People are very good at detecting patterns, if the data is presented to us visually. Soundience has deep expertise in creating rich visualizations of data, which allow insights that are elusive using purely analytic approaches.
We have always used visualization for exploring data in our modelling work, and our clients increasingly value this capability on its own, as making sense of "big data" becomes increasingly important.
Approach and Tools
Our approach is to rapidly build interactive web-based visualizations based on each problem, to shed light on the messages embedded in the data.
While our tools include standard graphics and statistics packages, most of our visualizations involve creating custom visualization using canvas and Scalable Vector Graphics (SVG) in the browser, and static images on the server.
Example 1: Legal Cases
Since we develop our own software in the course of our work, we are not constrained by the capabilities of standard packages such as Excel. We have developed a number of charting types to display different aspects of data, and continue to develop more.
For example, a project with a law firm involved trying to find patterns in the timing, load, and staffing profile of their projects. While summary numbers and traditional charts provided some initial insights, we needed to see the profile of each project during its life, including the level of involvement by different levels of staff (a key profitability driver). To this end, we developed "spike diagrams" that helped us to find the relevant patterns more quickly, and included these as part of a web interface on top of their data (see example at right).
We can work with you to understand the questions you want answered, the data you have available, and then deliver the analysis either as a document, or as a live interactive web portal that allows you to explore the data for yourself.
Example 2: Service Improvement
In helping a software client to improve its quality of service and reduce wasteful processes, we visualized call service records, to find which connections between departments typically accounted for the biggest delays in responding to customer complaints.
This visualization and analysis helped identify bottlenecks that led to an improvement in service capability, resulting in faster service at lower cost.
In the screenshot to the right, the colours on the horizontal graphs correspond to departments, the widths of bars represent time in the department, and the heights reflect hours of work.
In the screenshot at the left, the most common routings have been displayed as a network map, showing the total amount of time in each department (size of bubble), and the connections between these (cumulative transfers in the cases analysed).