This tool translates physical risk reduction (e.g., a commercial roof upgrade) into insurance-relevant outcomes under a specified policy structure.
It takes an event-based loss distribution, applies mitigation assumptions, and maps resulting losses through a policy layer to show changes in EAL, insured vs. uninsured loss, and where avoided losses sit relative to the deductible and coverage.
This demo focuses on SME commercial properties exposed to hail risk. All assumptions are user-editable and illustrative only.
This prototype has been shaped with input from practitioners at InnSure and others working in insurance and resilience.
Loss data
We need event-level loss distributions from catastrophe models (AIR, RMS, CoreLogic) or historical claims data. The current sample data is illustrative only. A data partner with access to calibrated hail loss curves for commercial property would immediately strengthen credibility.
Mitigation performance data
Current mitigation assumptions are expert-judgment placeholders. We need product-specific performance data from IBHS, FM Global, or manufacturer testing — ideally hazard-intensity-conditioned curves, not just a single reduction percentage.
Empirical validation
Before-and-after claims data for buildings with known roof upgrades would allow backtesting. Even a small dataset (20–50 properties) with pre- and post-upgrade claim histories would be valuable.
Expanded building archetypes
More occupancy types, roof systems, regions, and value ranges. Partners with commercial property underwriting data can help define realistic profiles.
Policy structure realism
The current policy engine is simplified. Input from underwriters on sublimits, aggregate deductibles, coinsurance penalties, and other common commercial property terms would improve accuracy.
Multi-peril extension
Expanding beyond hail to wind, flood, and wildfire requires new loss distributions and mitigation curves per hazard. Partners with multi-peril modeling capability can accelerate this.
Premium indication
A loss-cost-based premium estimation module could show how mitigation affects pricing. Requires actuarial input on expense loads, profit margins, and regulatory constraints.
Pilot deployments
We are looking for brokers, carriers, or resilience programs willing to pilot the tool on real accounts and provide feedback on usability and output credibility.
1. Where do the input data come from?
The tool operates on an event-level loss distribution provided by the user. Defensibility depends on the source: catastrophe models, engineering studies, insurer loss data, or validated public datasets. Sample data are illustrative only.
2. What does the tool actually do?
It applies mitigation assumptions to event losses, recomputes EAL and EP curves, maps losses through a policy structure (deductible, limit), and reports changes in gross, insured, and uninsured loss. It does not model hazard or damage directly.
3. How is mitigation represented?
Uniform scalar: base loss reduction, probability of failure, and maintenance degradation combine into an effective reduction applied uniformly. Hazard-intensity curve: reduction varies by event intensity (e.g., hail size), with failure and maintenance applied on top.
4. Does the tool include vulnerability or damage functions?
No. Losses are provided directly in the CSV. The tool does not convert hazard intensity into damage. All hazard-to-loss assumptions sit upstream.
5. How does the insurance layer work?
Simplified policy structure: deductible (flat or % of coverage), coverage limit, optional insured share mode. Calculates insured loss and uninsured (retained) loss.
6. Why might mitigation reduce gross loss but not insured loss?
If avoided losses occur below the deductible or above the coverage limit, insured loss may change less than gross loss. Identifying where mitigation sits relative to the policy layer is a central use case.
7. What is the archetype system?
Archetypes are preset building profiles (e.g., small warehouse with flat roof) that pre-fill asset inputs. They provide consistent starting points for common SME property types.
8. What are the two mitigation modes?
Uniform scalar applies a single reduction percentage across all events. Hazard-intensity curve maps hail size to reduction percentage, allowing more realistic intensity-dependent mitigation.
9. What is hazard share?
The portion of total risk attributable to the modeled hazard. If provided, hazard-specific reduction is translated into a total-risk estimate. Otherwise results are hazard-specific only.
10. What are the main limitations?
No hazard/damage modeling, simplified mitigation and policy, single-asset scope, no empirical calibration. Best understood as a translation and sensitivity tool, not a predictive model.