The data manager's job has never been easy, often presenting significant challenges, including data system rewrites, data security, regulatory compliance, and reporting. And the digital age, with a myriad of new and innovative data sources and more sophisticated analytic models, presents its own unique hurdles to implementing a successful data-management and data-quality program in the modern insurance enterprise.
Insurance data-management practices have evolved to include both a strategic and an operational focus across the entire organization. Additionally, insurers are facing key data-management challenges-such as data integration and the lack of satisfactory granularity-especially as demands for analytic solutions continue to evolve at a rapid pace.
Some of the key data and data-management issues facing insurance data managers today are:
- Senior management buy-in to the concept that data is a valuable asset
- Protection of proprietary business intelligence
- lack of adequate detail for more sophisticated analytics
- Timely system changes and data capture
- Matching data across diverse sources within an organization to create information more valuable than the sum of its parts
- Maintaining a coherent and coordinated enterprise-wide data-quality program
- Ensuring easy access to a central repository of issues related to corporate data-integrity
- Attracting and retaining appropriate personnel with the specialized skill sets needed for superior data management
Many issues have not changed much over time but have been magnified by the emergence of better technology and advanced analytics. The explosion of predictive modeling in the insurance industry has brought the issue of insufficient detail to the forefront over the past few years.
The data refinements needed for insurance business applications have added to privacy and confidentiality concerns. Also adding to data-management concerns are publicized security breaches, regulatory reporting requirements, and senior management's recognition of the critical role data plays as a corporate asset.
With the increased focus and use of analytics across the insurance industry, the need to build a consistent, synchronized, innovative, and scalable data-management function is imperative for the future operational performance and growth of insurers.
One of the most important factors for sustaining high-quality data and successful data-management practices is the attention by senior management teams within the insurance industry. A significant trend over the past two decades has been the recognition by C-level executives that data quality and data management are crucial to the achievement of both short-term tactical plans and long-term strategic goals.
Most successful participants in the property/casualty insurance industry understand that effective data management and high-quality data are keys to the creation of actionable information and analytics. Significant senior management visibility and commitment through major investments in the infrastructure have helped to implement data-management best practices within the industry's flourishing enterprises.
It has been said many times that data is the lifeblood of the insurance industry's products and services. As competition accelerates and the pressure increases to "right-price" risk, many insurer management teams are redoubling their efforts to develop and maintain high-quality data at specific levels of detail-especially for use in predictive models designed to refine policy pricing.
Management's understanding of the importance of data quality is evident in their questions. They are asking data stewards and managers about availability and usability of data to make informed business decisions, address regulatory compliance, and conduct strategic planning. For the most part, senior management appreciates that quality is not simply the validation of values within a specific data field or the relationships across related data fields. Indeed, data integrity is now the responsibility of all users, be they collectors, processors, aggregators, or analyzers. Therefore, clear standards and procedures for data capture, deployment, and reporting for all users within an organization are essential.
As the focus on quality and sound data management increases, a shift from a purely operational approach to a blended operational and strategic approach has emerged. Escalating sources and quantities of data are being demanded to satisfy the ever-expanding world of analytics. While purely operational demands for these functions still exist, many organizations are looking beyond individual projects toward strategic issues such as best practices to produce information that will yield competitive advantages.
In that capacity, modern data management has evolved from being independently focused on a particular tier of analysis to stressing a more holistic methodology. An increasingly incisive and strategic approach to insurance data quality allows users and management to review data from the perspective of how quality issues will cascade through multiple analyses. Additionally, external pressures have amplified scrutiny on completeness and accuracy of data. Such scrutiny often comes from the actuarial community in the form of Actuarial Standards of Practice and Disclosure requirements. It is also a result of the regulatory community looking for information to help fulfill their oversight responsibilities.
Insurance data management is changing in a fundamental way because of the heightened emphasis on on-demand analytics and predictive models. As competition increases within the industry, analytics and modeling are playing a much larger role in separating market winners from losers. A major challenge for data managers is to keep pace with the progressively complex data environment evolving from the new and innovative uses of data.
At the same time, those developments emphasize the shortcomings in the granularity of the data, the quality of the data, and the difficulties in manufacturing data by linking the various sources of information available in today's information age. Contemporary data sources in use by the insurance industry can include:
- Structured, semi-structured, and unstructured data
- Text, graphic, and pictographic resources
- Multimedia material
- Geospatial, climatologic, and atmospheric information
To implement decision-support analytics successfully within an organization, data managers must be part of the process, even lead, in overcoming the shortfalls and challenges of integration. Well-conceived and implemented data-management programs do have costs associated with them. However, the benefits are apparent within a relatively short period of time. Increased internal and external data needs, coupled with the inevitability of tight resources, reinforce the need to develop practical data-management methodologies. Especially in these uncertain economic times, properly supported initiatives will allow insurance organizations to maintain the data infrastructure needed to drive real-time strategic and market-focused decision making.