نوع مقاله : مقاله علمی پژوهشی
عنوان مقاله English
نویسندگان English
Objective
This study aims to develop a novel methodology for evaluating the efficiency of user interface (UI) pages in the digital advertising industry. By integrating the strengths of different data envelopment analysis (DEA) models with users’ neural response data, the study seeks to develop a model that can evaluate both the overall efficiency of a UI and the relative efficiency of its individual components within a short computational cycle. The proposed approach is intended to provide website owners and decision-makers with practical information for optimizing UI design, particularly advertising areas, and improving return on investment (ROI).
Methodology
This applied study was conducted using a case-study approach. The statistical population consisted of users interacting with a specific set of website UIs. Participants were selected using a non-probability random sampling approach, and data were collected by observing users’ eye-pupil and gaze responses while they interacted with the UIs in a realistic environment. Given the importance of visual perception in UI interaction, eye-tracking data were incorporated into the efficiency evaluation framework. The validity of the selected indicators and criteria was assessed through face validity and consultation with relevant experts and specialists. Network data envelopment analysis (NDEA) was then employed to construct and evaluate the proposed efficiency models. The models were implemented and tested using LINGO version 17.
Findings
The results demonstrate that integrating eye-tracking data with commercial and UI-related indicators provides a comprehensive framework for evaluating UI efficiency. The proposed NDEA-based model enables the assessment of overall UI efficiency while also distinguishing the relative efficiency of specific components, including the header, main content area, advertising sections, and unrelated or empty spaces. The model further incorporates desirable and undesirable inputs and outputs, thereby providing a more comprehensive representation of UI performance. The results indicate that identifying inefficiencies at the component level can provide decision-makers with more specific and actionable information for improving UI design and the effectiveness of digital advertising.
Conclusion
The proposed methodology provides a systematic framework for evaluating and improving website UI efficiency, particularly in the context of digital advertising. By combining eye-tracking information with non-parametric efficiency analysis, the model can help website owners identify inefficient UI components and develop targeted improvement strategies. The findings highlight the importance of optimizing advertising-related UI elements in response to the growing role of digital and social media marketing. Overall, the proposed approach provides a practical basis for benchmarking UI performance against competitors and supporting data-driven decisions aimed at improving advertising effectiveness and ROI.
کلیدواژهها English