Cytora, Warren Group Automate Underwriting
Cytora, an insurance technology firm, has partnered with real estate data provider The Warren Group. The collaboration will embed national property and transaction data directly into commercial insurance workflows, aiming to automate risk evaluation and accelerate decision-making for underwriters. The move is part of a broader trend of insurers adopting AI and data integrations to improve commercial property underwriting.
- The Warren Group, a family-owned company now in its fourth generation, has been collecting real estate and transaction data since 1872. Its data is sourced from public records at registries of deeds and town clerks' offices. - London-based Cytora, founded in 2012, has raised a total of $41.5 million over five funding rounds from investors including QBE Ventures and Cambridge Enterprise. - This partnership addresses a core inefficiency in the insurance industry, where underwriters can spend as much as 41% of their time on administrative tasks rather than on strategic risk analysis. - The collaboration is part of a significant market trend, with the global AI-powered commercial property insurance market valued at $167 million in 2025 and projected to grow to over $1.5 billion by 2033. - The integration will give underwriters access to The Warren Group's extensive datasets, which include building permits, property characteristics, sales history, and mortgage information. - Insurers are rapidly adopting such technologies, with one study indicating that the use of AI in underwriting is expected to increase from 14% to 70% in the next three years. - The use of third-party data directly addresses the underwriting challenge of "garbage in, garbage out," where inaccurate property valuations lead to incorrect risk pricing and potential underinsurance. - Cytora has recently established similar data-centric partnerships with firms specializing in geospatial intelligence and climate risk assessment, indicating a strategy of enriching its platform with diverse external data sources.