Before you choose your next PAS, decide what kind of insurer you want to be

Owen Greenwood, Client Partner – Insurance & Dan Telling CGO.

Dan Telling CGO in conversation with Owen Greenwood, Insurance Client Partner at Dufrain.

Mergers and acquisitions, legacy technology and the need for new capabilities are all prompting insurers to rethink their policy administration systems (PAS).

But when it comes to modernisation, where should insurers actually begin? And how can they make sure the investment delivers more than simply replacing one system with another?

As part of our series on PAS Migration and data transformation, Dan caught up with Owen Greenwood, Insurance Client Partner at Dufrain, to discuss the decisions insurers should make before embarking on PAS modernisation, the risks worth addressing early and the wider opportunities it can create.


Dan: What’s driving insurers to rethink their policy administration systems?

Owen: It’s a very common topic of conversation. It’s something all of our clients are thinking about at some point, if not on an ongoing basis.

There’s the continual challenge of modernising the overall technology estate, including policy administration systems.

Then you’ve also got M&A activity, which means you may have an estate of policy administration systems that you want to consolidate and simplify.

If you’ve got a big inorganic growth strategy, you’re likely to end up with multiple policy administration systems. How do you make decisions around consolidating those, and which platform should you choose?

Equally, existing systems don’t always get updated as frequently as insurers would like. They may be seeing new systems with features and functionality they’re not getting from their current platform.

So there are several triggers for modernisation.


Dan: When insurers decide its time to modernise, where should they start?

Owen: I think the natural go-to question is often, “Which system should we use?”

And that’s absolutely not the right way to be entering into the conversation.

The first thing to look at is what kind of insurance company do we want to be?

What does our operating model need to look like? What product lines are we going to focus on? How do we want to engage with our clients?

All of those questions around the broader people and process aspects of the business should be the initial part of the conversation.

You really want to be looking at how you want your people to work and engage with clients. And then you can start looking at what technology or solution might support that.


Dan: And what about how a new PAS fits into the wider technology and data estate? 

Owen: You really want to be looking at the overall ecosystem and the holistic picture of what your future state will be.

You need to understand how the PAS sits within that.

What are the limitations of the system you’ve chosen? What are the configuration and customisation options? And you need to be certain that the system is going to be fit for purpose for what you’re trying to achieve.

There are other things to consider around data security, APIs and any integrations you want to make.

All of this needs to be considered when you’re making a selection as part of that broader ecosystem.


Dan: There’s also a risk of simply carrying all that legacy complexity into the new environment, isn’t there?

Owen: Absolutely. With legacy systems, particularly policy administration systems, there’s a whole load of complexity that persists in the estate.

How do we make sure we don’t just lift and shift all that complexity? How do we actually modernise as we go through this?

If you think of it purely as a migration of data from A to B, from legacy to target, you’re really doing yourself a disservice in terms of the opportunity.

A policy administration system migration is the perfect opportunity to modernise your approach to business, but also to get your data into better shape.

It’s a prime time to be looking at improving data quality, implementing data governance and data management, and fundamentally changing the way you do business with the right tool selection.


Dan: We’re consultants, so we probably use the word strategy a lot! But how important is getting that upfront thinking right?

Owen: I’d say it’s really important. If you’re looking to maximise the business case and get real value from your migrations, you’ve got to spend that time and effort upfront looking at the overall strategy.

If you’ve got 20 legacy policy administration systems, there’s absolutely no need or reason to migrate them all wholesale.

You want to look at each PAS individually. What data does it contain? Which systems can you let run down? Which ones might need some of the data migrated while leaving other data behind?

You don’t necessarily always need to migrate all the detailed risk data, for example. You may only need to capture some of the higher-level policy information.

Once you’ve systematically worked through the systems in scope, you get a much clearer picture of the true scale of the migration.

What needs to move? What can be left to run down? What can be decommissioned?

That helps you get a more accurate view of where the value is and when it can be delivered.


Dan: What should insurers be asking about their data before making those decisions?

Owen: You want to get as much detail as you can, without boiling the ocean. Pragmatism is key.

How many policies does each system hold? What kind of policy data is there? Which policies are still live, and which have come to the end of their life?

Which records need to be retained, and which can be deleted or archived? You also want to understand the data quality, the data model and the structure.

From there, you can start to build an accurate picture of each PAS and what the actual migration requirements are.

There are dependencies to consider too, from a business, technology and delivery perspective. From a business perspective, can you simplify your products or lines of business? From a technology perspective, what’s the volume and quality of the data? And from a delivery perspective, what’s the right sequence? What dependencies exist between different data elements?

All of that needs to be considered.


Dan: These can be sizeable investments. How can insurers prove the approach before committing to a major programme? 

Owen: These are big decisions. Migration programmes can run for years, depending on the number of policy administration systems involved and the complexity of the data.

London Market and specialty lines data, for example, can be horrifically complicated and very challenging to migrate.

So it’s fairly common, and good practice I’d argue, to run a proof of concept or proof of value around the migration.

You want to test that the approach you’re planning to take will actually deliver the value you’re expecting.

That means taking a thin slice of the overall data you’re looking to migrate.

But when I say a thin slice, you want to get as much of the full end-to-end breadth of a product line as possible, including some of the edge cases.

That allows you to test the core of the migration and make sure the approach is going to work in the way you expect.

You’re proving to yourselves that the solution will work and deliver the value you want from the future state.


Dan: And how can insurers avoid starting from scratch every time they migrate another system?

Owen: PAS migrations are a significant part of what we do at Dufrain, and we’ve developed a clearly defined approach that we call our PAS migration factory.

We develop a migration engine that takes data from the source, transforms it, improves the quality, changes the data model and structure, and migrates it into the future state.

Once we’ve built that engine, we can run the migration again and again as part of development and testing.

That might be 10, 20, 30 or even 100 times or more if needed.

So when it comes to the final migration, we’ve potentially already run it hundreds of times.

But importantly, we’ve also got a tool that can be applied to future migrations for that client.

There will often be some variation in the customisation or configuration needed, depending on the next source system. But we’re not starting from a standing start.

A significant portion of the engine is already there, and that can make subsequent migrations much more efficient.


Dan: Beyond replacing the system itself, what other business value could insurers realise?

Owen: Often a PAS migration forms part of a broader transformation programme.

It’s the right time to be looking at data quality and data management, but also how you might make that data available elsewhere.

You might take a copy of the data or build pipelines into Microsoft Fabric, Databricks or another platform.

You can then use that data to gain other insights and valuable information about your business and your customers, and use it to drive other business decisions.

So the migration is a key part of what your ambition may be, but it’s also a great opportunity to use that data to drive other business value.

You’re moving beyond policy lifecycle management into customer management and predictive customer management.

And there’s a benefit for customers too.

If you’re speaking with a broker who has a full understanding of your insurance needs and portfolio, you’re going to have a much more seamless and valuable conversation.

Having that data available and accessible can benefit other parts of the business too.


Dan: Finally, if there’s one thing insurers should take away from our conversation, what would it be?

Owen: When we speak with clients, they’re often concerned about risks, costs and timeframes.

And that’s understandable. These are big, complicated and potentially risky migrations to undertake. But I’d encourage clients to look at this differently.

It’s a real opportunity to drive value and transform your business. If you take the right approach, you can do it quickly, effectively and efficiently, while reducing the risk.

Those concerns are entirely understandable, but they can be managed if you approach it in the right way and see it as the opportunity it really is.


Look before you leap – how Dufrain can help. 

PAS modernisation involves more than selecting a new platform and moving policy data. The decisions made early can influence migration complexity, cost, risk and the value insurers ultimately realise.

At Dufrain, we help insurers assess their legacy estates, plan and deliver PAS migrations, and identify opportunities to improve data quality and make better use of their data.

For some of our FAQs on PAS Migration – view Owens recent article here. where he answers 12 common questions about migration strategy, data complexity, testing and repeatable migration approaches.

Or get in touch with our insurance team to discuss your PAS modernisation plans.

View some of our client stories here.