Insurate

Insurate is a focused test of whether external operational data can identify materially different workers’ compensation risk inside class codes that traditional markets often treat as uniformly hazardous. The MGA targets construction, manufacturing, transportation, energy, agribusiness and related employers with guaranteed-cost premiums from $25,000 to $3 million. Its Safe-Tier approach combines claims history with hundreds of additional data points and predictive models, then uses the results for pricing and loss-prevention recommendations. The thesis is economically important: if safer operators are pooled with weaker peers, they subsidize loss and have less incentive to invest in risk control. Better differentiation could lower premiums for high-performing employers while improving carrier results. The central question is whether the data predict future claims rather than merely restate past success. Workers’ compensation has long development patterns, medical inflation, legal variation and exposure changes that can make apparent early advantage disappear. Novel variables can also be proxies for company size, geography, workforce composition or access to technology. Carrier and regulator confidence therefore depends on documented data provenance, stable definitions, explainable decisions, testing for unintended bias and out-of-time validation across multiple underwriting years. Insurate says its approach requires no employee monitoring or additional surveys and instead uses existing loss history and other operational data. That reduces friction and some privacy concerns, but it increases the importance of verifying third-party data, correcting errors and understanding when inputs have gone stale. Model scores should supplement—not obscure—underwriter judgment, payroll and class-code accuracy, management interviews and actual loss-control conditions. Markel announced a strategic collaboration with Insurate in June 2025 to enter middle-market workers’ compensation, following an earlier capacity and product-development relationship with SiriusPoint announced in 2022. A new carrier relationship can validate the proposition and broaden scale, but it resets performance cohorts. Historical results under a prior insurer, policy form, claims administrator or reinsurance structure are not automatically portable. The relevant evidence must separate carrier periods and show how claims, reserving, audits and safety interventions are governed. Insurate’s fractional chief safety officer service and AI-generated recommendations are potentially more valuable than selection alone because they aim to change risk after binding. Their impact should be measured against comparable insureds, with adoption and outcome tracking rather than anecdotal savings. Public disclosures do not provide premium, loss ratios, retention or state-level availability. The indicators worth watching are loss and reserve development by policy year and class, actual-to-expected claim frequency and severity by Safe-Tier band, score stability, model overrides, audit adjustments, claims closure, intervention adoption, carrier tenure, producer concentration, renewal retention and whether price advantages persist after several high-severity years.