The internal IT organization of one the largest professional services firms asked us to help build a custom sales dashboard to view key client and portfolio-level financial and operational metrics. The existing sales dashboard was using aging technologies to pull and aggregate sales and revenue data from slow internal systems. Some of these systems were built for data and report processing rather than real time web use, so the existing dashboard was slow.
We joined their team of developers to help architect and build out the services layer with the explicit goal of sub second response times. Our team built out the caching layer to enable storage of data from downstream systems; we also built processes to periodically refresh data from the internal sources as needed, including real time data from their enterprise service bus. Furthermore we built complex business logic to do financial calculations, real time currency conversion and helped set the standard for modern single page applications within the group.
The application was built using Web API and .NET OData web services surfacing 10+ internal data sources. To implement the caching layer we used ElasticSearch and leveraged its ability to store large amounts of unstructured data and aggregate it at considerable speeds. We also wrote Java plugins for ElasticSearch to deal with currency conversion and other computations prior to aggregation at the caching layer. The front end was written in Angular.
Our team worked on the project expanding the feature set and user base for 2.5 years helping establish it as a reference architecture for modern single page applications within the organization.