Whoever uses a digital application often decides on its quality within seconds. Until now, appealing design has often been seen primarily as a factor for first impressions or subjective satisfaction. The meta-analysis recently published in the "International Journal of Human-Computer Interaction" now shows that aesthetically designed digital applications can also improve the actual performance or behaviour of users when interacting with a system - for example in terms of speed, accuracy or efficiency.

"Our results show that aesthetics are by no means just decorative accessories," says study author Professor Meinald Thielsch, who heads the Chair of Work and Environmental Psychology at the University of Wuppertal. "Attractively designed digital interfaces can help people to complete tasks more successfully and efficiently."

For the study, the research team analysed 31 studies with a total of 18,794 participants. Only studies in which the differences between aesthetically appealing and less appealing designs were actually perceived by the users were taken into account. Overall, the results show a small to medium positive effect of aesthetics on performance - significantly stronger than first demonstrated in an earlier analysis from 2019.

Bridging the gap between research and practice

Whether privately, professionally or in education - billions of users interact with user interfaces every day. "Our analysis provides important insights for companies and organisations in particular," says Marleen Schlamann, psychologist and user experience consultant at MAI eresult. "Investing in good design not only improves the perception of digital products, but can also support their actual use. This is also an economically relevant factor if a digital application is used by a large number of users."

In practice, the researchers therefore recommend designing aesthetics in a targeted and evidence-based way. The integration of user feedback into the development process is key.

At the same time, they see their results as an important contribution to the further development of research in the field of human-computer interaction. The analysis revealed methodological gaps in previous studies. The authors summarise that many of them measure aesthetics imprecisely or neglect important influencing factors.

In addition, previous studies provided partly contradictory findings on whether attractive design tends to support or even distract. In contrast, the new meta-analysis finds no evidence that attractive design fundamentally impairs performance.

Comprehensive literature research

For the meta-analysis, the research team systematically searched the international research literature on the relationship between visual aesthetics and user performance. The starting point was around 40,000 potentially relevant publications, the selection of which was supported by AI-supported methods, among other things. The final analysis included 31 studies with a total of 18,794 participants. Among other things, processing speed, accuracy and efficiency in the use of digital interfaces were assessed. The researchers also analysed various influencing factors, such as the type and duration of interaction, measurement of aesthetics or specific design elements such as typography.

However, the researchers also emphasise that there is a need for further research. Although the analysis confirms the positive influence of aesthetics on performance, there are also clear differences between individual studies. In future, more precise and methodologically high-quality studies are therefore needed in order to better understand which influencing factors are particularly effective - and why.

The team itself has made a start by identifying ten influencing factors that could explain the effect of aesthetics on performance and could therefore serve as a guide for future studies.

Link to publication: Schlamann, M., Nestler, S. & Thielsch, M. T. Attractive Things Do Work Better: A Meta-Analysis on Visual Aesthetics and User Performance. International Journal of Human-Computer Interaction. DOI: https://doi.org/10.1080/10447318.2026.2664081