A leadership monograph
An insurance promise proves its value through a usable payment.
Chief Executive Officer · Australia · Volume I · 2026
Chapter I
In my first insurance job, I spent a morning locating a photograph that had been filed under a supplier's reference instead of the policyholder's claim number. The photograph mattered because it showed machinery before temporary repairs began. The business owner had already sent it. We had already received it. Yet the file appeared incomplete, and the next decision was waiting. I was twenty-two and had little authority, but I could see how an insurer's internal organisation could become somebody else's financial delay.
I joined Fallowmere Claims Services in August 1993 after completing a business degree at North Alvren Institute. The firm handled commercial claims administration for insurers. My early work involved collecting documents, reconciling estimates and arranging visits by loss assessors. I learnt the vocabulary of cover through actual files rather than product brochures. I also learnt to distinguish a fact that had not been established from a fact that had simply become hard to find. That distinction still matters when a management report calls a claim incomplete.
In January 1998 I moved to Kestervane Underwriting as a commercial underwriting manager. The change was deliberate. I wanted to understand why policies had been written in ways that later created difficult claims discussions. I supervised a small team pricing property risks for workshops and light industrial businesses. Some applications lacked the detail we needed; others offered reassuring descriptions that did not survive a site visit. I became interested in the quality of an insurance question: whether it elicited an answer that could support a future decision, rather than merely complete a form.
Orleston General Insurance recruited me as head of commercial claims in July 2004. The portfolio was larger, and the organisation had more specialised functions. That brought expertise but also opportunities for responsibility to fragment. Claims staff could wait for legal clarification while members waited for funds that were not in dispute. I introduced a review of such cases, initially concentrating on machinery and stock losses. We had to separate an appropriate partial payment from a premature conclusion about the entire claim. That distinction required experienced judgement and a record of what remained unresolved.
I returned to Kestervane Underwriting in January 2010 to lead its eastern regional business. The appointment put distribution, underwriting and service results under one remit. It also exposed a limitation in my earlier thinking. I had assumed that better explanations would resolve most disagreements. Sometimes the economic choice itself was unacceptable to a customer, however clearly it was explained. We withdrew from a small segment where the proposed terms no longer represented a credible proposition. Losing volume was preferable to selling a product that needed continual exceptions to remain attractive.
In September 2015 I joined Brindleharbour Mutual as chief operating officer. Member ownership appealed to me, but I was cautious about the language surrounding it. A mutual still has to price risk, maintain capital and make unpopular decisions. Its distinctive question is how those decisions are justified to people who are both customers and owners. I worked on claims hand-offs, broker service and repairer capacity. We consolidated certain administrative activities while retaining specialist judgement close to the circumstances of a loss. The programme involved 410 colleagues and an A$86 million annual operating budget.
I became chief executive in July 2020. Today I report to the mutual's board and sit on it as an executive director. I lead an eleven-person executive team and remain accountable for 780 colleagues, an A$1.28 billion annual gross premium book and A$2.4 billion of investment assets at June 2026. I do not present investment assets as money freely available for operating choices. Their purpose and obligations require the finance and actuarial disciplines that support the insurer's wider balance sheet.
The decisions that have occupied me most concern competing forms of fairness. A claimant wants certainty and prompt cash. A member renewing after several claim-free years wants an intelligible price. Members who will suffer losses next year need the institution to remain capable of paying them. Those interests overlap but are not identical. I have learnt to state whose position is improving, who is paying for the change and what evidence would cause us to revise the decision.
I have also changed how I respond when an operating measure improves. After we accelerated first payments, I asked whether the change affected only straightforward claims. Our analysts separated eligible cohorts and tracked reopened issues. That made the result less convenient to describe but more useful to manage. I would rather qualify a number than let colleagues organise their work around a misleading one. My reputation inside the business rests less on a single programme than on returning to those awkward qualifications after the initial presentation has finished.
I now want to contribute selectively beyond my executive responsibilities. A member-owned enterprise or a company whose service promise is tested under stress would be a relevant setting. I would bring experience of claims, pricing decisions and the distribution of cost between customer groups. I would not present myself as the board's actuarial expert or make an independence claim before the relevant relationships had been examined. The first useful conversation is about the contribution needed, the time available and the limits of my experience.
Chapter II
I ask what the person on the other side of an insurance decision can do next. An explanation may be technically correct and still leave a member unable to reopen a workshop or pay a repairer. My job is to preserve the discipline of the insurance promise while making its practical consequences impossible to overlook.
Ronan Kelveth
Chapter III
Chapter IV
§ 1
Our commercial property claims report once used final settlement time as its principal speed measure. That encouraged a complete view of each file, but it obscured cases where a substantial amount was already agreed while one smaller issue remained unresolved. A workshop could wait for cash to replace damaged equipment because a separate interruption calculation was still being examined. The file was progressing internally; the member could not act on that progress.
In 2023 I asked the claims director to establish a separate measure for the first undisputed payment. We limited the initial programme to defined commercial property cohorts and retained the final-settlement measure. Finance colleagues were concerned about reconciliation and recoverability of mistaken payments. Some experienced handlers worried that members would interpret a partial payment as agreement on the whole claim. Those objections shaped the process rather than being dismissed as resistance.
We created a short decision record identifying the accepted amount, unresolved elements and explanation given to the member. Experienced handlers received delegated authority within specified limits. Quality reviewers sampled paid and unpaid files, including cases in which a handler had decided against an interim payment. That last group mattered: reviewing only completed payments would have missed avoidable delay. We also tracked reopened issues and payment corrections so the new speed measure could not conceal a deterioration in accuracy.
By June 2025, median time to the first undisputed payment in eligible claims had fallen from 23 to 9 days. We did not describe that result as the settlement time for every commercial claim. The programme required A$1.1 million of additional training and review expenditure in its first year. It gave members earlier access to established amounts without pretending that unresolved questions had disappeared. I would now involve broker service teams sooner. Their early explanations sometimes differed from the wording claims colleagues had carefully tested.
§ 2
During the 2022 renewal cycle, Brindleharbour Mutual received 420 formal complaints about pricing in a reviewed commercial portfolio. Members were often told that market conditions explained the increase. That phrase combined several different causes and gave frontline staff little help when a member asked why a neighbouring business had received a different offer. The absence of a usable explanation also encouraged exceptions whose combined cost was difficult to see.
I commissioned a joint pricing and distribution review. We separated five components: expected claims, repair-cost assumptions, catastrophe protection, portfolio experience and individual exposure changes. We did not promise that every member could reproduce the actuarial calculation. We required a defensible account of the main movement and a clear route for correcting factual errors. The commercial team argued that more detailed explanations could create longer calls and expose inconsistencies in our own data. They were right about both risks.
We trained a first group of advisers with actual anonymised renewal cases and corrected the explanation where it broke down. A proposed concession had to identify its funding effect and the reason it was appropriate for that member. We also accepted that some business should leave if its desired price could not be supported. The board reviewed retention alongside the quality of the remaining portfolio, rather than treating a lower retention number as an automatic management failure.
Across two annual cycles, formal complaints in the reviewed portfolio fell from 420 to 255. Retention moved from 91% to 88%, and we recorded the lost premium rather than presenting the programme as an uncomplicated success. The important change was a more honest renewal conversation and a visible record of exceptions. I would have tested the first explanatory letters with a broader range of small businesses; our initial language assumed more familiarity with insurance cost components than several members had.
§ 3
Before the 2024 catastrophe-protection renewal, I asked how confidently we could assemble evidence for a large recovery. The purchasing paper compared capacity and price in detail. It said less about the location records and claims documentation that would support a submission after a widespread event. A sample of relevant historical files found missing location-level evidence in 18% of cases. The omission did not mean that every recovery would fail, but it made our stated readiness difficult to defend.
The finance director preferred to complete the data correction before changing the protection purchase. The reinsurance team was concerned that waiting would leave too little time to negotiate terms. I asked both teams to distinguish what could be corrected before renewal from what would remain an operating dependency. We took the board a combined decision: additional work on records and a protection structure better aligned with the exposures we intended to retain, at A$4.2 million more annual cost.
Claims specialists joined the renewal discussions and tested the proposed evidence requirements against sample files. We assigned owners to missing fields, created an exception register and required a documented recovery rehearsal using a synthetic event. The exercise exposed two hand-offs that were assumed rather than assigned. We corrected those responsibilities before describing the process as ready. The additional premium reduced funds available for other discretionary projects, including a planned distribution enhancement.
The subsequent file sample showed missing evidence in 4% of reviewed cases. I treat that as a control improvement, not evidence that a future recovery will be simple. Contract interpretation and event facts can still produce difficult discussions. The enterprise gained a shared understanding of what it had bought and what it needed to maintain. My principal lesson was to involve claims colleagues at the start of the buying process, before commercial options had already narrowed around a preferred proposal.
§ 4
In 2021, our supplier register suggested that we had several alternatives for urgent commercial property repairs. A closer review found that some firms depended on the same subcontractors and specialist equipment. Different contract names did not necessarily represent independent capacity. The problem became visible during a regional loss simulation in which several member premises needed similar work at the same time.
Procurement proposed adding more panel firms. Claims leaders preferred retaining a smaller group whose quality they knew. I asked for a dependency map and a trial that tested both positions. We identified the work for which genuine alternatives existed and the work constrained by scarce specialist capability. We then selected two additional regional firms for a limited pilot, retained inspections and paid for readiness arrangements where a credible capacity commitment could be established.
The financial trade-off was explicit. The pilot increased annual supplier-management and standby expenditure by A$780,000. We also accepted a smaller procurement saving than the original tender objective. In return, we required more transparent allocation rules and evidence of actual availability during a surge. Member communications explained when an interim measure or payment might be more practical than waiting for our preferred repair route. We avoided presenting cash as a universal substitute for a managed repair.
Within the pilot portfolio, median time from authorisation to repair commencement fell from 19 to 12 days over twelve months. Repeat visits for defective work remained at 3.2%, which meant the improvement had not yet solved every quality concern. We retained that qualification in the board report. The programme changed how we assessed supplier resilience: subcontractor dependencies and observed delivery mattered more than panel size. I would have included smaller rural jobs earlier, because travel and minimum-job economics created constraints that the metropolitan sample understated.
§ 5
As eastern regional director at Kestervane Underwriting, I inherited a niche property segment that contributed A$42 million of annual gross premium. The reported growth was attractive, but renewal files contained an increasing number of pricing and wording exceptions. Each exception had a plausible commercial explanation. Together they described a product different from the one our underwriting authority and operating capability were designed to support.
In 2012 I asked for a cohort review covering claims experience, referral volumes and the cost of servicing exceptions. The distribution head argued that we should invest in a revised product because brokers valued our responsiveness. The underwriting team believed that a narrower appetite would be safer. I proposed a staged withdrawal from the least coherent part of the segment while retaining clearly defined risks that met the intended terms. This sacrificed volume and required uncomfortable broker conversations.
We stopped new business in the affected cohort, honoured existing commitments and gave brokers advance notice of renewal decisions. A small transition team checked that communications matched the actual policy position. We did not use the withdrawal as a reason to rush open claims or impose retrospective conditions. I moved two experienced colleagues into broker support for the transition and accepted that their usual growth objectives would need revision.
The affected premium reduced by A$17 million over eighteen months. Referral volumes in the remaining segment fell by 36%, and the team could explain the retained appetite without relying on informal workarounds. I did not attribute subsequent claims movements entirely to the withdrawal; weather and claim development also affected the result. The decision taught me that commercial relevance requires a product we can administer consistently. If repeating it, I would establish the cohort review earlier, before the accumulation of exceptions had created expectations that were costly for both brokers and us to unwind.
Chapter V
Executive director and chief executive
Reports to risk and member interests committees; executive status
July 2020–present
Chapter VI
The time to pay an established amount and the time to conclude every question are different management obligations. Measuring both can make a complex claim more useful to the policyholder without disguising uncertainty.
Read the essayA member-owned insurer should explain who funds a pricing exception. Fairness requires an account of differences between members, including those whose losses will arrive after today's renewal discussion.
Read the essayCapacity and price do not describe the whole protection decision. The executive team must connect contract terms with location data, loss evidence and the people who will assemble a recovery after an event.
Read the essayAn invitation
I consider one external non-executive board role in a member-owned or service-intensive organisation, subject to clearance and conflict assessment. I welcome speaking invitations about claims, mutual economics and executive accountability. I am staying in my current executive role and do not accept individual claims appeals through this website.
office@ronankelveth.example