Appendix A. Case Study Conclusion

ETI has successfully developed the “fraudulent claim detection” solution, which has provided the IT team experience and confidence in the realm of Big Data storage and analysis. More importantly, they see that they have achieved only a part of one of the key objectives established by the senior management. Still left are projects that are intended to: improve risk assessment for applications for new policies, perform catastrophe management to decrease the number of claims related to a calamity, decrease customer churn by providing more efficient claims settlement and personalized policies and, finally, achieve full regulatory compliance.

Knowing that “success breeds success,” the corporate innovation manager, working from a prioritized backlog of projects, informs the IT team that they will next tackle current efficiency problems that have resulted in slow claims processing. While the IT team was busy learning enough Big Data to implement a solution for fraud detection, the innovation manager had deployed a team of business analysts to document and analyze the claims processing business process. These process models will be used to drive an automation activity that will be implemented with a BPMS. The innovation manager selected this as the next target because they want to generate maximal value from the model for fraud detection. This will be achieved when it is being called from within the process automation framework. This will allow the further collection of training data that can drive incremental refinement of the supervised machine learning algorithm that drives the classification of claims as either legitimate or fraudulent.

Another advantage of implementing process automation is the standardization of work itself. If claims examiners are all forced to follow the same claims processing procedures, variation in customer service should decline, and this should help ETI’s customers achieve a greater level of confidence that their claims are being processed correctly. Although this is an indirect benefit, it is one that recognizes the fact that it is through the execution of ETI’s business processes that customers will perceive the value of their relationship with ETI. Although the BPMS itself is not a Big Data initiative, it will generate an enormous amount of data related to things like end-to-end process time, dwell time of individual activities and the throughput of individual employees that process claims. This data can be collected and mined for interesting relationships, especially when combined with customer data. It would be valuable to know whether or not customer defection rates are correlated with claims processing times for defecting customers. If they are, a regression model could be developed to predict which customers are at risk for defection, and they can be proactively contacted by customer care personnel.

ETI is seeing improvement in its daily operations through the creation of a virtuous cycle of management action followed by the measurement and analysis of organizational response. The executive team is finding it useful to view the organization not as a machine but as an organism. This perspective has allowed a paradigm shift that encourages not only deeper analytics of internal data but also a realization of the need to incorporate external data. ETI used to have to embarrassingly admit that they were primarily running their business on descriptive analytics from OLTP systems. Now, broader perspectives on analytics and business intelligence are enabling more efficient use of their EDW and OLAP capabilities. In fact, ETI’s ability to examine its customer base across the Marine, Aviation and Property lines of business has allowed the organization to identify that there are many customers that have separate policies for boats, planes and high-end luxury properties. This insight alone has opened up new marketing strategies and customer upselling opportunities.

Furthermore, the future of ETI is looking brighter as the company embraces data-driven decision-making. Now that its business has experienced benefit from diagnostic and predictive analytics, the organization is considering ways to use prescriptive analytics to achieve risk-avoidance goals. ETI’s ability to incrementally adopt Big Data and use it as a means of bettering the alignment between business and IT has brought unbelievable benefits. ETI’s executive team has agreed that Big Data is a big deal, and they expect that their shareholders will feel the same way as ETI returns to profitability.