“The insurers that are pulling ahead are not the ones with the most sophisticated platforms. They are the ones where data management has become a boardroom conversation, not a back-office one.” Owen Greenwood, Client Partner, Insurance.
Data management has quietly become one of the biggest determiners of whether insurance transformation delivers outcomes or stalls in local optimisation. The leadership move is not more technology. It is establishing data management as an enterprise capability.
This is the second article in our Data in Insurance series. In Part 1, we explored data governance as the accountability layer for insurance transformation. Here, we move into the broader subject of data management: the discipline of making data usable, protected and fit for purpose across the organisation. If governance sets the direction and rules, data management is what helps insurers apply them in practice and turn them into business value.
“Every hour spent remediating poor data downstream is an hour that should have been spent on the business problem. Strong foundations do not slow transformation. They are what let it accelerate without falling over.”
Why data management is a transformation issue, not a hygiene programme
In complex insurance organisations, fragmentation is normal. Legacy stores sit across business units. Platforms have been built over long timeframes with limited interconnectivity. Data is handled differently across teams, product lines and functions. The result is familiar: transformation creates pockets of progress, but not always consistent enterprise outcomes.
That is why data management matters.
The challenge for leaders is not simply getting data onto a platform. It is making that data trustworthy, protected and usable across the organisation in a way that scales. In insurance, that matters more than ever because data is not a side issue. It sits at the heart of pricing, underwriting, claims, customer servicing, reporting, compliance and increasingly AI.
A useful way to explain the difference is this: governance is to data what project governance is to a programme. Data management is the broader discipline, more comparable to project management. It covers the wider planning, structure, execution and control needed to make the whole thing work.
The outcomes insurance leaders are really trying to realise
Data management should be assessed by outcomes that matter to leaders across the business, not by the data team’s internal activity.
- Better decision-making: Greater confidence in the data behind pricing, underwriting, claims and reporting, with fewer manual reconciliations and less debate over which version of the truth is right.
- Speed with control: Faster operational delivery without relying on exceptions, workarounds or manual fixes that create risk later.
- Information protection that stands up: Controls that are adopted, evidenced and applied consistently, especially where sensitive personal or medical data is involved.
- Less operational friction: Less rework caused by poor quality data, unclear ownership, inconsistent standards or disconnected processes across teams and systems.
- AI readiness: The ability to use structured and unstructured data safely and responsibly, with governance and evaluation built in from the start rather than bolted on at the end.
These are not data team outcomes. They are enterprise outcomes.
5 leadership moves to kick-start data management in insurance
1. Start with the business outcome, not the data activity
The best data management programmes do not begin with policy documents, technical inventories or abstract framework design. They begin with the outcome the business is trying to improve.
For an insurer, that may be more reliable pricing and underwriting decisions. This could mean faster, more consistent claims handling, stronger oversight of sensitive customer data, or the ability to move ahead with AI use cases because the organisation is clearer on what data it holds, where risk sits and how information is protected.
The point is not to make data management sound bigger than it is. The point is to make it relevant. Leaders should be asking: what business outcome are we trying to unlock, and what data management capability needs to exist to support it?
2. Treat data management as an enterprise framework, not a local fix
One of the biggest risks in insurance transformation is local success that never scales.
A business unit improves one process. A function creates a new control. A programme cleans up a specific data issue. But because ownership, standards and operating approaches differ across the group, the benefits stay local.
That is why data management has to be approached as an enterprise framework. Not because everything needs to be centralised, but because enterprise outcomes depend on consistency, repeatability and clear accountability.
3. Prove value in a bounded area first
Going global too early often creates disruption, resistance and unnecessary complexity. The stronger pattern is to prove value in a defined area first.
That might mean focusing on a claims data problem, a pricing and underwriting domain, or a high-risk unstructured data estate. The objective is not to stay small. It is to test what good looks like, demonstrate that better controls improve outcomes without stopping the business, and create evidence that supports wider rollout.
4. Let evidence shape priorities
Data management improves faster when leaders can see where risk, value and exposure actually sit.
That means understanding which quality issues affect decision-making, where access controls do not reflect real-world use, where retention rules are inconsistently applied, and where sensitive data is concentrated in ways the organisation cannot yet evidence or explain.
This is especially powerful in insurance because the biggest issues are often hidden in operational reality rather than policy. Evidence-led approaches help leaders prioritise what matters most instead of applying blanket controls that create noise, friction and low adoption.
5. Build adoption and repeatability from the start
If every rollout depends on heroic effort, it will never scale.
Sustainable data management depends on clear ownership, practical guidance, consistent definitions and an operating cadence that can be repeated across business units. Introducing policies without education invites workarounds and resistance. When accountability is unclear, risk grows and delivery slows, and when every area invents its own method, enterprise value stays out of reach.
For insurance groups, repeatability is what turns progress into capability.
“We have never seen a governance programme fail because it started too small. We have seen many fail because they tried to solve everything at once and solved nothing.”
A note on unstructured data: the accelerant for both value and risk
Unstructured data in insurance is not just documents. It is operational content tied directly to customer outcomes and regulated processes: claims notes, emails, call recordings, underwriting submissions, complaint files and policy correspondence.
This is often where sensitive data concentrates, retention risk hides, and AI ambitions quickly collide with the reality of control and evidence. It is also where insurers can unlock significant value if they approach it properly.
That matters for responsible AI too. Responsible AI becomes a drag when governance is bolted on at the end. When it is embedded into architecture, evaluation criteria and human oversight from the start, it stops being a brake and becomes part of how better systems are built. In practice, that makes responsible AI less a values-only discussion and more a risk, trust and operational readiness issue.
What this means for senior leaders now
For insurance leaders, data management is increasingly judged on whether it enables transformation that scales across business units, supports speed with control, strengthens evidence for risk and compliance oversight, and creates the conditions for AI adoption where it can deliver genuine business advantage.
Done well, data management becomes a business capability that reduces friction, improves trust and creates the foundations for responsible innovation. Done in isolation, as a series of local projects without a connecting framework, it produces pockets of progress that are hard to evidence, hard to scale and hard to sustain.
Is your data management capability building enterprise value, or managing a collection of local problems?
Most insurance organisations already have data management activity in place. The harder question is whether it is operating as a connected enterprise capability or a collection of local efforts that do not add up to scale.
Our Data Governance & Management Maturity Assessment uses a proven framework to identify where your data management capability is creating drag, where risk is concentrated, and what a scalable path to outcomes looks like, scoped to your organisation’s size and complexity.
Talk to us about our Data Governance Power Hour or Maturity assessment.
If you’d like to discover more more from our series:
Part 1: Data governance is the accountability layer for insurance transformation