Digital: On Demand Video Streaming
Our Retail client wanted to build an On-Demand Video Streaming service that was deployed on their digital platform for their global customers. The purpose of this project was customer retention and acquisition. Our team of Digital consultants and programmers created apps for mobile devices and integrated it with the online platform. On the back-end a common Content Management System was built that was supported on Oracle architecture. The legacy system application was migrated to the new web platform with new functional specifications. This also involved migrating the POS system to the new Oracle architecture.
Data Analytics: Upgrading Interactive Voice Response System
Our Banking Client wanted to harness the power of Big data to understand real time customer interactions across multiple disparate systems and the IVR for an Omni-channel view of the customer journey. The purpose of the project was Self Service improvements by researching the processes and inputs required to serve customers within the IVR. Our Big Data analysts studied the system for prompt variations, recognition parameters and data collection points to find out how the system was creating a barrier to the customer journey process. They used Hadoop, Pig, R Studio for ingesting and analyzing the data using Mango DB. Once we identified the bottlenecks, we presented our analysis to the client and the system was upgraded the next generation IVR.
Big Data Analytics: Identifying Fraudulent Data
Our client's customers were getting incorrect, fraudulent charges on their accounts. The costs resulting from these anomalies were huge and impacting our client's customer retention rate. They wanted us to identify the problem quickly whenever a fraudulent transaction happens through predictive modelling. Our team carefully assessed this opportunity and came up with strategy to source data sets from different applications into the Big data platform architecture built on Hadoop that would ingest multi-structured data streams from transactional records, chat, social media and other business process data sources. This data was used to scan real time transactional records and using machine learning algorithms they were able to spot anomalies and report to customers.