Review, then revolutionize
Map the current system through analytics
We needed to step into customers' shoes to experience how they buy Thermo Fisher Scientific products and services. Our first step was to create a blueprint that mapped all customer service processes and defined a detailed taxonomy that identified gaps and pain points.
Using our diagnostic tools, analytics, automation, and transformation methodologies – including ProcIndex, the Intelligent Automation Index, and Celonis' process-mining tools – we assessed the customer service value chain, made recommendations, and built the business case. That's how we created a transformation roadmap, setting out the process-improvement priorities using digital tools to bring in a refined automated ordering system.
But first, consolidation was key. We created a centralized data lake that brought data together from multiple ERPs, CRMs, and legacy systems from years of M&A growth into a single layer. This data lake underpinned the entire project.
We created a 'single pane of glass' that shows the entire lifecycle of an order with integrated track and trace. Workflow technology called Cora Orchestration – deployed on web service Amazon EC2 – replaces the multiple screens employees had previously used to track orders. Now employees can use a smart search platform on a single dashboard to pull information from the data lake, be it customer orders, product availability and pricing, customer-specific quotes, or invoices.
Our order management system, Cora OrderAssist, is now in place. Built on cloud-computing platform Amazon Web Services (AWS), it uses Amazon EC2, Amazon SageMaker, AWS Lambda, and Amazon Redshift to create a scalable solution that allows ThermoFisher to focus on business outcomes, rather than IT infrastructure.
By using document-automation technology from Esker, Cora OrderAssist automates processes and extracts and validates order information. Customer care representatives now have fewer orders to key in manually. And if there are exceptions or order changes at any point, machine learning and dynamic workflow manage them automatically.