Debunked: The top five data migration myths

Autor: Arjan Vis, Delivery Manager Migration, Keylane

It’s no secret that implementing a new core insurance system can be a massive undertaking, with the dread of data migration at the forefront of concerns. Moving policies, customer records, payments and other critical assets from your old environment to a new system is of the utmost importance.

Keylane’s experience in mission critical implementations has delivered more than 15 million successful policy migrations, and with that in mind, I’d love to talk about some of the most common myths I hear in the insurance industry, as well as offer my industry insider’s perspective on them.

Myth #1: “Moving policies is easy, you shift them from A to B”

I hear this quite often in initial meetings. However, the reality is that data cleansing is always necessary when undertaking policy migrations. Many, if not most insurers, take a policy migration as an opportunity to pre-process data, such as splitting policies, combining coverages from multiple policies and/or creating group contracts and packages.

Another aspect adding to the complexity is that policy data is context sensitive. Your new data structure needs to work with all systems that are using it, such as integration with vehicle registration authorities. An automated process needs to be developed in order to decommission insurances and re-register them with updated and correct information.

A successful migration strategy ensures data consistency across different systems and provides a full view of data, dependencies and business context. A solid understanding of your current data model is critical when it comes to mapping source records to a new target structure. How well documented is the current data? Is it up to date? Is technical expertise available in-house?
The Keylane migration methodology offers a structured and proven approach, and can unpack any legacy data, as well as transform it to new formats that support a core system migration with full data integrity.

Myth #2: “Off-the-shelf data migration tools achieve the same results”

While it’s true that there are many tools and applications available that can assist in data migration projects, such as ETL (extract-transform-load) tools, Robotic Process Automation (RPA), data governance and testing software (to mention a few), the reality of policy migration and management is far more nuanced, and all such tools have limitations and dependencies.

Setting up and configuring these tools may seem less technically challenging and demanding at face value, but implementing them to full effect requires a good knowledge of the starting point and the desired state, as well as an understanding of data dependencies, and exceptions in how to define the necessary data validation rules. Additionally, ETL tools have the shortcoming of not being able to understand the full context of the data. This means they won’t be able to transfer and maintain the relationship between payment data, reserve, documents, tasks and memos. This is a concern because, for a migration to be truly successful, these assets would typically need to be solved for with custom developed tooling.

The Keylane migration team makes use of standard software and a dedicated team specialised in custom development for process automation and validation. And the entire migration ecosystem, be it tools, software and data, are managed in a separate environment to ensure data security and privacy regulation. This makes it possible to guarantee migration quality, no matter how big and complex the scope is.

Myth #3: “All historical data and policies need to be migrated”

The belief that all non-current and older versions of policies need to be migrated is a very common industry myth. And while the Keylane migration methodology can easily accommodate for this where desired, most Keylane clients choose to only migrate current policy versions.

In the case of claims that are linked to older policy versions, they can be can easily retrieved, though given the infrequency with which older versions are needed (rarely, if at all), it’s a difficult business case to make, as the additional costs and time are hard to justify.

Myth #4: “You only need to think about policy migration right before go-live”

Before your business switches over and makes the leap to running your insurance processes on a new system, you will need to have all your data securely loaded into your new system. This is of course obvious. But to get to a point where you can rest assured that the new systems have successfully integrated all necessary data, you must have a solid migration strategy in place that comprehensively outlines the approach, resources, project phases, testing plans, validation and rollback procedures. Therefore, planning ahead is crucial to a successful migration, and failing to give adequate time to policy migration planning is a recipe for failure and inevitable disruption.

Axon implementations safeguard against these risks by including and setting up a day one policy migration project long before the date of planned go-lives. Keylane’s specialists also design, execute and refine the migration process throughout the entire implementation project, and run migration on the product database every two weeks, ensuring data quality issues and mapping errors are reduced to an absolute minimum.

Myth #5: “We have to do policy migrations ourselves”

It’s a fair point to say that that it doesn’t really matter who is responsible for policy migration, as long as it’s done on-time, within scope and meets all expectations in terms of quality and results. But insurance organisations that do undertake entire migration project themselves more often than not have several major blindspots, and top of list tends to be a severe underestimation of the challenges of sourcing the project correctly, and with the right competencies. A data migration project of such size is mission critical, and having the right resources, technical knowledge and experience is imperative to success. Put simply, an overall lack of experience adds major risks to any policy migration, but these risk become exponentially more acute with scale.

Another blindspot I’ve seen with insurers who go it alone with policy migration is with data reconciliation and regulatory compliance, which is always subject to external verification and intense auditing. Keylane offers peace of mind in regards to these factors with optimised and standardised best practices, guaranteeing compliance with all strict auditing requirements.

So there you are. My top five myths debunked from an industry insider’s perspective. And in all fairness, in an industry as old as insurance, a few myths are to be expected.

When I’m not debunking myths at Keylane, you’ll find me supporting colleagues and clients, while proudly helping to maintain our 100% successful track record for migrating client legacy data, alongside our gold standard approach to serving today’s insurance market.

Keylane guarantees successful migrations on behalf of its clients by enacting a strictly controlled process that includes planning, data cleansing, extraction, transformation, loading and post-migration testing and validation. Our methodical approach is orchestrated in close cooperation with the client, bonding company specific knowledge with proven methodology and tooling.