
57
133000
2000
02:30
trying to be CEO of Spaghetti, Inc. Right?
58
135000
3000
02:33
They don't spend time jockeying for power.
59
138000
2000
02:35
But there's another reason as well.
60
140000
2000
02:37
And the reason is that business students are trained
61
142000
2000
02:39
to find the single right plan, right?
62
144000
3000
02:42
And then they execute on it.
63
147000
2000
02:44
And then what happens is, when they put the marshmallow on the top,
64
149000
2000
02:46
they run out of time and what happens?
65
151000
2000
02:48
It's a crisis.
66
153000
2000
02:50
Sound familiar? Right.
67
155000
3000
02:53
What kindergarteners do differently
68
158000
2000
02:55
is that they start with the marshmallow,
69
160000
2000
02:57
and they build prototypes, successive prototypes,
70
162000
3000
03:00
always keeping the marshmallow on top,
71
165000
2000
03:02
so they have multiple times to fix when they build prototypes along the way.
72
167000
3000
03:05
Designers recognize this type of collaboration
73
170000
3000
03:08
as the essence of the iterative process.
74
173000
3000
03:11
And with each version, kids get instant feedback
75
176000
2000
03:13
about what works and what doesn't work.
76
178000
3000
03:16
So the capacity to play in prototype is really essential,
77
181000
3000
03:19
but let's look at how different teams perform.
78
184000
3000
03:22
So the average for most people is around 20 inches;
79
187000
3000
03:25
business schools students, about half of that;
80
190000
2000
03:27
lawyers, a little better, but not much better than that,
81
192000
3000
03:30
kindergarteners, better than most adults.
82
195000
2000
03:32
Who does the very best?
83
197000
2000
03:34
Architects and engineers, thankfully.
84
199000
3000
03:37
(Laughter)
85
202000
5000
03:42
Thirty-nine inches is the tallest structure I've seen.
86
207000
3000
03:45
And why is it? Because they understand triangles
87
210000
3000
03:48
and self-reinforcing geometrical patterns
88
213000
2000
03:50
are the key to building
89
215000
2000
03:52
stable structures.
90
217000
2000
03:54
So CEOs, a little bit better than average,
91
219000
3000
03:57
but here's where it gets interesting.
92
222000
2000
03:59
If you put you put an executive admin. on the team,
93
224000
2000
04:01
they get significantly better.
94
226000
2000
04:03
(Laughter)
95
228000
2000
04:06
It's incredible. You know, you look around, you go, "Oh, that team's going to win."
96
231000
2000
04:08
You can just tell beforehand. And why is that?
97
233000
2000
04:10
Because they have special skills
98
235000
2000
04:12
of facilitation.
99
237000
2000
04:14
They manage the process, they understand the process.
100
239000
2000
04:16
And any team who manages
101
241000
2000
04:18
and pays close attention to work
102
243000
3000
04:21
will significantly improve the team's performance.
103
246000
3000
04:24
Specialized skills and facilitation skills
104
249000
3000
04:27
are the combination that leads to strong success.
105
252000
3000
04:30
If you have 10 teams that typically perform,
106
255000
2000
04:32
you'll get maybe six or so that have standing structures.
107
257000
2000
04:34
And I tried something interesting.
108
259000
2000
04:36
I thought, let's up the ante, once.
109
261000
3000
04:39
So I offered a 10,000 dollar prize of software to the winning team.
110
264000
3000
04:42
So what do you think happened to these design students?
111
267000
3000
04:45
What was the result?
112
270000
2000
04:48
Here's what happened:
113
273000
2000
04:50
Not one team had a standing structure.
114
275000
3000
04:54
If anyone had built, say, a one inch structure,
115
279000
3000
04:57
they would have taken home the prize.
116
282000
2000
04:59
So, isn't that interesting? That high stakes
117
284000
2000
05:01
have a strong impact.
118
286000
2000
05:03
We did the exercise again with the same students.
119
288000
2000
05:05
What do you think happened then?
120
290000
2000
05:07
So now they understand the value of prototyping.
121
292000
3000
05:13
So the same team went from being the very worst
122
298000
2000
05:15
to being among the very best.
123
300000
2000
05:17
They produced the tallest structures in the least amount of time.
124
302000
3000
05:20
So there's deep lessons for us
125
305000
2000
05:22
about the nature of incentives and success.
126
307000
3000
05:25
So, you might ask: Why would anyone
127
310000
2000
05:27
actually spend time writing a marshmallow challenge?
128
312000
3000
05:30
And the reason is, I help create
129
315000
2000
05:32
digital tools and processes
130
317000
2000
05:34
to help teams build cars and video games
131
319000
2000
05:36
and visual effects.
132
321000
2000
05:38
And what the marshmallow challenge does
133
323000
2000
05:40
is it helps them identify the hidden assumptions.
134
325000
3000
05:43
Because, frankly,
135
328000
2000
05:45
every project has its own marshmallow, doesn't it?
136
330000
3000
05:48
The challenge provides a shared experience,
137
333000
2000
05:50
a common language,
138
335000
2000
05:52
a common stance to build the right prototype.
139
337000
2000
05:54
And so, this is the value of the experience,
140
339000
2000
05:56
of this so simple exercise.
141
341000
2000
05:58
And those of you who are interested
142
343000
2000
06:00
may want to go to MarshmallowChallenge.com.
143
345000
2000
06:02
It's a blog that you can look at
144
347000
2000
06:04
how to build the marshmallows.
145
349000
2000
06:06
There's step-by-step instructions on this.
146
351000
2000
06:08
There are crazy examples from around the world
147
353000
3000
06:11
of how people tweak and adjust the system.
148
356000
2000
06:13
There's world records that are on this as well.
149
358000
2000
06:15
And the fundamental lesson, I believe,
150
360000
2000
06:17
is that design truly is
151
362000
2000
06:19
a contact sport.
152
364000
2000
06:21
It demands that we bring all of our senses to the task,
153
366000
3000
06:24
and that we apply the very best of our thinking,
154
369000
2000
06:26
our feeling and our doing
155
371000
3000
06:29
to the challenge that we have at hand.
156
374000
2000
06:31
And sometimes, a little prototype of this experience
157
376000
3000
06:34
is all that it takes to turn us
158
379000
2000
06:36
from an "uh-oh" moment to a "ta-da" moment.
159
381000
2000
06:38
And that can make a big difference.
160
383000
2000
06:40
Thank you very much.
161
385000
2000
06:42
(Applause)
162
387000
2000
ABOUT THE SPEAKER
Tom Wujec - DesignerTom Wujec studies how we share and absorb information. He's an innovative practitioner of business visualization -- using design and technology to help groups solve problems and understand ideas. He is a Fellow at Autodesk.
Why you should listen
Tom Wujec is a Fellow at Autodesk, the makers of design software for engineers, filmmakers, designers. At Autodesk, he has worked on software including SketchBook Pro, PortfolioWall and Maya (which won an Academy Award for its contribution to the film industry). As a Fellow, he helps companies work in the emerging field of business visualization, the art of using images, sketches and infographics to help teams solve complex problems as a group.
He's the author of several books, including Five-Star Mind: Games and Puzzles to Stimulate Your Creativity and Imagination.
Tom Wujec | Speaker | TED.com