Overview of a UX Design Process

Three colleagues and I worked on a project to design a portable interactive museum guide, which would enhance the experience of museum visitors. A design is only effective if it meets the needs of its intended users; hence our starting point was to develop an understanding of the needs of museum visitors. This was accomplished by going to museums and observing visitors and their activities, using Robson’s framework to provide structure. We gained insight into the physical environment, the variety of visitor profiles and the nature of their interaction with the museum exhibits as well as with each other. On a secondary note, we also gained insight into how current portable museum guides are used in order to identify what works well and what requires improvement. Our observation data were complemented by two sets of structured interviews, the second conducted with visitors either during their visit or at the end of it. We aimed to increase the validity of our data through triangulation: data triangulation was achieved by gathering data from the British Museum and the V&A Museum; investigator triangulation was achieved by gathering data independently; and methodological triangulation was achieved by using observation and structured interview techniques.

This yielded a variety of quantitative and qualitative data that had to be analysed and interpreted in order to define a clear set of functional and non-functional requirements for the guide. A hierarchical codebook was developed to pick out the key themes from the qualitative data and, along with the quantitative data, were collated and analysed in a spreadsheet, using graphical representation to identify dominant themes and patterns. The results of the analysis determined our primary and secondary personas as well as the requirements specification. The visitor attributes with the highest values were representative of a large section of surveyed users and therefore were used for the characteristics and goals of the primary persona. Similarly, the secondary persona was derived from recurring attributes. The functional and non-functional requirements were chosen to support the characteristics and goals of the personas, a deductive process of specifying requirements that enable the successful and satisfactory completion of a given goal. We accomplished this by drawing on our creative thinking, general knowledge and established principles of Interaction Design such as affordance, visibility and consistency. While these translated well into usability goals, addressing user experience goals was less certain as they are purely subjective; the degree to which our design was satisfying, enjoyable or helpful could only be determined in the evaluation phase.

Given the popularity of smartphones, and thus apps, among our personas, it was used as the interface metaphor for our design. It provided a familiar structure in terms of having a Main Menu, Search function and page navigation. This also extended to the Map and Navigation functions. A mixture of four interaction types were used in the design. These included instructing the guide by selecting menu options on a touchscreen, conversing by typing in search queries, manipulating by physically waving the guide when using the NFC function and exploring by using the map and navigation functions. The primary interface type considered for the guide was touch, with the ability to tap and swipe offering flexibility and intuitiveness. Although gestures may be interpreted differently from one culture to another, which is of particular significance given the different nationalities of the personas, we believe a gestural interface such as touch is sufficiently global in adoption to minimise the use of incorrect mental models that do not match the proposed conceptual model. A haptic interface was also considered but decided against because we believe vibrotactile feedback would not necessarily enhance the museum experience and might cause dissatisfaction if visitors that did not require the function had to take extra steps to turn it off.

The conceptual design involved placing each persona in a scenario and developing a storyboard to capture the sequence of events. The main focus of the storyboard was on the screen of the guide; as it was a portable touchscreen device carried about by each visitor, the interactions taking place on the screen were of greater relevance than environmental factors. However, the boards that captured the navigation activities focused on the environment, as it was necessary to consider the physical layout of the museum.

Based on our initial data gathering and analysis, the two most important activities carried out by the personas were making a selection from a curated list of the top exhibits in the museum and using a map with navigation for directions to a specific exhibit. Hence these two elements were more fully developed in the detailed design. We opted for a paper prototype, as it was an effective way of not only depicting the content and layout of each screen but also simulating visitors’ interactions with the guide by changing the sequence of the screens in response to a given input.

The low-fidelity nature of a paper prototype was ideal for our purposes as it was practically free and could be easily created and modified without requiring technical or artistic skills. Designing the prototype in PowerPoint as opposed to hand sketching enhanced our ability to visualise each screen and efficiently explore alternative ideas. In terms of functionality, detailed design emphasises depth over breadth and therefore the prototype was vertical as it modelled the complete sequence of activities for each element. Since an actual portable interactive museum guide was not being built, we did not have to formally consider if the prototype was throwaway or evolutionary; however we still viewed it as a throwaway prototype as it was only used for concept development and not the final product.

The final phase of the design process was the evaluation although in principle, the cycle of requirements gathering, design and evaluation is iterative until the final product is ready. To this end, we elicited the support of four test users whose profiles were close to the personas. The aim of the evaluation was to detect design flaws that conflicted with the desired usability and user experience goals, garnering feedback that could subsequently be used to improve the design or explore alternatives.