I was a key practitioner on the UX team that redesigned UW Medicine’s patient experience a few years ago. The experience was redesigned in early 2017, but served with minimal changes from 2012 until 2016. Our process included evaluating the personas the client had already created, building a patient journey map, developing an IA for the new experience, and conducting a comprehensive content audit. I did the primary work of developing the new navigation, prototyped the new experience, helped design the usability study, and conducted interviews with half of the participants. I also identified key opportunities for improvement from the user testing, made changes to the prototype, helped present to doctors and other stakeholders, and collaborated with the developers who would be implementing the design.
The culmination of UW Medicine’s redesign of the patient experience was a low fidelity interactive prototype built in Axure. To determine content needs and design patterns, we had previously conducted sketching exercises with members of the web team, marketing team, and executive-level leadership. After that activity, we had a solid basis on which to start prototyping.
The prototype was deliberately low fidelity, but we included actual content and higher-fidelity interaction when necessary to get optimal results from the user testing. A design pattern was established for medical services, conditions, and locations, as well as more utilitarian functionality like making an appointment and finding out information about billing.
After we had come to an internal consensus about the prototype and presented iterations of it to the web team, I facilitated user testing of the proposed redesign, working from a colleague’s testing plan. We set tasks that focused on both content discovery and navigation, based on user goals set out in the initial personas and refined in patient journey modeling activities.
I performed user testing with five participants, five in person and five remotely, using WebEx to view their desktops and record the sessions.
Many of our participants were retirees and were used to the workarounds and compromises they had developed to meet their own needs with the existing UW Medicine website. Communicating the goal of the exercise and making them comfortable with the testing environment and technology was a challenge.
We gathered excellent data that I was able to incorporate into the prototype, making it a much stronger basis for development work. Some of these findings were conceptual, revealing insights into user mental models, while others were very tactical, revealing labels or microinteractions that needed further refinement.We also gathered a large amount of serendipitous information, which the web team was able to use to begin an evaluation of some of their other processes and systems.
I compiled the user testing data into a deck which we presented to our immediate stakeholders, leadership, and interested committees of doctors and staff. Findings were organized by task, presented with additional details, and then the action that should be taken to address each finding. Summaries of implications for other tasks in our work streams, like the content strategy, were also called out explicitly, to make the future use of this information more apparent.