Renewal Risk Hides in 60-Day Decision Windows
Selective retention margins hold when multi-year loss data drives underwriting
Renewal decisions happen fast—typically 60 days before expiration. Underwriters review current loss runs, hazard changes, and market appetite. But renewal risk is not one year—it's five. Loss trends that emerge over time (increasing frequency, severity creep, claim pattern shifts) get compressed into a 60-minute underwriting review. Selective retention requires recognizing loss patterns before renewal notice arrives.
Without multi-year loss data, you retain risks that should be cut.
Where capacity bleeds today
The bottlenecks AI removes
Loss Trends Scatter Across Files and Systems
Loss runs arrive in email throughout the year. Summary data gets entered in the management system; detail stays in PDFs. Underwriters pull individual files when renewal notice lands, often missing loss patterns visible only in 3–5 year trends. Seasonal spikes, frequency creep, and claims development patterns stay invisible until the renewal decision is already made. One-year loss data is insufficient for accurate retention decisions.
Selective Retention Requires Real-Time Portfolio Context
Marginal risks are acceptable when portfolio loss ratios are healthy and market appetite is strong. The same risks become unacceptable when portfolio deteriorates. Current market rates also factor into retention decisions—rate increases change underwriting appetite. Underwriters need instant access to portfolio trends, market movement, and placement economics to make selective retention calls. Spreadsheet analysis is too slow.
AI Renewal Underwriting MGA Surfaces Multi-Year Patterns
Bastion's Soldier ingests 3–5 years of loss runs per risk, automatically detecting frequency trends, severity creep, and claims development patterns. Market condition data (rate movement, capacity appetite) flows continuously. When renewal notice lands, underwriters see not just current loss data but trend analysis, portfolio impact, and recommended action. Selective retention becomes data-driven decision-making, not guesswork.
AI renewal decisions convert multi-year data into binding choices that stick.
| Dimension | Before AI | After AI |
|---|---|---|
| Loss Data Available at Renewal | Current year loss run only; historical files archived | 3–5 year loss history; trend analysis pre-calculated |
| Loss Trend Detection | Manual review; seasonal spikes and creep missed | Automated trend flagging; frequency/severity patterns visible |
| Renewal Decision Speed | 45 minutes per file; limited context available | 5-minute review; full trend and portfolio context loaded |
| Selective Retention Accuracy | 40–50% of risks retain or cut based on intuition | 80%+ decisions guided by quantified loss trends |
| Portfolio Loss Ratio on Renewals | Selective retention misses deteriorating subpools | Portfolio subpool visibility; proactive cut signals |
Retention decision quality improves 30-40% with historical data. Margin floor moves from 40% ceiling to 32% guaranteed floor.
Where this sits in the $84B pool
$30.8B of MGA revenue is AI-compressible. Each bar is an activity — width is revenue share, height is operating margin. This workflow sits where the bar lands. Click any other to explore it.
Co-operate, not consult
We take position in the workflows we automate.
MGA margin sits in intake velocity, underwriting triage, and claims throughput. We run these — not map them. Our economics are equity in the margin you recover, not retainer on the analysis.
Talk to a principalThe full $84B pool
See where the MGA margin moves.
Map every activity — width is revenue share, height is operating margin. Click any bar to explore that workflow.
View the profit poolHow much renewal decision cycle time is spent on data gathering vs. judgment?
Typically 60–70% of renewal review time is spent pulling files, compiling loss data, and formatting information for underwriting. Only 30–40% is actual judgment and risk assessment. AI handles data gathering in real-time, leaving underwriters to focus entirely on judgment. The net effect is 10–15 fewer hours per renewal cycle spent on administrative work.
What hidden loss trends does multi-year AI analysis reveal for renewal decisions?
Multi-year analysis reveals frequency acceleration (claims increase year-over-year), severity creep (average loss size trending up), claims development patterns (long-tail emergence), and seasonal concentration (peak loss months). These patterns are invisible in single-year loss runs but drive retention decisions. MGAs miss 2–3 deteriorating subpools per renewal cycle without multi-year visibility.
How does selective AI-driven renewal improve MGA loss ratios?
Selective retention cuts losing risks faster, before they deteriorate further. It also identifies high-margin renewal risks worth competing harder to retain. Portfolio-level feedback shows which underwriting cohorts (class, size, geography) are performing. Loss ratio improves 3–5% when MGAs cut losers based on trend data rather than renewal-notice-only snapshots.