We craft impactful and influential visualizations that tell data-driven stories about the dynamics of the life insurance industry.
The standarization inititive constitutes establishing a design language that incorporates a style guide and a set best practices in order to guide the creation of new visualizations. A style guild ensures that data is presented with a common voice and makes the data the focus, not stylistic differences. Our set of best practices guide visualization composition and behavior and our guidelines are based on recommendations from experts in the field of visualization and, if followed, optimize a visualization to communicate the underlying data.
The Data Visualization Team builds and maintains open source visualization libraries. These libraries support the visualization team, as well as business analysts and data scientists. The libraries enable easy and consistent production of standard visualizations and charts. The theme built into the library employs our style guide and and follows our determined set of best practices. Furthermore, examples and instructions are provided to facilitate adoption.
Our visualization process leverages techniques in User Experience Design, Human-Computer Interaction and Software Architecture. The design process is user-centered, task focused and the implementation process is technically advanced. The goal is to produce visualizations that enable users to easily and quickly gain insights, accomplish tasks, and are enjoyable to use. We also build reusable components, setup systems, and establish processes that contribute to an automated visualization pipeline. This visualization pipeline improves our efficiency at delivering future visualizations and dashboards.
Visualizations go through a prototyping process in which stakeholders and potential users provide feedback through multiple iterations of usability testing. The visualizations are optimized around the user's need to accomplish tasks and to gain insights from the data.
During the implementation phase, the technologies we use are dependent on the purpose of the visualization, required functionality, complexity tradeoffs and context of use. We address visualization performance by optimizing the logic that maps data values to visual space as well as selectively choosing the rendering method. Additionally we make use of data marts, tailor queries to their corresponding visualization and strategically making use of caching.