The Prettier Your Mockup, the Worse the Feedback You'll Get
The meeting started with everyone agreeing the prototype looked great. Then someone mentioned the button color. Then someone else thought the typography felt off. Forty minutes in, you hadn't touched the user flow, the navigation logic, or the information architecture — the things you actually needed to validate.
This is not a stakeholder problem. It's a fidelity problem. And it's been documented since 1996.
What Virzi, Sokolov & Karis Actually Found
At CHI 1996, researchers Randy Virzi, Jeffery Sokolov, and Demetrios Karis ran a study comparing usability testing across different prototype fidelity levels. The finding was clean: low-fidelity and high-fidelity prototypes surfaced substantially the same usability problems.
That's significant in one direction — low-fidelity isn't a shortcut that misses issues. But the more operationally important implication runs the other way: high-fidelity doesn't produce meaningfully better insight. You are spending design effort — hours, sometimes weeks — on a level of polish that generates no additional diagnostic value in usability research.
What high fidelity does produce differently is the character of feedback you receive.
The Anchoring Effect in Stakeholder Reviews
Anchoring is a cognitive bias where initial information disproportionately influences subsequent judgment. In design reviews, the first thing a non-designer perceives and can evaluate is visual aesthetics — color, typography, spacing, imagery. These are accessible. Everyone has opinions about them. Structure, navigation logic, and information hierarchy are harder to see and harder to articulate.
A polished prototype frontloads the accessible information. It trains attention toward aesthetics at the precise moment you need attention on structure. Stakeholders aren't being shallow — they're responding to what's visible and legible to them. When a design looks finished, feedback aligns to what finished designs get feedback on: surface details.
There's also a sunk cost mechanism. Polished work signals investment. Stakeholders who perceive that significant effort has already gone into a design feel social friction around suggesting fundamental changes. They'll recommend the button be darker or the layout be slightly different. They won't say "I think the entire information architecture is wrong" — not because they don't think it, but because it feels cruel to say it about something that clearly took a long time to build.
Rough work invites rougher critique. It signals that nothing is settled, that changes are welcome, that you're in exploration mode. That signal changes the quality of input you receive.
Design Fixation: How Polish Constrains the Designer Too
The fidelity problem doesn't only affect stakeholders. It affects the team doing the designing.
University of Michigan researchers Leahy, Daly, McKilligan, and Seifert published research in 2020 on design fixation — the phenomenon where exposure to existing solutions constrains subsequent creative thinking. The mechanism is simple: once you've developed a detailed representation of one direction, it anchors your cognitive model of the solution space. You're no longer exploring; you're refining.
High-fidelity prototypes accelerate fixation. The more time spent polishing a direction, the harder it is to abandon it. The design team has invested effort. The visual consistency has been established. The spacing grid has been calibrated. Walking it back requires dismantling work that felt, at the moment of creation, like progress.
Low-fidelity forces parallel exploration. When your first iteration is a rough sketch that took four minutes, you can generate three alternatives in twenty minutes and choose between substantially different approaches. When your first iteration is a two-week Figma build, you're committed to its bones before you've tested its logic.
What Non-Designers Actually See
There's a simpler version of this problem that doesn't require the full cognitive science.
Non-designers can only give feedback at the level of fidelity they can perceive. This isn't a limitation of their intelligence — it's a feature of what's accessible to someone who hasn't been trained to decompose visual design into its structural and aesthetic layers.
Show a non-designer a rough wireframe and they will engage with the structure. They will ask "why does this menu live here?" and "what happens when I click this?" because those are the only questions the sketch raises. The sketch has no button colors, so they don't comment on button colors.
Show the same person a polished mockup and they will engage with what they can evaluate. They will note that the blue feels cold or that the font seems small. They're not being difficult. They're answering the question the artifact appears to be asking: "how does this look?"
A PMC 2020 study on prototype types across different design contexts found that tangible low-fidelity mockups generated 77 percent useful structural feedback, compared to 65 percent for virtual high-fidelity designs. The polished version wasn't useless — it was just less efficient, and more likely to generate surface feedback that required additional filtering to find the signal.
Structure Before Style, Always
The practical implication is straightforward: match the fidelity of your prototype to the questions you're trying to answer.
If you're testing information architecture — what should be grouped, what should be reachable, what should be visible on first load — you don't need any visual fidelity at all. A sketch on paper with labeled boxes is sufficient. The Virzi study used paper prototypes. They found the same problems that polished digital prototypes found.
If you're testing visual design — brand expression, typography choices, color systems — you need fidelity. But you should be testing these in isolation, not bundled with structural questions. Structure reviews and aesthetic reviews require different cognitive modes from your stakeholders. Running them together optimizes neither.
The design system adoption research points to a related issue: design systems fail when teams adopt components without understanding the principles behind them. The same failure mode applies to prototype fidelity — teams reach for high-fidelity tools as default rather than asking what level of fidelity serves the specific research question.
The Brief Sketch as Research Methodology
There's a version of the fidelity conversation that frames low-fidelity as a shortcut — something you do when you don't have time to do it properly. This inverts the logic.
Low-fidelity prototyping is more methodologically rigorous for structural validation, not less. It controls for the variable you're not testing (visual design), which allows the variable you are testing (structure and logic) to produce clean signal. That's what good research method looks like.
The sketch isn't a compromise. It's the tool that produces the feedback you actually need before you've committed to a direction you can't easily reverse.
Show stakeholders rough work early and often. Show them polished work when the structural questions are settled. They'll give you better input in both phases — because you're asking the questions in the order that makes them answerable.
Cover photo by Anete Lusina via Pexels.