The Problem with Showing Your Work Too Well

Author: Caroline Petersen | Editor: Steve Korver | Imagery: Anastasiia Hriaznova

Reading time: 8 min read

Quick summary: Rough visuals, sketches, wireframes, and animatics elicit better and faster feedback than verbal descriptions because people can react to something concrete rather than something abstract. But how finished a thing looks is itself a message. Design research is clear that polish shapes what a viewer thinks they are being asked to do, and the risk is that stakeholders, particularly technical experts unfamiliar with the iterative creative process, cannot tell a rough concept from a final proposal once it is visible. The fix isn't less visualization. It's stating the stage of the work before you show it.

A few months ago, the team and I worked on a project for a client who was highly technical but had limited marketing experience. The client wanted us to explain a complex service offering, and the deliverable was a video.

Before building the final cut, we built an animatic: a rough, unpolished sequence used to test the concept and message before committing to full production. That is the animatic's role in the production pipeline. It exists to be criticized cheaply, before the expensive parts are locked in.

The client watched it and reacted as though it were the finished product.

Placeholder pacing and unfinished transitions, deliberately left in, were read as errors rather than as signals of an early stage. The expectation, once anything landed in the client's inbox, was that it needed to be complete, not close. Complete.

This created a specific problem. The client had real subject-matter expertise we lacked, and some of their notes were necessary and correct. But because they were reacting to a rough cut as if it were final, it became difficult to separate two very different kinds of feedback: “the concept is wrong” and “this frame is not yet polished.”

What the research actually says

This pattern is not unique to a single client. It has a long history in design and human-computer interaction research under the heading of “prototype fidelity”.

Marc Rettig's 1994 argument for paper prototyping, still one of the most cited pieces on the subject, is practitioner reasoning rather than experiment. His case against high-fidelity prototypes is that they take too long to build, resist change once built, and pull reviewers toward commenting on appearance instead of substance.

This reasonable advice is complicated by the experimental record. Virzi, Sokolov, and Karis (1996) conducted two studies comparing paper prototypes with working products and found substantially the same usability problems in both conditions, detected by similar proportions of participants. Rough prototypes did not hide problems. They concluded that low-fidelity prototypes remain useful late in development, not only at the beginning.

Sauer and Sonderegger (2009) found that fidelity and visual polish shape how people evaluate what they are shown. More attractive designs were rated as more usable. Lower-fidelity prototypes prompted what they call deficiency compensation: users mentally fill in the missing details, which is exactly what a well-briefed reviewer should do and what an unbriefed one does poorly.

Bill Buxton names the mechanism directly in his classic Sketching User Experiences (2007). A sketch and a prototype serve different cognitive functions: a sketch invites exploration, while a prototype invites evaluation. Conflating them leads teams to make decisions before the concept has been tested.

The reaction to a visual is so fast and so strong for a reason. The Heath brothers' Made to Stick: Why Some Ideas Survive and Others Die (2007) offers a useful way to describe the mechanism: concreteness is one of six properties that make an idea “stick,” because concrete information is easier for the brain to hold, recall, and manipulate than abstract description. However, the same concreteness that makes an idea memorable also makes it feel finished before it is.

Why this matters more with non-designers

Our client wasn't being difficult. Technical subject-matter experts are frequently more literal readers of visual fidelity than marketing-native stakeholders, precisely because their training rewards precision and completeness as signals of correctness. In engineering, science, and technical fields, a rough draft that looks unfinished usually is unfinished, and treating it otherwise is a liability. That instinct, correct in their domain, works against them when applied to a creative process built around iteration.

This is a pattern we see consistently across technology, cybersecurity, and federal clients: stakeholders with deep domain expertise and comparatively less exposure to creative development cycles are the group most likely to mistake a rough concept for a final proposal.

The same instinct, lower stakes

The same pattern shows up inside a team, where it is easier to see because it repeats weekly. It also costs almost nothing to fix. A team member explains, verbally, how they handle out-of-office coverage or how a task moves through review. The explanation is accurate. It is also hard to follow, because the listener has to build a mental model of something they cannot see and keep updating it as the description continues. The correction is almost always the same: screenshare it.

The research here is older and more settled than the work on prototype fidelity. Allan Paivio's dual coding theory (1971, 1986) holds that verbal and visual information travel through separate but connected cognitive systems, and that information encoded both ways is retained better than information encoded one way.

Richard Mayer's Multimedia Learning (2001, 2009) demonstrated this principle across dozens of controlled studies and called it the multimedia principle: people learn more deeply from words and pictures together than from words alone. A description asks the listener to construct the thing before they can assess it. Showing it hands them the construction and frees their attention for the assessment. If the thing you are describing already exists somewhere you can point to, point to it. Talking about it is the harder path for everyone in the room.

The fix is calling the stage

The solution is not to visualize less, or to visualize only once ideas are fully formed. Visualization still produces faster, sharper, more specific feedback than verbal description, and the research above supports that at every fidelity level. The fix is to make the fidelity signal explicit rather than implicit.

In practice, that means:

• Call the stage. “This is a rough animatic built to test the concept, not the finished cut” recalibrates what kind of reaction the room should give.
  
  
• Match visual roughness to the intended reaction. A hand-drawn sketch invites structural critique. A polished mockup invites approval or rejection. Choosing the wrong fidelity for the stage sends an unintended signal.
  
  
• Ask a specific, structural question rather than a general one. “Does this framework hold together?” draws a different kind of answer than “Thoughts?”
  
  
• Reserve full polish for the moment a genuine yes-or-no decision is needed, since polish is an evaluation-stage signal, not an exploration-stage one.

High fidelity early is sometimes the correct call. When the decision is genuinely aesthetic, when you are buying credibility with a skeptical exec, or when a stakeholder has shown they cannot read a sketch, the polished version is the honest one.

Fidelity used to be expensive

The above advice largely assumes it's okay for rough work to look rough. That world is closing.

Generative tools have not primarily made concepting faster, though they have certainly done that. What they have done is collapse the cost of looking finished. A first-pass script now arrives in clean prose with no seams. A layout concept arrives rendered, kerned, and color-corrected. A storyboard arrives as something closer to a finished frame. The output of ten minutes of prompting is visually indistinguishable from the output of two weeks of production, and it carries none of the visual cues that used to tell a viewer which one they were looking at.

Those cues were doing real work. The wobble in a hand-drawn frame, the placeholder timing in an animatic, the grey box where an image would go: none of that was a limitation to be engineered away. It was a signal, sent for free by the medium itself, of the stage of the work and what kind of reaction the work was asking for. Fidelity used to be expensive, which is precisely why it was informative.

Now it is cheap, which makes it noise. Every concept arrives wearing the same clothes as every final. A viewer, like our client, who misreads a rough cut is at least misreading something. A viewer looking at a generated concept has nothing to go on. The response is not to artificially rough up the work, which is condescending and transparent. It is to say the stage out loud every time, because the artifact will no longer say it for you. The framing was always good practice. It is now the only thing carrying the information. 

The bottom line

It is easier to criticize what exists than what does not. That is the argument for visualizing ideas early and often. But existing is not the same as finished, and research on prototype fidelity shows that viewers respond to how something looks, not just what it is.

None of this is about the sketch, the animatic, or the screen share for their own sake. The point of showing rough work early is to get real feedback from everyone who has a stake in the message before the message is locked in. That is how you pressure-test an idea properly: not by asking one person to bless it, but by putting it in front of the people who will each catch something different while it is still cheap to change.

We got there with the client in the end. The animatic never changed. What changed was the sentence we started saying before we press play.

Sources

Rettig, Marc. “Prototyping for Tiny Fingers.” Communications of the ACM, vol. 37, no. 4, 1994, pp. 21–27. https://cacm.acm.org/opinion/prototyping-for-tiny-fingers/

Virzi, Robert A., Jeffrey L. Sokolov, and Demetrios Karis. “Usability Problem Identification Using Both Low- and High-Fidelity Prototypes.” Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI '96), 1996, pp. 236–243. https://dl.acm.org/doi/10.1145/238386.238516

Sauer, Juergen, and Andreas Sonderegger. “The Influence of Prototype Fidelity and Aesthetics of Design in Usability Tests: Effects on User Behaviour, Subjective Evaluation and Emotion.” Applied Ergonomics, vol. 40, no. 4, 2009, pp. 670–677. https://doi.org/10.1016/j.apergo.2008.06.006

Buxton, Bill. Sketching User Experiences: Getting the Design Right and the Right Design. Morgan Kaufmann, 2007.

Heath, Chip, and Dan Heath. Made to Stick: Why Some Ideas Survive and Others Die. Random House, 2007. https://heathbrothers.com/books/made-to-stick/

Paivio, Allan. Imagery and Verbal Processes. Holt, Rinehart and Winston, 1971.

Paivio, Allan. Mental Representations: A Dual Coding Approach. Oxford University Press, 1986.

Mayer, Richard E. Multimedia Learning. 2nd ed., Cambridge University Press, 2009.

About Gallery Design Studio

Gallery Design Studio is a go-to-market creative partner for high-growth B2B technology companies. We work with a select number of companies to plan, structure, and produce the strategic visual content that moves complex deals forward. By invitation only. gallerydesignstudio.com