Usability testing with insurance agents
Overview
In early 2022, our team at Mobee Dick started a usability research project for UNIQA — one of the leading insurance companies in Poland. The UNIQA design team approached us to evaluate the usability of a new version of their portal for insurance agents.
My Role
I collaborated with Łukasz Roth — an experienced UX designer — and Patrycja Hojka, a mentor and UX researcher. Within the project, I was responsible for:
- Study design — based on the brief, kick-off meeting with the client, and the delivered prototype, I developed research goals and study structure, which I then supplemented with specific tasks and in-depth questions
- Moderating research sessions — I conducted remote, moderated usability tests, introduced participants to task contexts, observed their behaviour, directed their attention, and probed important topics
- Leading debriefing sessions — together with the client team, we discussed observations, insights, and potential solutions to identified issues; along with Łukasz, I facilitated the discussion and documented decisions
- Collaboration and iteration — I continuously communicated progress to both the client and internal team; I supported client engagement in the research process (including preparing best practice observation materials) and consulted on prototype iterations with the UX designer
- Reporting and presenting findings — I analysed and synthesised research results into an accessible report, which I then presented during the summary meeting
Research goals
At the start of the project, we broke down the general study goal — verifying the prototype's usability — into more specific areas:
- Checking how users understand content, messages, and terminology on individual screens
- Understanding how users perceive the categorisation of insurance products
- Analysing how users search for specific products, policies, and offers
- Testing the usability of calculator shortcuts and quick action features
- Evaluating alternative versions of the homepage
Method
Since the tested solution was in prototype form, implementation was deferred, and the client team had the opportunity to actively participate in the design and research process, we jointly decided to conduct the project using the RITE (Rapid Iterative Testing and Evaluation) model.
A total of 9 insurance agents recruited by the client participated in the study. Each session lasted approximately 1.5 hours. The Mobee Dick research team (moderator and observer) and a client representative attended the sessions.
After each series of 2–4 sessions, we met to discuss the research progress, draw conclusions, and plan subsequent iterations.
RITE research formula
Usability tasks
With clearly defined research goals and access to a working prototype, I prepared a usability test script that standardised the flow of sessions with participants.
Sample tasks and probing questions
The script was additionally verified during a pilot research session.
Moderation and observation
During the project, I served as both moderator and observer of the research sessions. We documented observations on a FigJam interactive whiteboard, and conducted sessions remotely using MS Teams.
One of the research sessions moderated by the author
Observation space on the FigJam interactive whiteboard
Analysis and synthesis
During debriefing sessions, we jointly reviewed the research script, discussing observations, insights, and potential solutions to problems encountered by participants. We met every 2–4 sessions, though we occasionally consulted after a single session.
After each debriefing, we documented the decisions made for easy access to the change history — this also applied to smaller iterations introduced between sessions.
At the client's request, we grouped observations into themes and subcategories. This structured data allowed us to develop specific strategic recommendations.
Categorization of observations including themes and minor threads
Sample summary of a debriefing session
Impact
Over the course of the project, we collected more than 700 observations regarding user behaviours, habits, and attitudes towards the tested prototype. We implemented and tested nearly 50 changes.
We selected the homepage version that best matched user needs. We answered most research questions while identifying additional areas for further exploration.
Next steps
Information architecture testing
Study participants did not agree on the categorisation of insurance products. To better align the structure with users' mental models, it would be valuable to conduct card sorting studies and then verify the results through tree testing.
Testing in a production (or staging) environment
Participants expressed concerns about the differences between the prototype and the final product, wondering whether the target solution would meet their expectations. To address these concerns, it is crucial to conduct in-depth testing after implementation — in a production or staging environment.