Stay ahead!
Click a tag to get email alerts when we publish related content.
In-force policy migration: Is artificial intelligence changing the equation?
Published on November 19, 2026
In digital transformation programs in individual life and health insurance, there often comes a point when the initial enthusiasm gives way to a much more difficult question:
“What will we do with the in-force policies?”
Modernizing new business is one thing. Replacing the system that administers half a century’s worth of contracts is another.
It is precisely at this stage that many insurers choose to slow down, break their transformation into smaller phases or keep their legacy systems running in parallel. This caution is understandable. But the emergence of artificial intelligence (AI) could be fundamentally changing the equation.
Migration: The costly last mile
Legacy policy administration systems are not simply aging databases. They contain decades of business rules, contractual particularities, exceptional processing requirements and knowledge that is often difficult to document.
This is what makes migrating in-force policies so complex.
A Deloitte study, Life and annuity insurers consider the path less traveled: Core-system modernization is becoming table stakes, conducted among executives in the life insurance and annuity sector, reveals that fewer than one-third of the insurers surveyed had completed some or all of their modernization initiatives. The study, published in June 2023, also notes that previous attempts to consolidate legacy systems have often proved complex, cumbersome and costly, with some failing to achieve the expected results.
This reality helps explain executives' caution. After investing substantial sums in a new platform, why risk disrupting a portfolio that continues to function despite its aging system?
The legacy system paradox
This caution, however, presents a paradox.
An insurer can modernize its sales processes, automate its operations and significantly improve the experience of advisors and clients while continuing to administer a substantial portion of its portfolio on a legacy system.
The transformation is real, but incomplete.
More importantly, maintaining legacy systems comes at a cost. These systems must continue to be operated, integrated, secured and maintained. The expertise needed to make changes to them is also becoming increasingly scarce as specialists in these technologies leave the workforce. In the same study, Deloitte points out that replacing technology talent is already a challenge for insurers that depend on legacy technologies.
The portfolio does not simply “die” within the system. It continues to require resources to keep it alive.
AI changes the nature of the problem
McKinsey published an article in April 2026, Can agentic AI (finally) modernize core technologies in insurance?, that deserves particular attention.
One of its observations is especially relevant to insurance: much of the difficulty in modernization lies not in building the new system, but in understanding the old one.
Which rule applies to a particular product? Why does this exception exist? Which processing procedure must be replicated? What data drives this decision? Does the behaviour observed in the system actually match the available documentation?
Traditionally, answering these questions requires specialized teams to analyze the system's code, data, specifications and behaviour.
McKinsey suggests that agentic AI can now play a role in several of these activities: reverse-engineering legacy code, understanding rules, mapping data, generating tests, analyzing anomalies, reconciliation and preparing for deployment. Naturally, potential productivity gains can vary considerably depending on the stage of the process and the level of automation achieved.
AI does not eliminate the risks associated with migration, but it could significantly reduce the work required to carry it out.
The business case needs to be revisited
For years, the question has essentially been: “How much will migration cost us?”
A second question must now be asked: “How much is it costing us not to migrate?”
Making this calculation is now essential to making an informed decision.
In another article published in 2019, IT modernization in insurance: Three paths to transformation, McKinsey data shows that insurers with modernized technology systems handle 40% more policies per employee than their peers using legacy technologies. Information technology (IT) costs per policy can also be 40% lower.
The article notes, however, that these benefits can be compromised when older systems are not actually decommissioned (the cost of migration decreases while the relative cost of maintaining the legacy system increases) or when the new platform replicates excessive complexity.
Migration, once perceived as an exceptionally risky undertaking, can thus once again become a rational business decision.
Successful modernization projects in Canada
It would, however, be unfair to portray the migration of in-force policies as a nearly impossible undertaking.
In December 2025, Canadian insurer ivari announced the migration of approximately 732,000 policies from its Ingenium environment to the DXC Assure platform as part of a multi-year modernization program. According to DXC, the migration was completed without business interruption. The company also reported a 22% reduction in operating costs and the launch of four new products following the migration.
UV Assurance provides another example. The Quebec-based insurer undertook the replacement of its legacy policy administration system to reduce manual processes and accelerate innovation. With OIPA (Oracle Insurance Policy Administration) and its accompanying digital solutions, UV Assurance launched six new products and accelerated application processing.
These examples tell an important story: migration is not impossible. However, it requires a structured approach, domain expertise and rigorous control over the migration process.
The end of preconceptions about in-force policy migration?
For a long time, retaining a system such as CAPSIL to administer in-force policies could seem like the most prudent option. Insurers would modernize new business and digital experiences while waiting for the legacy portfolio to gradually decline.
That strategy now deserves to be reassessed.
The issue is no longer simply whether an organization is capable of migrating its portfolio. It is now necessary to determine whether it has the technological means to make that migration sufficiently predictable, manageable and affordable.
AI could be precisely the factor that tips the balance in this decision.
In-force policy migration may no longer be a project to avoid at all costs.
It could once again become an investment whose value deserves to be seriously recalculated.
Stay ahead!
Click a tag to get email alerts when we publish related content.