Kromatic logo

The Real Startup Book

Evidence-based guides and research methods from Kromatic

Product Prototyping

A product team arranges paper sketches, a clickable prototype on a tablet, and a 3D-printed model along a low-to-high fidelity spectrum

In Brief

Product prototyping is a family of techniques for building representations of a product at varying levels of fidelity — from paper sketches that take minutes to create, to full-scale functional models that take months. The purpose of every prototype is the same: to answer a specific question about your product with the minimum investment necessary. The critical insight is to start at the lowest fidelity that can answer your current question, and increase fidelity only when lower-fidelity methods have confirmed you are heading in the right direction.

Prototyping is an evaluative method. Unlike generative techniques that explore what to build, prototyping tests whether a specific solution works — whether users can navigate it, whether the form factor feels right, whether the core feature delivers enough value to retain users.

Common Use Case

When you have a candidate solution and you want to test whether it works before committing engineering or manufacturing budget. You want answers about flow, form factor, or core-feature pull, and you want them at the lowest fidelity that can give you a real signal — paper before clickable, clickable before single-feature MVP.

Helps Answer

  • Can users accomplish key tasks with this design?
  • Does the physical form factor work for the intended use?
  • Is a single core feature valuable enough to retain users?
  • Does the full experience hold together when assembled from existing tools?
  • Are there usability problems that need to be resolved before launch?
  • Does the product work under realistic conditions?
QuantitativeEvaluativeProduct

Description

The fundamental mistake teams make with prototyping is starting at too high a fidelity level. They jump to a clickable prototype or a functional MVP before confirming that the basic concept, flow, and value proposition work. This wastes weeks or months on polished execution of a flawed concept, when a paper prototype shown to 5 users in an afternoon would have revealed the core problem.

The golden rule: never increase fidelity until the current level has confirmed your direction. If users cannot follow the basic flow on paper, building a clickable prototype will not fix the problem — it will just make the problem more expensive.

Choosing the Right Prototype

Pick the lowest fidelity that can answer your current question. The six methods below are grouped by fidelity from low to high. Move up the ladder only after the lower rung confirms your direction.

AI Prompt

I need to prototype [PRODUCT / FEATURE — ONE SENTENCE].

The riskiest assumption I want to test:  [WHAT YOU NEED TO LEARN]
Form factor:                              [DIGITAL / PHYSICAL / SERVICE]
Budget I can spend on this prototype:     [DOLLARS]
Time I can spend:                         [HOURS / DAYS]
Where I am in the fidelity ladder:        [JUST STARTED / PAPER PASSED / CLICKABLE PASSED / ...]
[CONSTRAINTS]:                            [REGULATED INDUSTRY, REMOTE TESTING ONLY, B2B vs B2C, etc.]

Recommend 1-2 prototyping methods from this family that fit my situation. For each:
1. Name the method and why its fidelity matches the question.
2. Name the riskiest assumption it would NOT validate.
3. Suggest the next-higher fidelity rung if this one passes.
Low Fidelity
  • Paper Prototype Hand-drawn screens or physical mockups that users walk through with a facilitator acting as the “computer.” Best used as the first round when the basic flow is still uncertain and you want answers in an afternoon.
Medium Fidelity
  • 3D Print A physical model customers can hold and examine to evaluate form factor and ergonomics. Best used when the open question is whether the physical form feels right before any manufacturing investment.
  • Clickable Prototype Interactive digital screens connected by click targets, with no real backend. Best used when users need to navigate detailed interactions and you want to run usability testing without writing code.
High Fidelity
  • Single Feature MVP A limited product version that delivers one core feature exceptionally well. Best used when the question is whether that feature alone provides enough value to retain users.
  • Mash-up A functional product assembled from existing third-party tools and services, capable of serving real customers without custom development. Best used when you need to test the full end-to-end experience before investing in a custom build.
  • Life-Sized Prototype A full-scale prototype with working mechanisms and real materials. Best used when full real-world performance is the only remaining unknown — and only after lower-fidelity tests have confirmed direction.

Cost-Reduction Techniques

Prototyping costs can escalate quickly at higher fidelity levels. Several strategies cut cost without cutting learning:

  • Pretend to own. Instead of buying equipment, materials, or space, borrow or rent. Use a coworking space’s 3D printer, rent a commercial kitchen for a food product, borrow specialized tools from a makerspace. Cuts cost 50 to 90 percent at every fidelity level.
  • Prototype only the critical path. Do not build every screen, feature, or component. Focus on the 2 to 3 tasks that test your riskiest assumption.
  • Use existing assets. Stock photos, open-source UI kits, template services, and pre-built components dramatically reduce time and cost at medium fidelity.
  • Time-box ruthlessly. Fix the time, stop when it expires. A prototype 80 percent finished in 2 days teaches nearly as much as one 100 percent finished in 2 weeks.