Representative End User Clients
Representative Automation Clients
Representative Software Clients
Client Profile: Pitney Bowes Inc. (PB) is an American provider of global e-commerce solutions, shipping and mailing products, location intelligence, customer engagement, and customer information management solutions. Based in Stamford, Connecticut, the company has approximately 16,100 employees worldwide. It is one of 87 existing firms that have been members of the S&P 500 since its creation in 1957.
Business Need: The company provides various services. The customer information management solution helps businesses manage customer information and the location intelligence solution helps companies convert geographic data into better customer experiences and smarter strategic decisions. The customer engagement helps transform the way clients engage with their customers - driving consistent interactions and targeted orchestrated connections. The shipping and mailing solution helps the clients increase the speed of mail process while cutting costs and the global e-commerce removes barriers in cross-border shipping. The data system in the company consisted of multiple data sources. The size of the data was huge considering the services offered and the disparate sources made analytics a difficult task. The data received was transformed into reports (roughly 10) by the system. Though static with limited information, these reports were essential in making business critical decisions. Hence, most of the analysis was done manually. The company wanted to integrate the disparate data into a single view and make this available to the stakeholders. The goal was to derive a centralized data store that gave real-time actionable information, was interactive, and provided insights into status, SLAs (service-level agreement), hub performance, and so on.
Solution Implementation: The team worked on the project for about 10 months to derive a solution. The system consisted of a complicated parcel process flow that had to be tracked. The process began with sellers shipping their goods to the PB hub; a lot of information in the form of status and other data would also originate from this. The PB team had to track all this and make sure that all the formalities such as export compliance and so on were completed. Once the goods were sent across to the requested destination, they would have brokers to clear these shipments at the customs. One of the features of the platform they required was to provide a complete view of the tracking - the placement of the parcel in this entire network and other data had to be tracked in real time. They wanted a centralized database to bring all this together. With this goal, the team set to work on a Big Data solution. They set out on a technology agenda using a combination of machine-to-machine technology to collect and source data. They deployed a Hadoop based Big Data solution to collate the data from multiple data sources and also built a data warehouse. They did this by collecting the data through Big Data platform. The team then added API management over the Big Data platform in order to make information available. For analytics, they used this data to get the desired outcome. They implemented production machine learning technologies to solve various problems in the data, such as the harmonized standard classification problems. They created data warehouses and worked with the businesses to identify the KPI (key performance indicator) dashboards they would like to see. They built a warehouse using a set of technologies using Big Data technology stack. The team also built a series of web accessible dashboards for business and operations teams to monitor key KPIs. The dashboard users could interact and have self-service drill down capability.
Benefits: With this solution the company was able to provide dashboards that were interactive and provided much greater flexibility to use. The dashboards provided the current status within four SLA categories real time - delivered in SLA, delivered out of SLA, undelivered, undelivered but in SLA. The reports could be filtered in real time by the partner, to gain information such as - which outbound hub, destination country, and vendors/shippers. This helped collate the data and process it through the data warehouse. This also helped the company in getting real-time analytics and real-time alerts.
Keywords: ARC Advisory Group, Orlando Forum 2015, Pitney Bowes Inc., Big Data Analytics, IoT