If at First You Don’t Succeed… Iterate Like Crazy


“Humility and design go hand-in-hand; it’s just in the nature of design.” – Austin Knight, Google Designer 1

You probably think you can beat a kindergartener at just about anything. Half-court basketball? No doubt. Historical trivia? Sure thing. Barbecue cook-off? Fuhgeddaboudit.

How about building the highest spaghetti tower that can hold up a marshmallow? Well, prepare for a slice of humble pie to go with that marshmallow, because I have some bad news: kindergarteners typically beat post-graduate students when given twenty minutes to build a tower out of spaghetti, tape, and string to support a marshmallow on top.2 That’s right–kids who can’t tie their own shoes outperform adults who can drive and vote.

How can this be?

It turns out that the two groups approach the problem very differently. The adults spend most of their time developing a plan and only a few minutes executing on it, while kindergarteners generally start building right way. The children experiment, learn what works and what fails, and then refine their techniques accordingly. In the end, they come up with more successful (and more interesting) towers.

We can learn a lot about designing from these kids.

Iterative Design

Whether we’re building a new sleep tracker, a mindfulness app, or a spaghetti tower, people tend to invest in (and get attached to) a thorough plan before actually building anything. But this approach is loaded with major assumptions about how end users will actually behave when interacting with our concepts (and remember, humans are irrational and make decisions primarily based on emotion, not strict logic).

So instead of acting like artists, who create sculptures that reflect the amazing images in our heads, we need to act like scientists, who assume our initial hypotheses have fatal flaws and are only starting points. Therefore, we need a process that allows us to gather feedback along the way–to test and retest and change course as needed.3 The process looks like this:

  1. Discover (Do everything you can to understand the problem space and make your first solution hypothesis.)
  2. Create (A description, a drawing, a prototype, actual product changes – something.)
  3. Test (Anything from asking for feedback to actual product metrics.)
  4. Refine (Take what you learned and make another hypothesis to test.)
  5. Repeat (Vigorously! Until you’ve proven you have a real solution that works for real people in the real world).

It seems simple enough. But there is one other component that is 100% required for this to work.

The iterative design process only works if you, the designer, check your ego. After all, your favorite new innovation might be the first thing on the chopping block. You think your new VR fitness app will entice users to stay in shape while jogging along virtual beaches, but user testing feedback says no one wants to feel dizzy AND risk jogging right into a wall. Overconfidence and bias blindness will tell you that they lack vision and understanding, and just don’t comprehend the brilliance of what you’ve created. Humbleness helps you recognize that maybe they’re right, and you need to tweak a few things.4 Arrogance tells you to knock down the six-year-olds’ spaghetti towers to show them who’s boss, while humbleness helps you realize that the kids’ methods might just be innovative and valuable.

Behavioral design requires us to iterate, and being humble allows us to accept what each iteration teaches us. It works, too–the proof is in the pudding, or in this case, in the marshmallows.

Good Design is Humble, austinknight.com, 2016
Why Kindergarteners Always Win the Marshmallow Challenge, Toffler Associates, 2016
What is Iterative Design? (and Why You Should Use It), Enginess, 2018
Humble Beginnings: Current Trends, State Perspectives, and Hallmarks of Humility, Social and Personality Psychology Compass, 2013

Photo Courtesy   via Gratisography. Thank you Ryan!

Scroll to Top