Claims Management Application Development: What It Costs and How Long It Takes
The phases, costs and realistic go-live dates of a custom claims system build, and how much of it a configurable PAS already has in place.
A dashboard nobody opens is wallpaper. The dashboards insurance teams use every day, the data each one needs, and how to tailor them in your PAS.
Monday, 8:45. The renewals team signs in and the system opens on the dashboard the vendor shipped: written premium for the year, a pie chart of policies by product, a line of new business by month. All of it is accurate, and none of it is what the team came in to do. They need the 37 policies that expire in the next 30 days without a renewal offer, so someone runs an export, filters it in a spreadsheet and starts working down the list.
By Wednesday the spreadsheet is out of date. Nobody looks at the dashboard again.
No two insurance businesses run the same workflow. An MGA writing three programs for two carriers checks binder limits and bordereau deadlines every week. A regional carrier's billing desk lives in its overdue invoices. A broker's account managers start the day with renewals and open claims across a dozen insurers. A policy administration system that gives all of them the same screen gives each of them something to work around.
The fix is to tailor the system to the work rather than the work to the system, and dashboards are the cheapest place to start. They are the first screen people see, and they decide whether the PAS is where work happens or where it is recorded after the fact. A tailored insurance policy dashboard turns the record everyone already maintains into the list each desk works from, which is where the efficiency gain actually shows up: fewer exports, fewer status calls, nothing worked twice.
Generic vs Tailored Insurance Dashboard
A count tells someone there is work; a sorted list that clicks through is the work.
Three things separate the dashboards people use from the ones they skip. Each belongs to one desk and is named after its job. It carries a queue, a sorted and filtered list that clicks through to the record, with the counts around it. And it is cheap enough to change that the team fixes it rather than working around it. That last property decides the other two: if a change is a development ticket, the dashboard stays generic.
Most operations need five or six dashboards, one per desk, plus the view a policyholder gets. The table is the short version; the sections below list the data points each one has to read.
| Dashboard | Who opens it | The question it answers |
|---|---|---|
| Underwriting and referrals | Underwriters, underwriting managers | What is waiting for a decision, and how long has it waited? |
| Renewals and servicing | Account managers, servicing team | Which policies need action before they lapse? |
| Billing and collections | Finance, credit control | Who owes what, and what is overdue? |
| Claims | Claims handlers, claims managers | What came in, what is stuck, and what is it costing? |
| Portfolio and management | Leadership, MGA principals | Is the book growing profitably, product by product? |
| Policyholder dashboard | Customers, in a portal | What am I covered for, what do I owe, where is my claim? |
An underwriting dashboard needs the work in front of the desk, ordered by urgency, with the reason each case is there.
This is the dashboard that would have saved the Monday spreadsheet.
Collections depend on timing, so this board only works if it reads live data; a list from last night's job gets the same customer chased twice. Our comparison of insurance billing software covers the systems behind this desk.
The claims management desk reads the most volatile numbers in the business, so this board earns its place by being current.
Leadership is the one audience that does want totals, as long as they are broken down far enough to act on.
The customer's view is a dashboard too, usually inside a client portal, and it is the one with the most measurable effect on retention. J.D. Power's 2025 Claims Digital Experience Study, reported by Insurance Business in December 2025, found that insurers deliver adequate digital updates on a claim just 22% of the time. Among customers who rated their digital claims experience poor or just OK, 52% were likely to leave or not renew; among those who rated it excellent, 4%.
Desk dashboards answer “what needs doing”. The other question, asked on every call, is “what is going on with this policy”, and answering it usually means four systems and a reference number read out letter by letter.
Openkoda answers it with a Policy 360 view: one page per policy, built from the same live record that billing, claims and documents use. A band of four numbers sits at the top, the ones a caller asks about first: annual premium, balance due, open claims and days to renewal. Two of them are alarms rather than facts, since a balance that is not zero and a claim count that is not zero both mean a longer conversation.

Below the band, in the order someone on a call needs them, come the coverages with each limit and deductible, every instalment with what is paid and a button to record a payment, the claims on the policy, each period's history, and every document and email produced for it.
What a Policy 360 View Brings onto One Page
Sections of the Openkoda Policy 360 view.
What makes the view customizable is the action bar at the top. Endorse, cancel, renew, file a claim and record a payment are there by default, but none of them is hard-wired into the page: each is a small workflow attached to the product. Adding a new one, such as an Email documents to client button that sends the current certificate to the insured, takes a button step, a message step and an end step in the workflow editor. The Policy 360 tutorial builds exactly that, in fourteen steps.

So a request like “we always email the documents when a customer asks” stops being a feature request for a vendor and becomes an afternoon's work for the person who wants it. The same applies to the fields shown, since the data model behind each product is configuration too.
Desk dashboards in Openkoda are built from seven kinds of tile: a list, a number, a chart, a saved report, a note, the viewer's own tasks and quick-action buttons. Each tile points at an entity in the PAS, such as policies, invoices or claims, and the board is arranged on a twelve-column grid by dragging. A report written in plain English with Reporting AI can be saved and embedded live, and a whole board can be generated by describing in a sentence who it is for and what they care about. Any dashboard can then be made the screen the team lands on when they sign in.

None of this needs a developer. The person who knows what the billing desk needs builds the billing desk's board, and changes it when the desk's work changes. The dashboards tutorial walks through one board tile by tile.
The trade-off is the same as for the platform as a whole. Openkoda does not carry the decades of accumulated functionality of the large suites, and a team that wants a vendor to design every screen for it will find more ready-made screens elsewhere. What it gives instead is customization with no ceiling, on a policy management system whose data model, workflows and screens are shaped to the business rather than the other way round.
Whatever the desk using it acts on today: for underwriting, referrals and their age; for renewals, policies expiring without an offer; for billing, overdue invoices. A good dashboard pairs a few counts with a sorted, filtered list that clicks through to the policy record.
The customer's view of their own insurance, usually in a client portal: cover and limits, documents, payment schedule, claim status and renewal date. It cuts routine calls and, according to J.D. Power, the quality of digital updates is closely tied to whether customers renew.
In a configurable PAS, yes. In Openkoda, dashboards are assembled from widgets that point at policies, invoices or claims, arranged by dragging, or generated by describing them in plain English. In less configurable systems, new dashboards and fields still go through the vendor or an integrator.

The phases, costs and realistic go-live dates of a custom claims system build, and how much of it a configurable PAS already has in place.

Four ways to build a policy management app: vibe coding, a greenfield build, an enterprise suite or a configurable PAS with AI, compared on speed, cost and risk.

Lemonade's bot handles about 55% of its claims. Chubb wants 85% of major processes automated. What insurance automation software runs today, and where people stay.
Book a live, personalized demo with our product team - tell us your use case and see the platform work with your data. No commitment.