← /notes

Prototyping

Created Dec 23, 2024 designcraftmaking

A drawing lies. It shows what you intend, not what you’ll get. The prototype speaks truth: this is how large it actually feels, how heavy it actually is, where your hand actually goes. The gap between representation and reality closes only through making.

David Kelley at IDEO institutionalized this as “fail fast, fail often.” Build a rough version in hours, not weeks. Put it in someone’s hands. Watch what happens. The information you gain from a crude prototype exceeds what you learn from months of analysis. Drawings describe intentions; prototypes reveal consequences.


Prototypes serve different purposes at different stages. Appearance prototypes test looks — proportions, surface finish, color. Functional prototypes test mechanism — does it actually work? Integration prototypes combine both — close to production, testing everything at once. Early prototypes should be rough and cheap; late prototypes should be close to final.

The Japanese concept of genchi genbutsu — “go and see” — applies here. Toyota engineers test designs by making and handling physical parts, not just reviewing simulations. The body knows things the mind doesn’t. A part that looks right on screen may feel wrong in the hand. You can’t iterate on sensation without sensation.


The temptation is to skip prototyping because you’re sure you’re right. This is always wrong. Every experienced designer has stories of the thing that worked perfectly on paper and failed immediately in reality. The chair that was uncomfortable. The handle that was awkward. The mechanism that jammed. The prototype would have caught it.

Prototyping is thinking with your hands. Each version is a question: what happens if? The material answers. Your job is to listen and iterate. The final design emerges from conversation, not prescription.

Related: craft, feedback loops, constraints, tacit knowledge, design