How it works
Eight steps, and the honest answer is that step three is most of the work.
Everything else takes about twenty minutes. The loss data takes as long as it takes to get a loss run from your broker or carrier, and there is no shortcut around it. A feasibility study built on estimated losses is a guess with a chart attached.
The platform saves after every step, so you can open a study now, get to step three, and come back when the loss run arrives.
Company profile
Legal entity, home state, and the number of insured entities. The home state determines self-procurement tax exposure if you were to insure through an offshore captive; the entity count matters for risk distribution and, if you are contemplating an 831(b) election, for the diversification test.
Exposure values
Payroll, revenue, vehicle count, property values. Exposure is the denominator that separates a rising loss total from a growing business — without it, five years of increasing losses look alarming even when the loss rate has been flat.
Historical loss data
Five or six years of loss history, valued at each development age, split by line of business. This is the part that decides whether the analysis is defensible. A carrier loss run contains everything needed. Case-incurred figures and claim counts matter: incurred develops faster than paid, and without counts frequency cannot be separated from severity.
Current programme
What the commercial market charges today, and the deductible or SIR you already carry. This is the baseline every projected saving is measured against.
Risk tolerance
How much retained loss the business could absorb in a bad year without distress, and the confidence level you want the premium funded to. This is a judgement about your balance sheet, not an actuarial output, which is why we ask rather than assume.
Balance sheet
Capital available to commit, and enough of the balance sheet to test capital adequacy. A captive capitalised at exactly the domicile minimum has no cushion, and the severe-adverse scenario will show you what that means.
Structure
Pure, group, cell, or micro-captive. This filters the domicile comparison — not every domicile permits every structure — and determines whether the 831(b) diagnostics run.
Run the analysis
Chain-ladder and Bornhuetter-Ferguson projections, a seeded Monte Carlo simulation at 10,000 iterations, domicile scoring across ten domiciles, tax and regulatory diagnostics, five-year pro formas under three scenarios, and a verdict with every threshold published. Every input and assumption is written to the audit vault as it happens.