MODELLING USER BEHAVIOUR AS A TOOL FOR IMPROVING UX DESIGN

Abstract

While working with any-purpose web resources, we regularly encounter the same scenario where the interface appears user-friendly, and its logic is properly designed, yet users still make mistakes and leave the site. It seems evident with time that the issue is not about the interface and its design itself, but about the problem of the user’s unawareness about how to work with it during the session. Besides, the user behaves typically differently from the way it is described in the manual. The study presents the author’s method for investigating user behaviour, based on the investigator’s observations and analysis.

The methodology was aimed at distinguishing stable behavioural models of typical action scripts with the focus on repeating errors and micro-responds occurring during the interaction with the web source. Unlike the traditional approach to the UX access, based on visual and functional optimization, user’s behavioural modelling is considered as the forecasting tool for reaction prediction during the final project stage. The core idea is that the notorious behaviour should be transformed into useful and project-oriented decisions. This includes the decisions towards users’ simplifications, the accent’s shift towards the interface, and the reduction of the number of significant choice points and the implementation of support mechanisms as well. Significant results were obtained while the implementation of three practical issues, namely, e-learning systems, e-service portal, and an e-commerce website. The number of repeating errors was significantly decreased because of the proposed accessibility, as well as the assessment completion time decrease for 20-30% on average. The same productivity refers to the improvements of the interface’s convenience.

The study’s scientific novelty is determined by its statement that the opportunity of user’s behaviour modelling application as a real-interaction-oriented single user experience is more beneficial than the use of idealised patterns for modelling. The practical value of this study is embodied in the chosen methodology’s appropriation as a practical designer’s, UX analytics, and developers’ tool for better UX comprehensibility and predictability in various contexts and users’ strategies. In the end, its accessibility is a game-changer in modern UX design quality.