In fact, most companies tend to underestimate the amount of effort needed to gather, cleanse and standardize legacy data to be migrated to the new systems. Therefore, it is important for insurers to get a handle on their data early to give them as much time as possible to maximize data quality, integrity and usefulness to the business. While any list of key principles runs the risk of oversimplifying the complex process of data migration, we have identified five key elements shared by every successful data migration initiative.

Data Quality Rules: The quality of the migrated data is the most important factor in a successful migration. Complex unit and assembly tests are run prior to each delivery of converted data. Design conditions are monitored with each delivery, as well, with exceptions raised and analyzed. Automated tools and processes, along with repetitive, industrialized deliverables allow the insurer to focus on complex conversion design challenges.
Keep the Process Traceable: We use a secure web portal to provide clients with direct access to all deliverables, including data analysis, design documents, and validation reports. Design documents are written in what we call "pseudo-code" feeding directly into the build; clients have total visibility in the design and, therefore, in the development of conversion routines. Meanwhile, thorough documentation of the conversion fulfils audit, data privacy and client requirements.
Fast is Good: We accelerate the delivery timeline whenever and wherever possible, for instance by reducing the time spent on migration planning and using automated tools from design to test based on full code generation. A higher degree of automation allows for a shorter migration timetable and provides better abilities to adjust to changes and address evolving issues.
Re-use Encouraged: When properly implemented, data migration efforts have the potential for a high degree of re-use beyond the immediate conversion objectives. The data conversion framework and execution should be tied to an overall application rationalization that lowers the total cost of ownership.
The proof is in the data: It is important to execute dry runs and converted data delivery to support system tests on the full volume of production data. While data quality diagnostic tools can be utilized to perform initial data cleansing, it is important to have the capability to repeatedly execute conversion routines on the production data set numerous times prior to the actual production conversion. This will enable to identify exceptions, confirm data relationships and allow for future data quality issues to be addressed prior to adversely affecting customers.

In our experience, policy conversion can be a divisive topic. Different groups within the organization bring different perspectives and expectations regarding the benefits of conversion; sequencing of events; the conversion timeline; and the resources that should be devoted to the effort. Typically, there is little enthusiasm for major mobilization of resources, with insurers preferring to allocate resources to other transformation efforts.

Special Reports /

Data Migration Done Right: Lessons from the Factory Floor

Related Stories

Recommended For You