011 min
Why the aesthetic-usability effect matters
A team runs a usability test on a new checkout. The participant gets stuck twice, takes four minutes on a task that should take one, and then rates the ease of use as "very high". The team has two choices. They can believe the rating and ship, or they can believe what they watched.
The aesthetic-usability effect is the reason those two signals disagree. People's opinion of how easy something is to use is partly an opinion of how it looks. That has two consequences for the people who build products.
The first consequence is an advantage. A visually appealing design earns patience. Users are more willing to try it and more willing to forgive small problems in it. The second consequence is a hazard for research. A participant who likes the look of a design will say it is easy even when their behaviour shows it is not, so problems stay hidden in the data.
The cost of ignoring the effect runs in both directions. A team that treats visual design as decoration gives up goodwill that costs little to earn. A team that reads positive comments about visuals as proof of usability ships problems that its own test had already recorded.
022 min
What the aesthetic-usability effect says
Nielsen Norman Group defines the effect briefly. The aesthetic-usability effect refers to users’ tendency to perceive attractive products as more usable. The important word is perceive. The claim is about what people believe about a product, and not about how fast or accurately they can use it.
Two measurements are involved, and keeping them apart is the key to the concept.
- Perceived usability: how easy a person thinks the system is. It is collected by asking, usually on a rating scale.
- Actual usability: how well the person performs. It is collected by observing, for example by timing a task and counting errors.
The effect appears when the rating of attractiveness tracks perceived usability more strongly than it tracks actual usability. In plain terms, people's feelings about looks leak into their feelings about ease.
A psychological explanation has been offered by the researchers who tested it. Tractinsky, Shoval-Katz and Ikar (2000) observed that their results resemble those found by social psychologists regarding the effect of physical attractiveness on the valuation of other personality attributes. In that research, a person's physical attractiveness changes how other personal qualities are valued. The same kind of carry-over may apply to objects. The halo effect is the general name for that pattern, and the aesthetic-usability effect is one case of it.
A practical reading follows from this. When users call an interface "simple" or "clean", they may be reporting an impression of ease that the visual surface produced, and they may not be reporting the number of steps it takes.
031 min
Where the effect was first measured
The effect was first studied in human-computer interaction in 1995. NN/g's history of the effect describes the study: researchers Masaaki Kurosu and Kaori Kashimura from the Hitachi Design Center tested 26 variations of an ATM UI, asking the 252 study participants to rate each design on ease of use, as well as aesthetic appeal.
The finding is a comparison of two correlations. They found that the correlation between the participants’ ratings of aesthetic appeal and perceived ease of use was stronger than the correlation between their ratings of aesthetic appeal and actual ease of use. The authors titled their paper "Apparent Usability vs. Inherent Usability", and presented it at the CHI '95 conference.
Kurosu and Kashimura concluded that users are strongly influenced by the aesthetics of any given interface, even when they try to evaluate the underlying functionality of the system. The ATM was a good test object because every variation did the same job, so any difference in rating came from the interface and not the task.
Don Norman explored the concept in depth in his 2004 book Emotional Design, where he applies it to everyday objects.
042 min
Replications and the case for causation
A single study with one interface type in one country is weak evidence. Two later studies addressed that.
A replication in a second country. Tractinsky (1997) ran three experiments to validate and replicate the Kurosu and Kashimura result in a different cultural setting. His abstract reports that the results support the basic findings by Kurosu and Kashimura, with very high correlations between the perceived aesthetics of the interface and the perceived ease of use of the system before use. The correlations were different in size between Japan and Israel. Tractinsky wrote that this suggests cross-cultural differences, but these were not in the hypothesized direction. The effect held in both places, and its size varied.
A test of cause. A correlation does not say whether looks change the rating or the other way around. Tractinsky, Shoval-Katz and Ikar (2000) used a computer program built as a stand-in for an ATM and asked participants for ratings before and after they used it. Before use, the correlation between perceived aesthetics and perceived usability was strong. After use, the strong correlation remained intact. They then varied the real attributes of the system and found a result that supports the effect: the degree of the system's aesthetics affected post-use perceptions of both aesthetics and usability, whereas the degree of actual usability had no such effect.
Read that last sentence slowly, because it is the strongest evidence in the area. In that experiment, changing how good the system looked changed what people said about its usability. Changing how usable it actually was did not change what they said about its usability.
Both studies used an ATM, a short task with a single goal. That limits how far the conclusions reach, which the next section takes up.
052 min
Where the evidence is contested
A well-known later experiment found the effect can reverse. Tuch, Roth, Hornbæk, Opwis and Bargas-Avila (2012) asked 80 participants to use one of four versions of the same online shop. The versions differed in interface aesthetics (low or high) and in interface usability (low or high). Participants had to find specific items and rated the shop before and after use, on perceived aesthetics and on perceived usability.
They report that aesthetics does not affect perceived usability. In contrast, usability has an effect on post-use perceived aesthetics. In their shop experiment, then, the direction ran backward: a hard-to-use shop was rated as less beautiful after people used it.
The authors give a condition and a possible mechanism. The "what is beautiful is usable" notion can be reversed under certain conditions, which they name as a strong usability manipulation combined with a medium to large aesthetics manipulation. They suggest the user's affective experience with the usability of the shop might serve as a mediator: the frustration of poor usability lowers ratings on perceived aesthetics.
How should a practitioner read two experiments that disagree? The studies differ in task, in how large each manipulation was, and in how long participants used the system. A fair summary is this:
- Before use, appearance shapes the expectation of ease. The evidence for this is consistent.
- During use, real difficulty creates frustration, and frustration lowers how good the design seems.
- The effect is therefore most likely to hide problems that are small, and least likely to hide problems that are large.
NN/g reaches a matching practical conclusion. A pretty design can make users forgiving of minor usability problems, but not of large ones.
062 min
How the effect distorts usability testing
The effect is not only about customers. It is also a risk to people who run research. NN/g describes the familiar frustration: a facilitator watches a participant struggle, getting stuck and failing tasks, and then the participant comments only on the site's great colour scheme.
NN/g gives two cases from its own sessions.
- Arcadis. A participant visited the website of a consultancy that used large background photos on many pages. At first they responded positively: "First thing, popping into this webpage, I see this beautiful, colorful image." After struggling to complete several tasks, they revised the opinion and said the full-screen image was pretty awesome once and probably annoying the second time. The initial praise came before the problems and the revision came after them.
- FitBit. A participant met many issues while shopping on the FitBit site, ranging from minor annoyances to serious navigation flaws. She completed her task, but with difficulty. In a post-task questionnaire she still rated the site's ease of use very highly, and explained the rating with the colours and photographs. NN/g says that the positive emotional response caused by the aesthetic appeal of the site masked its usability issues and made them harder to identify.
The two cases show the effect from both sides. In the Arcadis session the problems eventually overcame the beauty. In the FitBit session the beauty hid the problems from the participant's own summary.
The remedy is to give less weight to what people say about ease and more to what they do. NN/g's closing advice is to pay attention not only to what users say, but also to what they do. Measure task success, time on task and errors, because those cannot be raised by a pleasant colour scheme.
072 min
How the effect shows up in product and design work
Prototype reviews. A designer shows a polished mockup and a rough wireframe to two stakeholder groups. Reviewers may respond to the finish of the polished one and not to the structure of the flow, which is the same in both. Showing wireframes first keeps feedback on the flow.
Usability test planning. A researcher writes the task list so that success is measured by behaviour: did the participant finish, how long it took, where they hesitated. They keep the rating questions but treat the answers as impressions, which is what the evidence says they are.
Judging a redesign. A team that measures only first-visit impressions may overstate the benefit of a new look. The Arcadis session shows that approval can fade with repeated use, so a check on later visits is worth adding.
Prioritising work. A PM weighing a visual refresh against a fix to a confusing form should not treat the two as the same kind of investment. The refresh raises patience for small problems. The form fix removes a large problem, and the research says large problems are the ones beauty cannot cover.
Heuristic reviews. An expert who runs a heuristic evaluation on an attractive interface should be aware that their own judgment is also subject to the effect. Walking through the task step by step, with a checklist, reduces the pull of a good first impression. A clear visual hierarchy is a case where looks and function help each other, because the ordering that makes a page look orderly also tells the reader what to do first.
081 min
How to use the effect and guard against it
To use it:
- Treat visual quality as part of usability work and not as a final layer. A consistent grid, clear type and sensible spacing buy goodwill for every later interaction.
- Spend the visual effort where first impressions form: the landing page, the first-run experience and the first screen of any new feature.
- Let the beauty serve the content. NN/g notes that the effect is at its strongest when the aesthetics support and enhance the content and functionality.
To guard against it:
- Separate the two measurements. Ask for ratings of looks and ratings of ease as different questions, and compare each against observed behaviour.
- Test with a rough version as well as the polished one when the question is whether the flow works.
- When a participant praises the visuals, ask what they were trying to do and what happened. The answer tells you whether the praise was about the task or about the surface.
- Look for the gap between the rating and the behaviour. A high rating on a session with errors is a signal to investigate, and not a reason to relax.
091 min
When the effect stops helping
The effect has limits, and each limit has a tell.
- Large usability problems. NN/g says the effect covers minor problems and not large ones. Tell: users abandon a pretty page after repeated failures.
- Repeated use. In the Arcadis session, the images that pleased the participant at first annoyed them on later pages. Tell: positive comments in the first minutes and negative comments later.
- Function sacrificed for style. NN/g says that when products suffer from severe usability issues, or when functionality is sacrificed for aesthetics, users tend to lose patience. Tell: a redesign that looks better and takes more steps.
- Experienced frustration. The 2012 experiment suggests that frustration lowers the rating of beauty. Tell: complaints about looks that began only after users had trouble with a task.
- Tasks where the cost of error is high. Where users must get the result right, a calm look does not substitute for clear labels and safe defaults. The evidence in the ATM studies comes from low-stakes tasks, so extending it to high-stakes tasks goes beyond what they measured.
Do not read the effect as a reason to avoid plain interfaces. A plain interface that works well can still be rated as easy to use. The effect says beauty adds a margin of patience, not that beauty is required.
101 min
Aesthetic-usability effect compared with nearby ideas
| Idea | What it claims | Where it differs |
|---|---|---|
| This pageAesthetic-usability effect | ||
| PsychologyHalo effect | ||
| MarketingAffect heuristic | ||
| Not in the library yetDelight |
The aesthetic-usability effect is also different from usability itself. Usability is how well people can complete tasks. The effect is about how usable people think a product is, and the whole point of the concept is that the two can come apart.
?6 questions
Questions people ask
What is the aesthetic-usability effect?
Does a beautiful design make a product more usable?
Who discovered the aesthetic-usability effect?
Why is the aesthetic-usability effect a problem for usability testing?
Is the aesthetic-usability effect always true?
How is the aesthetic-usability effect different from the halo effect?
▶3 videos
Watch
Understanding Aesthetic-Usability
PRINCIPLES of UX Design | EXPLAINED | Tamil | #uxdesign #basics
the design show
12 UI/UX Laws You MUST KNOW 🧠 | Become a UI/UX Designer in 2024 | Saptarshi Prakash
Saptarshi Prakash
§5 sources
Sources
Show all 5 sourcesHide sources
Moran, K. (2024). The Aesthetic-Usability Effect. Nielsen Norman Group.
Kurosu, M. and Kashimura, K. (1995). Apparent usability vs. inherent usability. CHI '95 Conference Companion. Described in source 1.
Tractinsky, N. (1997). Aesthetics and apparent usability: empirically assessing cultural and methodological issues. CHI '97.
Tractinsky, N., Shoval-Katz, A. and Ikar, D. (2000). What is beautiful is usable. Interacting with Computers 13(2), 127-145.
Tuch, A. N., Roth, S. P., Hornbæk, K., Opwis, K. and Bargas-Avila, J. A. (2012). Is beautiful really usable? Toward understanding the relation between usability, aesthetics, and affect in HCI. Computers in Human Behavior 28(5), 1596-1607.





