Claims data is an underwriting input—not an actuarial afterthought

Orion180’s public filing describes ongoing claims and portfolio feedback being used to refine models and adjust risk appetite. The principle extends beyond technology-led property underwriting: claims contain the earliest operating evidence that an underwriting thesis may need to change.

Yet claims information often reaches underwriters as a lagging summary. By then, the organization may have repeated the same decision across another renewal cycle.

Build the learning loop

A useful loop distinguishes signal from anecdote while preserving urgency.

• Tag claims to underwriting attributes and original decisions.
• Review frequency, severity and near-miss patterns separately.
• Name the threshold that triggers portfolio review.
• Record the action, owner and expected effect.
• Measure whether the change improved subsequent experience.

Leadership ownership

Claims learning belongs in a recurring cross-functional forum. Underwriting, actuarial, claims, operations and product leaders should agree on the signal and the response—not exchange reports in sequence.

The fastest signal can also be the noisiest

Early claims observations may reflect random variation, reporting behavior or operational changes rather than a broken underwriting thesis. Acting too quickly can create unstable appetite and inconsistent treatment across renewal cohorts.

The learning loop needs both urgency and statistical discipline: a method for escalating potential signals, testing alternative explanations and recording why management acted—or chose not to act.

The counterpoint: waiting for credibility can be expensive

Traditional actuarial thresholds may arrive after an MGA has repeated the same decision across substantial premium. Qualitative claims evidence, near misses and adjuster observations can justify targeted review before a pattern is fully credible.

MGA Index expects leading operators to separate “investigate” thresholds from “change appetite” thresholds. Earlier curiosity need not mean premature portfolio action.

• Create different thresholds for inquiry, intervention and full appetite change.
• Retain the rationale when management declines to act.
• Measure the exposure written between first signal and final decision.

A feedback loop needs decision rights

Orion180 describes using claims and portfolio feedback to refine models and appetite. The principle extends beyond one property platform: claim facts become strategically useful when they can be connected to the original underwriting evidence and reach a person authorized to change selection, price, terms or risk control.

A claims dashboard without thresholds and ownership remains observation. The operating design should name which patterns trigger review, how confidence is assessed and how the resulting action is recorded.

The countercase: anecdotes can distort appetite

Vivid claims can dominate judgment before credible frequency or severity patterns develop. Underwriters need timely detail, but actuarial context and cohort discipline protect the portfolio from reacting to noise.

The answer is a tiered loop: immediate escalation for material control or wording failures, periodic analysis for developing patterns and documented review before broad appetite changes.