Prototype Fidelity
Prototype fidelity is the intentional degree of realism applied to selected dimensions of a prototype so it can answer a specific product-design question efficiently.
Download Markdown (.md)Soroush’s Product Wiki · https://wiki.soroush.uk/topic/prototype-fidelity/
01What it means
Prototype fidelity describes how closely a prototype represents the intended product, but it is not one ladder from rough to polished. Visual detail, content, interaction, data, breadth, technical behavior, and environmental context can each have different fidelity. A prototype should be realistic where the research question requires it and deliberately simplified elsewhere.[1][2][3]
- 01AbstractStructure or concept only
- 02SelectiveRealistic on the tested dimension
- 03RepresentativeKey behavior and content feel credible
- 04Production-likeHigh realism where evidence requires it
Fidelity is not one ladder: visual form, content, interaction, data, breadth, technical behavior, and context can sit at different points.
02Why it matters
Teams often equate visual polish with progress. That can make participants reluctant to criticize, encourage stakeholders to debate styling, and consume effort on details unrelated to the risk being tested. Choosing fidelity by dimension helps a team create convincing evidence without accidentally implying that every part of the concept has been resolved.
03When to use it
Make an explicit fidelity decision whenever a prototype will support research, technical exploration, alignment, or planning. Low visual fidelity may suit structure and concept comparison; realistic content and interaction may be essential for comprehension or task studies. High technical fidelity is warranted when latency, device constraints, or integration behavior is the question.
04How it works
- Write the decision or uncertainty the prototype must inform and identify what participants need to believe.
- Select fidelity separately for visual form, content, interaction, data, breadth, technical implementation, and context.
- Build the smallest representation that preserves the behavior or judgment under study.
- Explain simulated boundaries to observers without revealing the expected answer, and capture breakdowns caused by the prototype itself.
- Retire, revise, or increase fidelity only when the next learning question requires it.
05A practical example
A logistics team is testing whether dispatchers can resolve route conflicts. The prototype uses production-like shipment data and realistic drag behavior, but only three screens, neutral styling, and a simulated optimization service. Because data density and response timing drive the decision, those dimensions are convincing; brand polish and full navigation are intentionally absent.
06Where it breaks
A prototype always changes the conditions being observed. Missing content or latency can create false failures, while polished visuals can create false confidence. Mixed fidelity may confuse participants unless boundaries are clear. Even a realistic prototype does not demonstrate production reliability, accessibility, security, or maintainability.
Common mistakes
- Calling an entire prototype low- or high-fidelity without identifying which dimensions matter.
- Adding visual polish because it is easy while leaving critical content or system behavior unrealistic.
- Treating positive prototype performance as proof that the production product is ready.
07Sources
- What Do Prototypes Prototype?Stanford University HCI Group · Foundational paperWhat Do Prototypes Prototype?
- UX Prototypes: Low Fidelity vs. High FidelityNielsen Norman Group · Professional guidanceUX Prototypes: Low Fidelity vs. High Fidelity
- Unveiling the Multiple and Complex Faces of FidelityCambridge University Press · Peer-reviewed paperUnveiling the Multiple and Complex Faces of Fidelity