Prototype the Argument Instead of Winning It

Design teams get stuck on interaction model debates more than almost anything else. Not because the problem is hard, but because the debate is verbal. Two people hold two mental models, neither has authority over the other, and every round of discussion makes them more invested in their own position.

The meeting ends without a decision. The next meeting starts where the last one ended.

There is a faster way out: stop arguing and build both.

The Deadlock

On a companion app for a medical device, the setup flow needed a structural decision. The team had two camps. One wanted a multi-step wizard: guide the user through one piece at a time, one screen per task, a clear sense of progress. The other wanted a single-page layout: all configuration visible at once, no navigation between steps, the full picture available at any moment.

Both were defensible. Wizards reduce cognitive load on first use and prevent users from missing required fields. Single-page layouts preserve context and let experienced users navigate freely. Neither argument was wrong.

Nobody had the authority to call it unilaterally. And because the debate was verbal, it kept producing more words instead of more clarity.

What Prototyping Both Actually Does

I prototyped both across five wireframe iterations and brought them to the team.

The result was not a vote or a majority opinion. It was a shared observation: this particular setup flow is used infrequently.

A user might complete it once at a pharmacy, then not touch it again for months. When they return, they need to understand what they are looking at immediately. The single-page layout made the full configuration visible on arrival. The wizard had already moved that information off-screen.

The deciding question was not "which pattern is better?" It was "what is the user trying to hold in their head when they come back to this?"

For a rare, high-stakes task, context is more valuable than step reduction. A wizard optimizes for first-time guided completion. A single-page layout optimizes for re-orientation on return. Given the actual usage pattern, that distinction settled the debate.

Both team members aligned. Not because one persuaded the other, but because the prototype made the question concrete.

Why Verbal Debates Calcify

Mental models shared in words are almost impossible to resolve. When you describe an interaction verbally, the listener constructs their own version of what you are describing.

Your wizard and their wizard are not the same wizard. Your single-page form and theirs have different layouts, different hierarchies, different visual weights. You are not actually discussing the same thing.

So you talk past each other, and each round of discussion requires defending your mental model against an attack you cannot fully see. That defense becomes personal. Positions harden before anyone has looked at the same thing.

A prototype puts both people in front of the same artifact. Now you are both reacting to the same thing. Disagreements become about what you are actually looking at, not about competing abstractions. That is a much more tractable problem.

The Condition That Makes It Work

This only works if you build both options honestly. A prototype that is clearly a strawman of the option you oppose does not produce evidence. It produces a setup for the answer you already wanted.

The test is whether the person who advocated for the losing option would say the prototype represented their idea fairly. If they would not, you have not replaced the debate. You have just moved it downstream.

Honest prototyping requires holding both options with equal care while you build them. This is uncomfortable if you have a preference. It is also the only way to generate evidence that actually ends the conversation.

What This Replaced

Before this project, I would have tried to win the debate through better argument. A cleaner rationale, a stronger analogy, a reference to a design principle or a case study from somewhere else. That approach has a ceiling. Smart people disagree with smart arguments all the time.

The shift was realizing that the goal is not to win the argument. The goal is to dissolve it by making the question answerable through something other than opinion. When you build both options, you replace the argument with a shared observation.

You still need the observation to be correct. The prototype test does not guarantee the right outcome. But it makes the conversation about the product instead of about the people having the conversation, and that is almost always an improvement.