An assessor checking a builder’s repair quote works through it line by line, labour against materials against local rates, the same way it has been done for decades. It is careful work. But as claims volumes and complexity increase, manual review alone can leave anomalies undetected, a challenge that PropertyExpert is addressing with AI-powered property claims management and that has earned the company a place among Insurance Business’ 5-Star Insurance Innovators 2026.
The company was founded in Germany in 2012 and has developed an approach that combines AI, data and human expertise across property claims management. It reached a significant scale milestone in 2024, processing more than one million claims in a single year across 80 insurers, before expanding internationally in 2025 into the United States, Australia and Thailand. PropertyExpert now employs more than 200 specialists across claims management, technology and data worldwide, per the company’s own figures.
Milad Razavi, country manager ANZ and managing director of PropertyExpert Australia, is direct about what the platform was designed to achieve when it launched in this market. “When we launched PropertyExpert in Australia and New Zealand, we were not trying to replace the expertise of claims assessors,” he says. “We were trying to give them something they have never had before. A systematic, consistent way to validate every line item in every quote, regardless of volume or complexity. The results have been immediate. More than half of the quotes we process contain anomalies that would never have been caught through manual review. That is not simply a technology story. That is a leakage story, and it has real dollar consequences for every insurer carrying property risk in this market.”
The problem PropertyExpert set out to solve in the Australian market is not new. Assessors still receive unstructured documents from builders and repairers and review them by hand, a process that has not fundamentally changed in decades, a pattern Deloitte’s analysis of AI and human judgement in P&C claims identifies as common across the wider industry, not unique to any one market.
Razavi says the issue was never a lack of skill among assessors. “The problem is that the volume and complexity of what they are being asked to review has grown well beyond what any manual process can handle consistently,” he says. “AI changes that equation. Not by taking over, but by doing the heavy lifting so that human judgement can be applied where it actually matters.”
Mechanically, that means digitising incoming scope documents, cost estimates and invoices regardless of format, then evaluating them against real claims data and market benchmarks using a combination of fixed rules and AI-powered analysis, according to PropertyExpert’s product documentation.
Anomalies and pricing deviations are automatically flagged, with straightforward cases cleared for fast approval and irregular ones routed for review with the supporting justification already prepared. This human-in-the-loop approach mirrors what CSIRO’s national research on AI adoption in Australian insurance identifies as a priority use case for the sector: automated claims processing paired with responsible governance.
But scope and quote validation is only one part of PropertyExpert’s broader approach. Its platform supports the property claims lifecycle from First Notice of Loss (FNOL) through claims assessment and validation to fraud detection, recovery and payment, combining AI-powered technology with human expertise across the process.
The underlying principle is that information captured at one stage of a claim can remain useful at the next. Rather than repeatedly collecting, interpreting and processing information in isolation, data can support decisions through the claims lifecycle, from the first information received through to the final outcome.
That broader end-to-end approach is central to how PropertyExpert positions its technology: individual applications solve specific problems, while the information generated throughout the claim creates additional value when it can be used across the process.
For Razavi, the effectiveness of quote validation in Australia and New Zealand also depends on localisation. “Our pricing intelligence is calibrated to Australian and New Zealand repair costs,” he says. “Our platform understands local trade categories, repair methodology and regulatory requirements. That localisation is what makes the outcomes meaningful. A 21% average saving per validated invoice is not a figure you can achieve with a generic tool. It comes from deep market knowledge combined with AI that knows what to look for.”
The human-AI division of labour extends to how the platform performs at scale. PropertyExpert processed more than one million property claims globally in 2025 across four countries. The company reports end-to-end claims processing within hours and a return on investment exceeding 12:1.
For Razavi, the wider implication is that AI-powered claims technology is already moving beyond experimentation. It can deliver measurable outcomes today while changing how insurers think about the claims process itself.
This direction is consistent with Insurance Business’ broader coverage of the technology reshaping APAC insurance in 2026. Razavi advocates a model in which technology provides consistency and scale, while human expertise remains focused on complexity, context and judgement.
Quote validation is therefore not an endpoint but one example of a broader shift in property claims management. As AI becomes increasingly capable of connecting information across the claims lifecycle, the opportunity extends beyond processing individual tasks faster. It gives insurers the ability to identify issues earlier, make better-informed decisions and use information generated during claims more effectively across the process.
For PropertyExpert, that development points towards an even broader challenge: how technology, data and human expertise can help insurers manage increasingly complex property risks while keeping them insurable in the future.
This article is sponsored by PropertyExpert Australia
