BCAA: UX Case Study
Improving online insurance purchase through autonomy, assistance, and automation.
The British Columbia Automotive Association (BCAA) is one of Canada’s most trusted insurance providers for home, auto, and travel. BCAA partnered with Invoke to redesign their Home Insurance online purchase experience to make it faster, more personalized, more accessible, and to stay competitive against new insurance products and services.
Sketching + Wireframing
I. Understanding the Product Strategy
Based on a kick-off workshop, user interviews, and competitive audits, our strategy team established the new Product Vision and created a set of Hypotheses which would be the foundation of the improved designs.
Product Vision: To provide homeowners, condo owners, and renters the peace-of-mind and confidence of accurate and high-value insurance policies — offering autonomy, assistance, and automation along the way.
Users that input partial data and receive complete data will save time and effort. They are more likely to re-engage, customize, and buy an insurance policy.
Users that customize a quote to improve accuracy and/or lower the price are more likely to buy an insurance policy.
Users that identify a quote that is suitable to them are more likely to re-engage, customize, and buy an insurance policy.
Users that submit additional content for review and verification by BCAA agents are more likely to feel confident about, customize, and buy an insurance policy.
Users that access in-line educational content (ex. tooltips) are less likely to require customer service support and are more likely to buy an insurance policy.
Users that utilize additional online channels (ex. chat) to get answers to outlier questions are less likely to abandon their quote and more likely to buy an insurance policy.
Together, we created a Conceptual User Flow showing where each hypothesis would be most relevant.
II. Wireframes, Rapid Prototyping, and User Testing
Key Players: UX Designer (me), UX Strategist, Project Manager
To get the most actionable results, we focused rounds of user testing on specific hypotheses or sections of the flow. Each two-week sprint included planning, wireframing, 2-3 user test sessions, and a review.
Pre-Estimate Screens / Autocomplete
For the first design sprint, I created mid-fidelity wireframes and user tested the screens leading up to the initial insurance quote. This included the Getting Started screen and basic home insurance questions such as home address and features.
We asked which areas users most expected to be pre-filled or autocompleted, and recorded their reactions when they saw certain fields autocomplete upon typing. I then revised the screens based on our findings, and carried the changes over to the next sprint.
Post-Estimate Screens / Customization
In Sprint 2, I created wireframes and user tested the screens following the initial quote. These included more specific questions about the user’s home, additional coverages, discounts, and a price flyout showing how the initial quote is affected by the new info.
We asked users how they felt about the amount and level of detail being asked from them at this stage, and whether they found the flyout useful. In general, users appreciated the flyout and were happy to answer additional questions if it meant receiving a more accurate estimate.
Estimate Screen / Self-Selection
The estimate or “initial quote” screen was one of the more challenging parts of the project, since drop-offs are typically high at this point. We hypothesized that users who identify a quote that is suitable to them are more likely to customize and buy an insurance policy, so we wanted to provide “Good, Better, Best” options for them to choose from.
Before jumping into design, we needed to answer some Big Questions:
What makes a policy “Good,” “Better,” or “Best”? Do “Good” and “Better” provide the most value, while ‘Best’ includes the most add-ons?
What key attributes and differentiators should we include for comparison?
Against whose quotes do we expect BCAA’s to be compared?
Can we produce a singular recommendation based on specific data-points captured from the pre-estimate screens?
Do we expect estimate customization to afford adjustments, plus additional coverages and discounts? How will this affect the post-estimate screens?
After a couple of working sessions with the BCAA team, I was able to craft and test a prototype of the Estimate and Price screens. We asked users if they understood the difference between the “Good,” “Better,” and “Best” options, and whether they trusted the recommendation given by BCAA. We rated their likelihood of customizing their selected policy, and learned what type of information they were looking for while selecting additional coverages.
I also adjusted all the pre- and post-estimate screens based on feedback from the first two rounds of user testing.
Addressing User Pain Points through User-Generated Content
By this stage, we had received consistent feedback that the most difficult part of the flow was the “Home Details” section, where users were asked if their Kitchen and Bathroom finishings were “Basic,” “Custom,” or “Deluxe.” Users found the definitions vague, and did not know which answer to select if their finishings fell somewhere in between.
We proposed solving this pain point through user-generated content, by asking users to upload photos so BCAA can accurately tag and categorize their kitchen and bathroom finishings. During the user test, we learned that users were happy to share photos of their home if it meant increasing the accuracy of their quote, receiving possible discounts, and for reference during future claims.
III. UI Mockups
Key Players: UX Designer (me), UI Design Contractor
Using the BCAA Brand Guide and with the help of a design contractor, I converted all our wireframes into hi-fidelity mock ups, as well as designed the final Payment screens. (Note: Icons would be added directly by the development team from BCAA’s existing library).
In this sprint, we tested users’ reactions to the UI, paying special attention to readability and accessibility.
IV. Content Writing and User Support
Key Players: UX Designer (me), UI Design Contractor, UX Copywriter
The final sprint focused on the refining the UX copy and incorporating hybrid support through in-line tooltips, live chat, and contact information.
Promoting User Autonomy through Education and Hybrid
Based on findings from the previous user tests, I worked with our UX Copywriter to tweak the copy and determine which areas of the flow needed tooltips the most. We then tested our assumptions by asking users what type of information they expect to see before clicking a tooltip, and whether the actual information provided was helpful.
We observed which areas users were most likely to use the chat feature and asked if they were satisfied with this level of online support. I also presented different placements and treatments of the tooltips, chat box, and contact information, and noted the users’ preferences.
V. Mobile Design and Developer Handoff
Key Players: UX Designer (me), UI Design Contractor, Project Manager
After incorporating the latest user feedback into the UI and copy, I packaged and annotated all the desktop designs along with mockups of key mobile screens for developer handoff.
In 12 short weeks, I was able to create, revise, and refine wireframe prototypes; conduct six rounds of users testing; and deliver hi-fidelity desktop and mobile mockups of key screens. I worked closely with Invoke’s Lead Strategist and spent two days a week at the BCAA head office, where I collaborated directly with their Home Insurance and in-house development teams.
Through this project, I learned the importance of defining and understanding product strategy, the power of iterative design, and how to balance business requirements with the user’s needs and feedback. The process was intense and laborious, but I am confident of the quality and thoroughness of my designs.