CREATION OF BIG DATA STRATEGY – THE CASE OF UNITED HEALTHCARE
Table of Contents
Introduction 2
Business Strategy for Big Data 2
Business Initiatives, Tasks and Objectives 4
Requirement of Technology Stack 5
MDM and Data Analytics to Support Decision System and Business Intelligence 7
NoSQL for Big Data Analytics 8
NoSQL Databases and Utilization in Big Data 9
Role of Social Media in the Decision Making Process 9
Big Data Value Creation Process 10
Conclusion 11
References 12
Introduction
Serving almost 51 million people, United Healthcare had a crucial requirement of big data for better understanding the healthcare needs of the customer and for maintaining integrity in payment. This ensued adequately paid claims and prevention of fraudulent claims by the use of analytics. There are several well- suitable use cases for big data solution in healthcare. Some research or academic focused institutions of healthcare either use big data in advanced research or consider to experiment it (Violino, 2015). Such institutes are drawing upon graduate students, statisticians, and data scientists for wrangling upon the complications of big data. However, it is worth noting that there are various challenges to be addressed in big data regarding its overall current distributions. The two biggest challenges involved from the use of big data in healthcare are requirement of technical expertise in using it and absence of integrated, robust security that surround it (Manzini & Maranesi, 2014).
Since recent years, the ability of technologies in business intelligence for providing predictive, current and historical views regarding business operations on the basis of collecting, extracting and analyzing business data for the improvement of decision making has enhanced (Raghupathi, 2014). Moreover, big data analytics and big data are stirring the interest of practitioners and researchers similarly. This report will conduct a detailed market research and present how there can be use of big data in the overall scope of business intelligence and decision support system for United Healthcare. The report will clearly map the business strategy of United Health with its business tasks, objectives and initiatives. Further ahead, the report will define the requirement of technological stack and analytics and data architecture for big data. The report will also address the necessary requirements of advanced analytics for supporting the selection of business strategy. Understanding the role played by social media in decision making process, the report will discuss about the process of value creation in big data.
Business Strategy for Big Data
Data are crucial but competitive edge and performance improvement arising out of analytics models allow the managers of United Healthcare for predicting and optimizing outcomes. Carrying more significance, the approach with maximum effectiveness to build a business strategy begins by the identification of business opportunity and determination of areas for improving the performance (Escobar, 2014). By the help of such hypothesis led strategy, United Healthcare develops root models and faster outcomes in practical relationships of data that can be understood broadly by the managers.
(Fig: The Estimated Journey of United Healthcare with Big Data Business Strategy)
(Source: Raghupathi, 2014)
The strategy should have a clear understanding about the specific decisions made by managers for aligning the initiatives with wider company objectives. Adequate communication with the frontline managers of United Healthcare will confirm that the tools and analytics are as per the current decision systems. Such a big data driven business strategy will help in the effective management of a number of trade-offs. There can be varying efforts that are highly dependent on the desired time line and goals of the company. Adjustment in mindsets and cultures typically hold the requirement of multifaceted approach including role modelling and training from the leaders. There can be specific metrics and incentives for the reinforcement of appropriate data driven behavior (Demirkan & Delen, 2013).
Business Initiatives, Tasks and Objectives
The fundamental idea of United Healthcare lies in ensuring that on receiving a claim, the company should make payment of correct value that also includes the prevention of fraudulent claims. In the current scenario, the company has the ability of identifying mis-paid claims in a consistent and systematic manner. The company is being encouraged for embracing analytics of big data and moving forward to broaden the scope of aspects such as customer experience, care provider networks and clinical networks. Following are the key business objectives for the big data business strategy (McVay & Skaife, 2014):
Based on factory, focus on the identification of inadequate claim
With the help of Ayasdi Models, True Fraud and Claims Review, provide a number of potential fraudulent cases that have to be tackled by the company
In accordance with the entire IT strategy, an effective strategy of healthcare data analytics will be aligning with each and every need of the business. Issues related to analytics include the requirements of human resources, software and hardware for supporting the needs of analytics at United Healthcare (Mehrjoo & Pasek, 2014). The useful analytics should be strongly aligned with the business and clinical information that decision makers use. There must be well- functioning of data analytics and technology for mining every key insight out of the data. There will be incorporation of the following components for the big data business strategy of United Healthcare (Violino, 2015):
Context of Quality and Business
Users and Stakeholders
Data and Processes
Techniques and Tools
Training and Team
Infrastructure and Technology
United Healthcare should focus on enhancing its presence on social media as there is an inherent interrelation between peer-to-peer decision making and human relationships. As a result of social media, there is often superseding of proximity in the factor of trust through like mindedness or relativity. This is extremely crucial for enhancing the success related to big data management and optimization being aligned with the business operations of United Healthcare (Chung, 2014). There is significance of course based security specifically in the setting of healthcare. If there is going to be placement of PHI data in a platform of Big Data, there is a crucial requirement of considering capabilities of low level encryption for performing it at the level of IO. Security still lacks robustness and maturity in the platforms of NoSQL (Rosemann, 2010).
Requirement of Technology Stack
There must be measurement of current, long term and short term analytical needs at United Health prior to acquiring the related capabilities. The exact choices of technology will be highly dependent on the overall needs that includes the necessity and requirement of analytical insight. Such choices may also end up hinging over the resources of the organization similar to a budget in payment for people and systems with the appropriate skills for performance of work (Escobar, 2014). The below table provides a detailed overview regarding all components of technology stack.
Component
Category
Description
Data Sourcing
Application
Most applications of big data will be requiring data source from other interfaces and databases.
Analytics
Processing or Application Type
A common application of advanced analytics helps indicting the requirement for solution of big data.
Operations (HDFS)
Processing or Application Type
Big Data is able of supporting the needs of operation such as risk management and complex event processing of real time for monitoring the patients.
Distributed Processing
Processing or Application Type
This is the core process of big data in which there is execution of task on several computers in a grid or cluster.
Representation (Amazon Simple Storage Service)
Infrastructure
There are a number of way for representing data in a platform of big data such as relational, graph, key value combinations, and wide column stores.
Persistence (NoSQL)
Infrastructure
This indicates how there will be persistence of data or if it will be present in the platform of big data. When massive real time data is processed, there will be a need for persisting the entire data by simply grabbing and processing the needs. There can be utilization of MPP RDBMS, distributed filesystem, and NoSQL for the persistence of big data.
Platform (Hadoop MapReduce)
Infrastructure
There can either be use of proprietary or open source databases and software, and proprietary or commodity hardware. It will also be a good option to leverage a private or public cloud.
Security (Apache Hive)
Management
There is significance of course based security specifically in the setting of healthcare. If there is going to be placement of PHI data in a platform of Big Data, there is a crucial requirement of considering capabilities of low level encryption for performing it at the level of IO. Security still lacks robustness and maturity in the platforms of NoSQL.
Development and Management
Management
There are a number of different technologies of proprietary and open source management and development that will be involved in the equation of Big Data.
Table: Requirement of Technology Stack
MDM and Data Analytics to Support Decision System and Business Intelligence
Master data management is known for representing the objects of business shared throughout a number of applications in transaction (Shaban & Tronci, 2014). Such a data is known for representing the objects of business for the execution of transactions. Maximum value of the business is related to the management of analytical and transactional master data. Under the scope of MDM, the cleaning of operational data helps in improving the efficiencies of operational applications and the involvement of business process for using such applications (Raghupathi, 2014). Oracle, supported by acquiring Hyperion, will be considered for providing the most comprehensive solution of MDM on the current market. The fixture of poor quality of data from the source and management of constant change is the core base of MDM solution. MDM solution of Oracle will be referred by United Healthcare for eliminating poor quality of data under the heterogeneous landscapes of IT application (Elder, 2010).
The MDM of Oracle will be providing strong pre-built models of data for supporting architectures with service orientation and operational workloads (Packard, 2016). It will be helpful in providing tools like secure and fast parameter of search engines, prevention and elimination, duplicate recognition, survivorship on data attribution, management of hierarchy, standardization of data, synchronization of data, and management of real time change (Demirkan & Delen, 2013). There will be employment of interfaces to data augmentation of third party while addressing the providers of standardization. There will be construction of cross- references for golden centralized data and federated data. There will be availability of quality customer data for applications of enterprise resource planning and management of customer relationship. In addition, quality product data will be provided to ERP applications and product lifecycle management department of United Healthcare. There are several efforts involvement for dealing with the fundamental issue of business intelligence as spread across the market (Turetken, 2014).
NoSQL for Big Data Analytics
The landscape of big data includes a number of buckets. There can be application of MapReduce and Hadoop programming for offline data where there is batch based processing. The category of NoSQL is instead course and is inclusive of several data models such as key value store, graph data, document store, and column data store (Violino, 2015). In this case, the model of document store will be providing a wider functionality and scale. The technology of cloud computing will allow CIOs for lowering the overall expense of ownership when new systems and technologies are deployed and tested. With the rise of operational expenses, TCO is continuously being a major concern for the respective authorities (Gillet, 2015). Without the involvement of new capital costs, there can be deployment of new technologies through the cloud.
There can be utilization of NoSQL database as the platform-as-a-service of big data for private deployment of cloud. Some of the external sources of data will be regional health information, and health information organizations, health information exchanges and national health information network (Chen et al., 2012). There can be varying efforts that are highly dependent on the desired time line and goals of the company. Adjustment in mindsets and cultures typically hold the requirement of multifaceted approach including role modelling and training from the leaders. There can be specific metrics and incentives for the reinforcement of appropriate data driven behavior. In accordance with the entire IT strategy, an effective strategy of healthcare data analytics will be aligning with each and every need of the business. Issues related to analytics include the requirements of human resources, software and hardware for supporting the needs of analytics at United Healthcare (Chung, 2014).
NoSQL Databases and Utilization in Big Data
NoSQL databases are known to be supportive of dynamically designed schema with the offer of key potential for increasing customization, scalability and flexibility in comparison with relational software (Raghupathi, 2014). This makes the databases appropriate for systems of content management, Web applications and additional uses that involve non- uniform data in large amounts with the requirement of often updates and variation of field formats. Being highly dependent on the business issue being solved, the IT decision makers of United Healthcare will be comparing the advantages of NoSQL relational and software databases. The IT professionals involved in the decision making process require a careful evaluation regarding if the available options of NoSQL are in alignment with the business requirement. NoSQL databases can be utilized for the purpose of supporting business analytics, cloud computing and big data applications (Demirkan & Delen, 2013). There can also be deployment of NoSQL tools for offering advice regarding the decisions and processes of selecting and implementing technology. Though there can be a significant use of database technology under NoSQL in environments of big data, there is a crucial requirement of relational databases and data management platforms of different types.
Role of Social Media in the Decision Making Process
United Healthcare is working in an environment in which it has less control on the reputation of its own services, products and brand as the opinion of the customers dictate the management and sale of this reputation. By the overall utilization of social media, prospects and customers appear to be having an extremely instantaneous platform to discuss about the overall knowledge, experiences and ideas (Dumbill et al., 2012). At an increased rate, the utilization of social media has started to play a crucial role in the process of decision making as these are used as mediums and tools for engaging them in the process of decision making. The social aspect for decision making has started to enhance this strength impressively while coming in connection with the various dynamics of communications, marketing and management of customer relationship in United Healthcare (Grant, 2016). Information holds the capacity of travelling at the velocity of business than ever before. SMS, email, professional networks and practical communities are among crucial tool enabling multi-channel accessibility for suppliers, partners, customers and employees.
United Healthcare and its operations have involvement in the long promised international collaborative and virtual work environment (Escobar, 2014). Professional networks and online communities are arguably change the way in which business in conducted with the viral creation of new ecosystems as per different communities and their personal preferences. Professional networks are responsible for facilitating vast networks, connections and interactions of individuals as they are enabled to collaborate at all points of time. United Healthcare should focus on enhancing its presence on social media as there is an inherent interrelation between peer-to-peer decision making and human relationships. As a result of social media, there is often superseding of proximity in the factor of trust through like mindedness or relativity. This is extremely crucial for enhancing the success related to big data management and optimization being aligned with the business operations of United Healthcare (Malik, 2013).
Big Data Value Creation Process
There are four main internal steps required by United Healthcare in order to harness big data for the creation of value. These steps are execution of data, experimentation of data, contextualization of data and democratization of data (Manyika, 2010).
Democratization of data is referred as the step of integrating data throughout the company for allowing maximum employees to understand and access information as required in a specific time. The overall data and its volume can appear challenging but there is scope of alleviation when the benefits to extract insights out of data are borne by employees. There will be transmission of significant knowledge throughout the business for enabling collective and wider application of data as a positive association for better creation of value (Raghupathi, 2014).
Contextualization of data is the step of assignment sense as an approach for interpretation of data in all of the executed actions. United Healthcare will be collecting different categories of data related to changing needs of customers, shifting preferences, demand in the market, and behavior of customer. This is crucial for identifying context based clues in gaining a holistic perception of the customers.
Experimentation of data is the step of promoting continuous experimentation and “trial and error” with the data while monitoring the main changes. It is suggested that the organizational culture based on trial and error, when combined with a higher scope of data access, provides a better chance for the transformation of value out of big data (Schroeder, 2017).
Finally, execution of data is the step of transforming insights of data in actions to identify the new opportunities for increasing the engagement of customer and hence, create value. There are differences in the execution of big data insights across firms with major dependence upon the execution ability of United Healthcare (Mandl & Kohane, 2014).
Conclusion
Big data has changed the business processes of decision making and the evolution is still taking place. However, as big data exceeds the capabilities and capacity of analytics, reporting and conventional storage, there is a significant demand for new approaches of problem solving. As strong advanced, computing database technologies converged with social networking, mobility and wireless data, there lies a possibility of bringing together and processing big data as a number of profitable approaches. As reflected in the report, all constituents of United Healthcare will be affected by big data for predicting the behavior of these players, encouragement of desirable behavior, and minimization of behavior that is less desirable (Sherman & Imhoff, 2015). In big data, there can be an inexpensive and quick optimization, refinement and testing of related applications which will further be supportive in radically changing the research and delivery of healthcare.
The leverage of big data will surely be involved as a solution to control the spiraling costs of healthcare. There will be a number of implications of this strategy for payers, researchers, providers and other constituents of the healthcare business. This will significantly impact the engagement of people with the ecosystem of health, specifically when there is an involvement of social networking, mobility, globalization, regionalization and external data (Violino, 2015). There can be collection of external data from a number of different medical systems spread across various countries and regions. Effective work on disparate repositories of data will help United Healthcare in identifying best practices and local knowledge, while leveraging them at regional and global level. While progressing with certain initiatives, there will be more availability of external data which will further turn out to be a challenge of integration.
References
Chen, H., Chiang, R. H., and Storey, V. C. (2012). Business intelligence and analytics: From big data to big impact. MIS quarterly, 36(4).
Chung, J. H. (2014). Big-data applications in the government sector. Communications of the ACM, 57(3), 78-85.
Demirkan, H., and Delen, D. (2013). Leveraging the capabilities of service-oriented decision support systems: Putting analytics and big data in cloud. Decision Support Systems, 55(1), 412-421.
Dumbill, E., Croll, A., Steele, J., Loukides, M. and Slocum, M. (2012). Planning for big data. 1st ed. O’Reilly Media, Inc.
Escobar, G. (2014). Big data in health care: using analytics to identify and manage high-risk and high-cost patients. Health Affairs, 33(7), 1123-1131.
Elder, G. (2010). Comparative cost-effectiveness analysis of two MSF surgical trauma centers. World journal of surgery, 34(3), 415-419.
Gillet, D. (2015, May). Graspeo: a social media platform for knowledge management in NGOs. In Proceedings of the Seventh International Conference on Information and Communication Technologies and Development (p. 63). ACM.
Grant, R. M. (2016). Contemporary Strategy Analysis Text Only. John Wiley & Sons.
Malik, P. (2013). Governing big data: principles and practices. IBM Journal of Research and Development, 57(3/4), 1-1.
Manyika, J. (2010). Clouds, big data, and smart assets: Ten tech-enabled business trends to watch. McKinsey Quarterly, 56(1), 75-86.
Mandl, K. D., & Kohane, I. S. (2014). Finding the missing link for big biomedical data. Jama, 311(24), 2479-2480.
Manzini, R. and Maranesi, F., (2014). A decision-support system for the design and management of warehousing systems. Computers in Industry, 65(1), pp.175-186.
McVay, S.E. and Skaife, H., (2014). Does ineffective internal control over financial reporting affect a firm’s operations? Evidence from firms’ inventory management. The Accounting Review, 90(2), pp.529-557.
Mehrjoo, M. and Pasek, Z.J., (2014). Impact of product variety on supply chain in fast fashion apparel industry. Procedia CIRP, 17, pp.296-301.
Packard, R. M. (2016). A history of global health: interventions into the lives of other peoples. JHU Press.
Raghupathi, V. (2014). Big data analytics in healthcare: promise and potential. Health information science and systems, 2(1), 3.
Rosemann, M. (2010). Handbook on business process management. Heidelberg: Springer.
Ramanathan, U., (2014). Performance of supply chain collaboration–A simulation study. Expert Systems with Applications, 41(1), pp.210-220.
Shaban, A. and Tronci, M., (2014). Inventory control system based on control charts to improve supply chain performances. International Journal of Simulation Modelling, 13(3), pp.263-275.
Schroeder, J. (2017). UnitedHealthcare | MapR. [online] Mapr.com. Available at: https://mapr.com/customers/unitedhealthcare/ Viewed: 5th September 2017.
Sherman, R. and Imhoff, C. (2015). Business intelligence guidebook. 1st ed. Waltham, Massachusetts: Morgan Kaufmann.
Turetken, O. (2014). The current state of business intelligence in academia: The arrival of big data. CAIS, 34, 1.
Violino, B. (2015). How Big Data Keeps United Healthcare Nimble. [online] Health Data Management. Available at: https://www.healthdatamanagement.com/news/how-big-data-keeps-unitedhealthcare-nimble Viewed 6 September 2017.