Patricia Evanoff

Discovery

I worked with the client conducting discovery workshops to understand their business goals, audience and user goals, and subject matter background to guide the work. I created an audience analysis & business objectives discovery findings deck.

Guy Carpenter was mainly using tabular spreadsheets and wanted visualize them into interactive dashboards to make them easier to understand for their clients as an advisory tool. I worked with the stakeholders and business analyst to understand the data and key performance indicators to be able to visualize the key data stories and structure the dashboards

Here are some of the requirements and use cases we defined:

• show the client compares to peers on average as well as in various cohorts and characteristics

• show different risk characteristic dimensions like peril, region, and line of business

• highlight how the data quality changes over time, such as a year over year comparison

• give the ability to filter and drill in further to invite clients to engage and consult with Guy Carpenter for further advisory services

Streamlining visualizations
Stakeholders from Guy Carpenter reviewed with us some example explorations for the Data Quality dashboard to help us understand the data they were looking to visualize. They sought our expertise to turn complex data into streamlined dashboards and visualizations.

The vision of this was to create 3 initial dashboards with a structure that could be expanded with more in the future. The 3 dashboards were in the following areas of catastrophe modeling risk data:

• Data quality

• Exposure composition

• Loss Drivers

Initial sketches
I began ideating for the data visualization with some low fidelity sketch concepts on the data stories we wanted to communicate.

Wireframes

Initial Wireframes & User Testing

Next, I started to get into the more detailed definition of the dashboards with high fidelity wireframes. This established a structure for the dashboards as well as initial passes at the visualizations.

I reviewed the wireframes with stakeholders and then used them to conduct qualitative user testing interviews and gather user feedback.

See the user testing wireframes

Integrating Feedback
I processed the user feedback and translated it into updates to the wireframes. The biggest area of change was to rethink the nightingale charts. While the nightingale charts were visually impactful and capable of showing multiple dimensions of data at once, they lacked common familiarity that was easily understandable to users. These were updated with a variation on bar charts.

See the final wireframes

Accessible Data Visualization Design

In addition to creating a clean, modern look consistent with the brand, I paid special mind to accessibility for the visualizations. I met WCAG AA contrast for color, but I also tried not to rely solely on color for differentiation. One example of this is in the line graph, which uses a pattern in addition to color.