Defining and Refining Your Prototype Plan

Learning Objectives:

• Define the prototyping phase of design thinking and explain the rapid prototyping process
• Practice developing critical questions and applying knowledge to create experiments

The famous “marshmallow tower” exercise asks groups to build a free-standing structure out of spaghetti, tape, string, and a marshmallow. The goal is to create a tall structure with the marshmallow on top.

How would you approach this problem? Quickly think through the beginning of the process: List one major assumption you would want to test, and one possible issue you foresee.

You may choose to attempt this optional exercise on your own, or with coworkers, friends, or family. You’ll need:

• 20 pieces of spaghetti
• 1 yard (3 feet) of masking tape
• 1 yard (3 feet) of string
• 1 regular-size marshmallow

You have 20 minutes to build the tallest possible tower you can. It must support a marshmallow at the top.

Most groups do not succeed at creating a standing structure out of spaghetti. Interestingly, elementary school students typically do better than business executives.

How would you have approached the marshmallow tower? Or how did you approach it? If you have encountered this challenge before, you know that one key assumption prevents most teams from succeeding—they assume that the marshmallow is light and that the spaghetti can support it easily. In reality, the marshmallow is dense relative to the spaghetti holding it.

Teams that spend most of their time sketching a design or discussing it often don’t realize this until it’s too late, and their tower collapses in the final minutes. The best way to design the tower is from the top down: start by testing if a small tower can support the marshmallow, then gradually add spaghetti to make the tower taller, observing if it remains stable. This process allows you to learn how to build the tower to support the marshmallow effectively.

The Marshmallow challenge highlights two key themes of prototyping:

1. Asking the right critical questions.

2. Designing quick experiments to find the answers.

For this reason, we define a prototype as a model for facilitating learning. On the concept poster, you listed critical questions and assumptions related to the innovation’s desirability, feasibility, and viability. The purpose of prototyping is to answer these questions.

Like the marshmallow tower, instead of attempting to construct one large pilot prototype near the end of development, you must begin by probing smaller but essential questions and testing them as quickly as possible.

Let’s transition to a different scenario: Before Netflix became a streaming service and media production company, it disrupted the video rental market by distributing DVDs through the mail.

To demonstrate how prototypes answer critical questions, let’s explore a question related to the feasibility of the Netflix model. Back when Netflix was just a concept for a rent-by-mail service, one critical question was: Can users safely receive and return DVDs through the mail?

What Netflix co-founders Marc Randolph and Reed Hastings did was simple: they took a standard envelope, slipped a CD into it, and mailed it. Two to three days later, the CD arrived intact, proving that other disks could survive the mailing process. This was a prototype that quickly answered one critical question. Most importantly, it likely took them only minutes to create.

Prototyping is an active and imaginative process. Once you’ve identified a critical question, the next step is to ask yourself: What is the fastest possible way to test this question? The Netflix experiment was successful because it focused clearly on one critical question. Instead of creating a finished envelope with perfect adhesive, branding, or other details, they aimed to answer just one thing: Will a disk break in the mail?

If the earliest prototypes succeed, you can then progress to more detailed versions. Testing too many questions at the same time, however, can cause you to overlook key assumptions. And, as with the marshmallow challenge, your concept may collapse if those assumptions are not addressed early.

Prototypes help us test critical questions and explore assumptions, as well as identify possible issues with ideas. The goal of prototyping is to learn more about a concept as quickly as possible, and preferably at a low cost.

When you are ready to prototype, you will have identified numerous assumptions and risks related to desirability, feasibility, and viability. The time period right before the launch of a new product, service, business model, or strategy is far too late in the process to begin identifying and testing these assumptions.

The traditional approach of “research, develop, launch” is what leads to spaghetti towers falling for lack of testing—and failed product launches.

Ideally, the development and testing cycle will occur numerous times before launch occurs.

These iterations may seem like a costly lengthening of the process, but it is much costlier to leave critical questions unaddressed.

The following picture is of Glass, formerly known as Google Glass: A smart device mounted on eyeglasses to provide a hands-free digital experience.

To learn how Google began testing some of the critical questions you identified, review the following excerpt from Tom Chi’s TED-Ed talk, “Rapid Prototyping Google Glass.”

My name is Tom Chi. I spent two years building the user experience team for the Google X division, a place I affectionately called the Department of Science Fiction because of the futuristic projects we tackled—self-driving cars, Google Glass, and more.

For those unfamiliar, Google Glass is a device that overlays digital elements onto your field of vision while keeping you engaged with the real world. Unlike looking at your phone, which isolates you, Google Glass allows you to stay present while accessing digital information.

Let me ask a simple question: How would you prototype this experience? How long do you think it would take to create the first working version of a heads-up display?

Answer: One day.

Here’s what we built. The magic piece was a coat hanger bent into a specific shape. The top loop went around your neck, the bottom rested on your chest, and it supported a piece of plexiglass with a sheet protector (the kind you use for book reports). This setup, combined with a pico projector and a netbook, allowed us to simulate digital overlays on the real world within a day.

Next challenge: How would you interact with this device, given there’s no mouse, keyboard, or touchscreen? Maybe something like the hand gestures in Minority Report?

How long to prototype this? Answer: 45 minutes.

We used two hairbands, fishing line, a chopstick, a binder clip, and a pen. The fishing line connected your hand movements to a clicker via tension, enabling forward and backward navigation. In under an hour, we could swipe through image galleries and emails with hand movements, mimicking the interaction experience.

Key insight: The prototype revealed the social and ergonomic awkwardness of this method, something we couldn’t have predicted just by thinking. This led to the second rule of prototyping: Doing is the best kind of thinking.

Last challenge: How do you make a heads-up display comfortable to wear?

Using basic materials—modeling wire, clay, and paper—I created a glasses model that simulated the weight of electronic components. This quick experiment taught us that weight perception depends largely on how it’s distributed. By placing weight behind the ears, we leveraged the ears as a fulcrum, reducing pressure on the nose.

This insight not only informed the design of Google Glass but also uncovered something fundamental about glasses in general. If your glasses feel heavy, gently push on the back—they’ll feel much lighter.

Lesson: Rapid prototyping with basic materials uncovers critical insights quickly, often revealing solutions you couldn’t predict through thought alone.

The prototypes for Google Glass are a great demonstration of Tom Chi’s three principles for rapid prototyping:

1. Find the quickest path to the experience.
2. Doing is the best kind of thinking.
3. Use materials that allow you to move at the speed of thought.

These principles encourage you to strategically plan prototypes and answer critical questions sooner than you might expect. Some realizations won’t arrive through empathy and imagination alone. In addition, rapid prototyping reduces your attachment to ideas that might not work.

Imagine that you work for a car company, and you are managing the development of touchscreen controls for the dashboard of new vehicles.

There are numerous critical questions to explore in the early development of such a concept. For example:

• Can drivers operate the touchscreen controls safely?
• What features do drivers actually want on the dashboard?
• How much traditional functionality (for example, physical buttons and dials) is still needed?
• Can people of different ages and sizes operate it?

One quick path would be attaching a tablet to the dashboard to test different functionalities and configurations. You could also conduct the same experiment in a driving simulator, which would replicate more dangerous conditions without risking personal safety or materials.