A prototype is not a product. It is a question in physical or visual form. You build a prototype to learn something specific, not to demonstrate how polished your design skills are.

This is the most misunderstood stage in design thinking. Teams consistently over-invest in prototypes, spending weeks on high-fidelity mockups when a paper sketch would have answered the same question in an afternoon. The rule of thumb: build the cheapest thing that will test your riskiest assumption.

Match prototype fidelity to the question you need to answer — not your available tools

The Purpose of Prototyping

After ideation, you have one to three promising concepts. But concepts are abstract. You need to make them tangible enough that real people can react to them.

A prototype serves three purposes:

Test assumptions. Every concept contains assumptions about user behavior, technical feasibility, and value. A prototype lets you test those assumptions before committing real resources.
Communicate ideas. A tangible artifact communicates a concept far more effectively than a verbal description. "Let me show you" beats "Let me explain" every time.
Reveal gaps. The act of building exposes questions you did not think to ask. "What happens when the user clicks back?" or "What if there are zero results?" These edge cases emerge naturally when you try to make something concrete.

Fidelity Levels

Fidelity refers to how closely the prototype resembles the final product. There is no "right" fidelity level. The right level depends on what you are trying to learn.

Low Fidelity: Paper and Sketches

Paper prototypes are hand-drawn screens or physical models made from cardboard, paper, and tape. They look rough on purpose.

Use low fidelity when:

You are testing the overall concept or flow, not specific interactions
You want users to focus on the idea rather than the visual design
You have multiple concepts to test and need to build all of them quickly
You are early in the process and expect significant changes

Low-fidelity prototypes have a hidden advantage: people give more honest feedback on rough work. When something looks polished, testers feel bad criticizing it. When something looks like it was sketched in five minutes, they feel free to say what they really think.

Medium Fidelity: Wireframes and Clickable Mockups

Wireframes are digital layouts that show structure and flow without visual design. Clickable mockups (using tools like Figma, Sketch, or even PowerPoint) allow users to tap through a flow.

Use medium fidelity when:

You are testing navigation, information architecture, or multi-step flows
You need stakeholders to understand the concept without explanation
You have validated the core concept and are refining the experience

High Fidelity: Realistic Mockups and Functional Prototypes

High-fidelity prototypes look and sometimes function like real products. They include visual design, real content, and interactive elements.

Use high fidelity when:

You are testing emotional responses and brand perception
Visual design is a core part of the value proposition (luxury products, creative tools)
You need to test with people who cannot distinguish between a prototype and a real product (some B2B stakeholders, consumer testing panels)
You are preparing for a final round of validation before development

For more on choosing the right approach and building quickly, see our Rapid Prototyping guide.

Prototyping Methods for Different Concepts

Digital Product Concepts

Paper sketch test: Draw 4 to 6 key screens on paper. Walk a user through the flow by swapping papers as they "tap" on elements.
Clickable prototype: Use Figma, Adobe XD, or similar tools to create linked screens. Users can tap through on a phone or computer.
Wizard of Oz: Build a front end that looks functional, but a human manually performs the backend operations. Useful for testing AI or algorithm-driven features before building the technology.

Service Concepts

Role play: Act out the service experience with team members playing the roles of service providers and customers. Surprisingly effective for revealing awkward moments and gaps.
Storyboard: Draw a comic-strip version of the service experience, showing key moments from the user's perspective.
Pilot: Run the service manually for a small group before investing in systems and processes.

Physical Product Concepts

Foam or cardboard model: Build a rough physical model to test ergonomics, size, and basic interaction.
3D print: For concepts where form factor and grip matter, a 3D print provides realistic enough feedback.
Functional breadboard: For products with electronic components, build the functional prototype separately from the form factor prototype. Test each independently.

The One-Question Rule

Before building any prototype, write down the single most important question it needs to answer. Not three questions. One.

"Will users understand the value proposition from the landing page?" is a question. "Will users click the signup button?" is a question. "Does the checkout flow feel trustworthy?" is a question.

When you try to answer multiple questions with one prototype, you end up with something too complex and too expensive, and the feedback you get is muddy. Build the simplest thing that answers your most critical question.

Common Mistakes

Over-investing in fidelity. If you are spending more than a few days on a prototype, you are probably building too much. The goal is to learn quickly, not to impress.

Falling in love with the prototype. Once you have invested effort in building something, it is psychologically hard to throw it away. But prototypes are disposable by design. If testing shows the concept does not work, the prototype has done its job by saving you from building the wrong thing at full scale.

Not prototyping the risky parts. Teams tend to prototype the parts they are most confident about. But the purpose of prototyping is to test uncertainty. Identify your riskiest assumption and build the prototype around that.

Skipping prototyping entirely. Some teams go straight from ideation to development. This almost always results in expensive rework because assumptions that seemed obvious turn out to be wrong when real users interact with the product.

What Comes Next

Take your prototype into the Test stage. Put it in front of the real people from your empathy research and watch what happens. Their reactions will tell you whether to refine, pivot, or move forward with confidence.