Centralize Data to Accelerate Core System Replacement
By Jack Plunkett
So, you've invested in a modern core system. Congratulations. As you are no doubt aware, this is the critical first step in your business transformation, ensuring swift reaction to market changes, and allowing for faster delivery of new products to market. But ties to the old days prevent completion of the transformation. The truth is, you simply can't let your legacy core system go — not because of an emotional tie; rather, it still houses your all-important historical data.
Unfortunately, far too many insurers find themselves in this position. They are ultimately blocked from achieving true business transformation because they are bogged down with worries about what data to pull from their legacy systems, how to sustain legacy systems to access important data that cannot be migrated, and how to support the business during transition. The good news is that insurers have the option of a few different approaches to achieve their data migration goals. However, some approaches will yield more favorable results than others.
The Best Approach for You
Insurers typically have three options for data migration, and their choice depends greatly on their business culture, goals, and appetite for risk. The following provides a bird's-eye view of these options, along with pros and cons to help inform your decision.
Option 1: Non-Conversion/Migration
For those insurers that fear the risk and effort associated with data migration may be too great, this approach may be best for you. With the "Non-Conversion/Migration" option, insurers leave their legacy systems running and treat them as data sources. Over time, however, the costs involved in not sun-setting the legacy system will continue to grow, as will associated risks involved with trying to keep outdated systems functioning. From a total cost perspective, this is the most expensive option long term as the cost of maintaining the legacy burden continues indefinitely, and as noted, this cost will continue to grow.
Option 2: Full Conversion
A bit more aggressive, the "Full Conversion" approach sources data from multiple locations and populates the new core system data model. Typically, a Full Conversion is attempted after the initial core system go-live due to the challenges associated with migrating to a constantly evolving data model. However, this means that the data migration must take place into a live system, with all policies and/or claims and billing in process, which presents its own set of challenges.
On the plus side, insurers can retire their legacy systems after the data migration is complete, but the complex system's ecosystem that consists of proliferating leads and feeds will not materially change and will incur ongoing expenses until modified. From a cost perspective, this approach will require significant conversion programs that will not be further leveraged and the company has not lowered its overall maintenance costs as these have shifted from the old systems to the new one. Ongoing maintenance burdens have been shifted but not reduced and will grow over time as additional technology is introduced into the stack.
Option 3: Operational Data Store (ODS)
The third option involves extracting the data from all necessary sources into an operational data store or integration repository. With an ODS, insurers can populate the new core system starting with renewals, and use the ODS as the new enterprise-wide source for all reporting and functionality that previously required system-to-system or point interfaces. Ultimately, this option will simplify the system landscape and allow an insurer to unlock data from across the enterprise.
However, while this may seem like the ideal approach, it does not come without its own set of challenges. Traditionally, such projects have been hard to get right for many reasons; namely, that it is time-consuming to obtain cross-functional consensus on the ODS design and expensive and time consuming to design one from scratch and perfect. This will typically require a multi-year, multi-functional effort. Once this consensus has been reached and the coding completed and data loaded and tested, the business and its needs can proceed and the company can start to reap the benefits.
After weighing these three options carefully, the ODS approach seems to be a clear winner, with Full Conversion coming in at a close second. But before committing to it, consider yet another option: a pre-built ODS that integrates with the new core system.
The Power of One
Ultimately, insurers must select the option that best suits the culture and goals of their business. If they are looking to gain profound operational insight from their new core systems to transform transactional data into knowledge that will guide their daily operations and better address strategic decisions, centralizing their historical data in a pre-built ODS is an ideal choice. Starting with a pre-built ODS; insurers can concentrate on optimizing for a competitive advantage rather than building an enterprise data model from scratch.
With this approach, a central hub is established that unifies, standardizes, validates and stores data from the former patchwork of legacy systems as well as external sources. As the leads and feeds are corrected to come from the central source, those legacy systems can finally be turned off. With this environment simplified, ongoing maintenance is reduced and key processes and functionality is now permanently insulated from technology change. When the ODS is integrated with the new core system, historical and external data can now be conformed, organized, and centralized, and users may take advantage of the advanced analytics and BI tools that are available today.
Effective data management is no longer a nice-to-have. It's mission-critical. In order to transform your business and remain competitive in today's fast-paced insurance market, insurers must optimize and fine-tune their entire business, not just the front end running on core systems. Modern data management infrastructure—namely, a centralized, enterprise-wide operational data store—integrated with a new core system, completes the transformation and sets the savvy insurer on a sure path to continued success.
Jack Plunkett is a P&C technology executive with over 35 years of experience in the industry. He is a senior industry advisor for Guidewire Software and has spent the preceding 16 years as the president of Millbrook. Prior to co-founding Millbrook, Plunkett spent much of his career with Royal Insurance (now Royal Sun Alliance) as well as serving as the CIO for several regional insurance carriers.