Test — let users break it

Put the prototype in front of real users and treat their confusion as data.

Why it works

Testing closes the loop by replacing your prediction of how people will react with their actual behavior. Because you built cheaply, you can afford to be wrong and iterate. The mechanism is disconfirmation: you are actively hunting for where reality contradicts your model, not seeking applause.

How to do it

1. Watch users use it without explaining or rescuing them; their struggle is the finding.
2. Ask what they expected to happen at each step, not whether they "liked" it.
3. Feed the surprises back into Define or Prototype — testing is a loop, not an ending.

Evidence

Mechanistic. Usability testing with a handful of users reliably surfaces most major problems; the broader principle of iterating on real feedback is well established across disciplines. The "small number of users finds most problems" claim traces to a specific mathematical model of problem discovery and to empirical work showing diminishing returns after roughly five test users. (mechanistic)

The famous "5 users finds 85% of problems" figure is a usability heuristic, not a hard law; treat it as directional.

Sources

• Nielsen, J., & Landauer, T. K. (1993). A Mathematical Model of the Finding of Usability Problems. Proceedings of the INTERACT '93 and CHI '93 Conference on Human Factors in Computing Systems, 206-213.
• Virzi, R. A. (1992). Refining the Test Phase of Usability Evaluation: How Many Subjects Is Enough? Human Factors, 34(4), 457-468.

Common mistake

Fishing for validation — asking "do you like it?" and counting polite yeses as success, instead of watching where users actually get stuck.