Claims teams spend too much time managing the process instead of the claim
Claims is where the promise of insurance becomes real. It's also where you find out, very quickly, how an insurer actually operates.
When a customer lodges a claim, they're not experiencing your transformation roadmap or your target operating model. They're experiencing the mechanics of how work moves through the business. How quickly someone can find the right information. Whether decisions move cleanly between teams or get stuck somewhere in the middle. Whether anyone can confidently explain what happens next.
In many insurers, that experience is still far more operationally difficult than it should be. Not because people aren't trying. Usually the opposite; there are capable, experienced people working hard to hold things together. The problem is the shape of the environment they're working in.
Claims is complex work, but the way many insurers run it adds to the burden
Most claims leaders already know this, so why is it that claims still feel so administratively heavy, even after years of investment in digital programmes, workflow tools, rules engines, process redesign, and platform modernisation?
Part of the answer is obvious enough - volumes increase, customer expectations increase, regulatory scrutiny increases. But underneath all of that is something more structural. For many insurers, claims is still being run as a collection of operational steps stitched together by people, rather than as one connected workflow.
You see this in the amount of manual coordination required to keep claims moving. You see it in the unofficial spreadsheets people maintain because the "real" operational view of the claim doesn't fully exist in any one system. You see it in the quiet institutional knowledge that sits in people's heads because no system captures it reliably. You also see it when the payment side of a claim has to be managed separately from the assessment itself, creating extra handoffs between claims, finance, approvals and customer communication.
All this is a problem of course, because when experienced people leave, they take the IP and the workflow with them.
When work is scattered across systems, coordination becomes the work
The operational pain in claims doesn't come from one large system failure, but from accumulation; small frictions, repeated constantly across the lifecycle of a claim.
Policy details in one place, documents somewhere else, workflow in another tool. Payment approvals happening separately. Reinsurance tracking sitting outside the claim file. Audit history technically existing somewhere, but reconstructing what actually happened requiring time and institutional knowledge that not everyone has.
When every individual component exists but in separate places, claims teams become human integration layers. They chase documents because workflow doesn't actually manage requirements. They manually reconcile statuses between systems because nobody fully trusts the handoffs. They maintain side trackers because they need a usable operational view of what's happening. They translate assessment decisions into payment actions because those two things live in entirely different operational worlds. They check whether a payment has been approved, whether it has been released, whether it has failed, whether a recovery applies, and whether anyone has updated the customer or adviser.
After a while, organisations normalise this. It becomes "just how claims works."
I think the industry has accepted too much of that friction for too long. And I say that not as a criticism of the people running these operations (usually they're doing something genuinely impressive to keep things moving) but as an observation about where the effort goes. A lot of very capable people are spending significant parts of their day managing the process rather than working the claim.
Why this matters more now
For years, a certain amount of claims inefficiency was tolerated. As long as claims were being assessed and paid within acceptable timeframes, operational friction was a known cost of doing business.
That's becoming harder to justify for several reasons. Operational load has increased, and even modest volume growth creates disproportionate pressure when the underlying workflow is labour intensive. You can't absorb extra claims by working harder if the coordination overhead already accounts for a substantial portion of your team's capacity.
Customer expectations have also shifted permanently (and will shift even further as Gen Z enters the purchasing window for insurance). Customers don't mentally separate "claims servicing" from the rest of their experience with an insurer. They just notice whether communication is clear, whether progress feels visible, and whether the organisation appears to know what it's doing. Once a claim has been accepted, payment becomes one of the clearest signals of service. Delays, unclear status updates, or disconnected payment processes can quickly undermine confidence, even when the assessment decision itself was handled well.
And then there's automation and AI, which brings me to something I think is important, and that I don't see discussed enough. A lot of insurers are exploring AI-assisted claims operations. Some of those pilots are genuinely interesting. But many are being built on top of fragmented operational environments where workflow, financial activity, governance controls, and operational context are still disconnected across systems. What's the saying? AI on top of chaos is still chaos.
You can apply AI to isolated pieces of a fragmented process. But you don't get meaningful operational improvement from doing that. What you mostly get is automation that sits on top of the same coordination problems that already exist, except now those problems are slightly harder to see because part of the process looks modern.
My view is that the insurers who benefit most from automation over the next decade won't necessarily be the ones with the most ambitious AI programmes. They'll be the ones that quietly improved their operational foundations first, and then found that automation had somewhere useful to land.
The friction usually sits between steps, not inside them
One thing I've noticed in claims environments is that the biggest slowdowns happen between systems and teams rather than inside them. Some examples:
- An assessment is completed, but the financial process begins somewhere else.
- A reviewer picks up a claim and can't immediately see what changed since the last touch.
- A task sits in a queue, untouched, because ownership across teams isn't clear.
- Finance approvals happen by email because the workflow itself doesn't enforce the control.
- Reinsurance recoveries are tracked outside the claim record, so settlement timing becomes harder to manage as the claim progresses.
- A claim is ready for payment, but the team still has to confirm payee details, approval status, payment method and benefit splits.
- A payment is made, but the claims team cannot easily see the full payment picture alongside the claim record.
Individually, none of these sound like massive problems. Collectively, they determine the operational reality of a claims team every day. They also shape customer outcomes in ways that compound quickly, particularly in situations where the customer is already under stress and has little tolerance for feeling like their claim has disappeared into a process they can't see.
This is why I think claims modernisation discussions sometimes focus too heavily on interfaces and customer channels while underestimating operational workflow design.
The more important question is whether a claim can move from notification through assessment, review, payment, and closure within one connected operational flow, or whether the organisation is still relying on manual coordination to fill the gaps between systems.
Four things that tend to characterise claims operations that run well
There's no universal claims operating model, as every insurer carries different product complexity, governance structures, and technology constraints. But when you listen to the views of seasoned claims professionals, a few patterns seem to appear.
The claim has to become the actual unit of work, not the handoff between systems
This sounds obvious until you look closely at how many environments still operate. Many insurers technically have a claims system, but the real unit of operational work remains the handoff - the email, the queue item, the spreadsheet row. Operationally, people are still stitching together context mentally every time they touch a claim.
The stronger environments organise around a single operational case that contains workflow, assessments, requirements, notes, tasks, documents, financials, and status history together in context. When that's working, the question teams ask changes. Instead of "which system do I need to check?", it becomes "what's happening on this claim?" That sounds like a small thing, but in practice it changes the day quite significantly.
Requirements need to be part of workflow control, not administrative tracking
Most claims teams already know what evidence they need. Medical reports, certificates, forms, supporting documentation; the list usually isn't the problem. The problem is that requirements still sit in notes, checklists, emails, or in someone's memory, rather than functioning as active operational controls.
That creates inconsistency quickly. Claims progress before information is complete. Review quality varies between assessors. Operational discipline depends too heavily on individual vigilance rather than being designed into the process.
When requirements are embedded directly into workflow logic (visible on the claim, able to prevent progression while mandatory information is outstanding) you stop relying on the right person having the right habits.
Assessment and payment need to operate as one connected flow
The separation between claims assessment and claims payment creates a surprising amount of operational drag, and it's a place where visibility tends to break down.
Assessors complete their work, and then the financial process begins elsewhere - different systems, different queues, different ownership. What's approved, what's pending, what's blocked, what's been paid, and why: in many environments, getting a clear picture of all of that requires going to several places.
The stronger operations connect lodgement, assessment, provisions, approvals, payments, and reinsurance positions into one operational sequence. The visibility that creates matters operationally, and it matters for governance. There's also a practical benefit for complex cases that run for months: when the financial story of a claim is visible in context, the people managing it make better decisions.
Usability is not a separate concern from operational performance
Claims teams don't work in neat, linear process flows. They triage constantly and they switch context. They revisit the same claims repeatedly over weeks or months. They manage a simple life insurance claim at the same time as a sensitive income protection case that requires careful handling. They deal with interruptions all day.
A surprising number of claims systems still seem designed around the assumption that users calmly progress through one structured sequence at a time. That disconnect creates operational fatigue and operational fatigue creates errors, shortcuts, and the kind of slow degradation of process quality that's hard to measure but easy to feel.
Good operational design here is less about making screens look modern and more about reducing cognitive friction. Clear queues, visible ownership, fast access to context, sensible task flow, fewer hidden dependencies. When those things improve, teams spend less time managing the process and more time actually working claims.
The same is true for claims payments. If people have to leave the claim record to understand payment status, approvals, exceptions or reconciliation, the system is adding cognitive load at one of the most sensitive points in the process.
Modernising claims is not really about replacing screens
A lot of claims transformation programmes still approach this primarily as a technology replacement exercise. New portals, new interfaces, new workflow tooling.
Sometimes that genuinely improves operations. Sometimes it relocates existing fragmentation into newer systems - and the operational experience doesn't actually change much, even if it looks different.
A better indicator is whether the workflow itself is becoming more connected. Can a claim move from notification through assessment, review, settlement, and closure without constant manual coordination? If the answer is no, the friction is still there. It just looks more modern on the surface.
None of this means insurers need to achieve perfection immediately. Most carry years of operational complexity and historical process decisions that can't be untangled overnight. But there are specific areas where reducing fragmentation tends to create disproportionate value: workflow ownership, requirements handling, task visibility, financial controls, auditability, payment orchestration, reinsurance visibility. These are the foundations. When they improve, a lot of other things (including automation) become easier to do well later.
Where I think this is heading
Claims will always be operationally demanding work. It involves vulnerable customers, financial decisions, governance requirements, and day-to-day operational execution. That complexity doesn't disappear.
But the gap between modern customer expectations and the operational reality inside many claims environments is widening, and it's becoming harder to ignore. There’s a good chance that the industry's push toward automation will expose those workflow weaknesses more clearly over time, not less.
The next generation of claims operations, I think, will look less like a collection of specialised systems and more like connected operational environments built around workflow visibility, configurability, and governance. Claims payments will need to be part of that same connected environment, linked to assessment decisions, approval controls, financial positions, reinsurance activity, customer communication and audit history.
Not because those ideas are fashionable, but because they make the work easier to run well, and because the alternative (highly skilled people spending their days managing the process rather than working it) is becoming increasingly hard to sustain.
When claims operations work properly, customers barely notice the machinery behind the process at all. They experience clarity, responsiveness, and a sense that the organisation knows what it's doing, at exactly the point where that matters most.
About the Authour ~
Tim von Dadelszen is the Chief Product Officer at Simfuni; an insurtech orechestrating modern life insurance operation.