The project team consisted of the CSS core team and Unic experts. Following our project vision, “A step ahead of the user”, we started looking for answers to the questions above. Based on the human-centered design framework, we used various research methods over the course of the projects that helped us get a clearer picture of the users.
During the analysis phase of the project, we exchanged a lot of information with the CSS team for customer experience management and market research. They had already drafted specific personas. These personas were modeled after prototypes of user groups with the same needs, characteristics, mindset, behavioral patterns and demographic data. The CEM team also provided us with findings from the current personal customer survey on the use of the website. This survey identified customers’ difficulties as well as the potential of the existing website.
Website visitors always leave a footprint. To learn more about the current situation on the website, we decided to do an analytics audit. As well as the css.ch website, we also audited all other CSS applications involved. Other focal points included the usage itself and the context of how, when and how often visitors use the website.
The data from the analytics audit helped us better understand user journeys. Data on visits, frequency, duration of sessions and usage of the search function and blog enhanced our understanding of the users.
Qualitative Online Survey
From 22 November to 4 December 2018, we performed a qualitative online survey with 41 respondents. We used Microsoft Forms as a tool and invited individuals to participate. Our goal with the online survey was to learn about people’s reasons to switch to a new insurer and the context surrounding the insured person.
The qualitative survey showed that the main reason for a switch to a new health insurance provider is cutting costs for basic and supplementary health insurance. About a third of the respondents had switched insurers in the two previous years.
Individual life events influence the choice of and decision for supplementary insurance. For instance, insured persons look into their health insurance and coverage when they move to a different canton or when they are planning a family. They reevaluate their needs based on their current situation. In general, respondents stated they want to be satisfied with their health insurance and are more likely to switch when they are not satisfied. The key result from the survey was that customers always act in a very specific context and based on their particular needs.
The findings gathered using this method helped us draft the information architecture for personal customers and the content structure with users’ needs in mind.
We used rapid prototyping for the development of the conceptual framework and the detailed concept. This formative evaluation method is intended to iteratively identify usability issues together with users, ideally within a day, and to improve the prototype. We applied this method with users to evaluate the usability of the mobile navigation concept and the login/logout process.
Mobile Navigation Concept
We used the mobile-first approach to design the navigation concept for personal customers in March 2019. Very early on, we validated three navigation principles with five test individuals: Drill down, accordion and mixed navigation.
There was no clear-cut test result: The test individuals could not solve one of six test tasks or required help to do so. Users had difficulties with navigation: Depending on the navigation principle, the entire structure or only part of it was visible. There were issues with the “back” link and with opening the accordion. Some of the navigation items were not self-explanatory and posed an additional hurdle for the test individuals. The findings helped us improve navigation and usability as well as user prompts.
The questions for the next iteration of the navigation concept were: How do users get to css.ch? Do users move locally, in a context-related way or globally? Do we want to offer context-based content such as teasers in the menu?
When developing the conceptual framework, we had several meetings on the issue of login levels. In May 2019, we used rapid prototyping to evaluate the concept with users. In the detailed concept, the login/logout process for the my.css.ch customer portal was evaluated with five test individuals. We used a prototype designed with Axure for this test.
A key finding from rapid prototyping was that the understanding and expectations of “logout” were vastly different. The test showed room for improvement in the process from a user perspective. The findings were included in the concept and technical implementation. After the test, the core team was undecided on whether to continue with the login-level approach. Due to new technological possibilities, the team chose a different option in the end.
Over the course of eight months, we as a team were able to improve our understanding of the customer. We collected and analyzed empirical data in every project phase, based on the human-centered design norm, which consistently places the focus on the user. We included our findings in the evaluation of our solutions.