The analysis, powered by ZestyAI’s proprietary Z-HAIL™ model, highlights the rising cost of hailstorms and underscores the urgent need for more precise risk assessment in the insurance sector.
The financial impact of severe convective storms (SCS)—which include hail, tornadoes, and high-wind events—has been growing rapidly, with damages in 2024 alone estimated at $56 billion, surpassing losses from hurricanes. Despite this increasing threat, many insurers still rely on traditional catastrophe models designed to estimate exposure at the portfolio level rather than at individual property level. These older models often miss the specific structural and environmental factors that lead to real-world losses from hail events.
“Catastrophe models have helped insurers understand where storms may strike and how losses might add up at a portfolio level,” said Kumar Dhuvur, Co-Founder and Chief Product Officer at ZestyAI. “But they weren’t built to assess risk at the individual property level, and they often miss the specific conditions that drive hail damage. By analyzing the interaction between structure-specific features and local storm patterns, we can distinguish risk between neighboring properties—enabling smarter underwriting, more precise pricing, and better protection for policyholders.”
ZestyAI’s Z-HAIL model blends climate data, aerial imagery, and property-specific characteristics using advanced machine learning techniques to deliver granular predictions. These reflect not only the physical condition of each structure but also the storm activity in its immediate area. This hyper-local precision equips insurers with the insights needed to refine underwriting and pricing strategies in even the most hail-prone regions of the country.
According to the analysis, 12.6 million structures in the U.S. are considered high risk for hail-related roof damage. These properties represent a total potential exposure of $189.5 billion in roof replacement costs. The top five states by dollar exposure are Texas ($68B), Colorado ($16.7B), Illinois ($10.8B), North Carolina ($10.4B), and Missouri ($9.5B). In contrast, the states with the lowest exposure include Maine ($4.7M), Idaho ($12.8M), New Hampshire ($18.5M), Nevada ($49.3M), and Vermont ($64.7M).
ZestyAI’s ability to assess risk at a micro level has been proven through real-world case studies. In one example from Allen, Texas—following a storm with 2.5-inch hailstones—Z-HAIL analyzed 483 insured properties. The model identified no losses among those rated “Very Low” risk, despite being in the same neighborhood and exposed to the same weather conditions as higher-risk homes. This demonstrates Z-HAIL’s ability to distinguish between properties in close proximity and highlights its value for insurers seeking to improve risk selection accuracy.
Z-HAIL is currently approved for use in 14 U.S. states, including all five with the highest hail-related exposure. Additional regulatory approvals are pending as insurers seek to adapt to the growing threat of severe weather with more precise, data-driven tools. As hail events increase in both frequency and severity, ZestyAI’s findings call attention to the urgent need for more accurate and granular risk models that move beyond legacy systems.
