In a recent issue of Tech Decisions, my colleague John A. Koepke, senior executive in Insurance at Accenture, talks about the importance of data migration to the industry. Whether acquiring another business, upgrading a legacy system or outsourcing a business function, it is inevitable that carriers will need to deal with migrating data from one platform to another.
The importance of data migration
Accenture has been studying data migration for the insurance industry for more than a dozen years. Our experience has shown us that most companies underestimate the amount of time and effort required for successful data migration.
Challenges to data migration include inaccuracies or inconsistencies in data added over many years by many people. Further, the expertise required to deconstruct a legacy system is not always found within a business, making this process more difficult.
Successful data migration is fast and smooth. Migrating data sooner, rather than later, means that the return on technology investment—as well as other benefits—can be realized sooner. A smooth data migration is trouble-free, reproducible, traceable, and of high quality.
Five key elements for successful data migration
Data migration is a complex process, and can be mitigated by keeping five key elements in mind:
- Data quality is paramount. The higher quality the data, the more successful the migration. Complex tests are applied regularly to ensure the quality of the data.
- Good records are important. When Accenture consults on data migration initiatives, we use a secure web portal to provide clients with all records, analysis and reports of the entire process.
- Speed is good. The faster data migration happens, the sooner the benefits can be realized. By automating as many steps as possible, the process is streamlined and made more efficient.
- Reuse is encouraged. The framework for a successful data migration can be applied toward other applications, lowering the cost of the project
- Testing is critical. Dry runs and data testing are critical to the viability of the actual production data. These testing routines must be run numerous times and on a regular basis to identify exceptions, confirm data relationships and provide early warning of any data quality issues.
Implementing a successful data migration
Insurers can reduce the cost and risk associated with data migration by applying an industrial model that can be applied to other processes. This process must be proven, flexible and minimize the required effort from key resources.