01
Eight vectors, every time
Financial fidelity, governance, regulatory, market, litigation, ESG, emerging exposure and sector risk are assessed on every submission. No vector is dropped for time.
D&O AnalyzerIQLive
Four hours of D&O research.
Now five minutes
Eight underwriting risk vectors assessed from authoritative primary sources, every finding cited and explained, and a complete audit trail behind the decision. Built to stand up in a Lloyd’s performance review.
5 minper company assessment
8underwriting risk vectors
100%of findings source-cited
The reality
What this looks like
in most firms today
- Hours per submission spent across Companies House, court records, filings and regulatory registers.
- No standard framework, so two submissions assessed by two underwriters cannot be compared.
- Under time pressure, whole risk vectors are skipped — and nobody records which ones.
Rising litigation costs, equity volatility and a shifting regulatory frontier have made D&O one of the hardest classes to price. The research constraint is not expertise. It is hours in the day.
Where the existing tools stop
Where tools stop
Research aggregators return documents. Pricing needs assessed signals, weighted and comparable across a book.
Emerging exposures — AI liability, ESG fiduciary risk, ECCTA — are not in any legacy assessment framework at all.
What it takes
What underwriting needs is the same eight vectors assessed identically on every submission, with the evidence attached.
How it delivers value
How D&O AnalyzerIQ earns its place
in the workflow
02
Primary sources, cited
Assessment runs against authoritative primary sources, and every signal carries the source it came from. The underwriter can check any finding in one click.
03
Standardised and comparable
The same assessment framework produces the same output structure every time, which is what makes submissions comparable across the book.
04
The underwriter owns the decision
The system assesses and evidences; it does not bind. Every output is reviewed, and the review is recorded with the assessment.
Capabilities
The specification, without decoration
- Eight-vector risk assessment from company name to evidenced underwriting report
- Financial fidelity — liquidity, solvency and financial health assessment
- Governance audit — board composition, oversight and compliance review
- Emerging exposure coverage for AI liability, ESG fiduciary risk and ECCTA as standard
- Full XAI audit trail: every source, signal weighting and human decision timestamped
The approach
How it actually works
Written for the technical evaluator rather than the buyer. If you are going to put this in front of an architecture review board, this is the section they will read.
Source acquisition before assessment
Primary sources are gathered and attributed first. Every downstream signal is bound to the document it came from, which is what makes the audit trail real rather than reconstructed.
One vector, one assessment pass
Each of the eight vectors is assessed independently against its own criteria rather than in a single sweeping prompt. Independent passes are why the output is comparable between submissions.
Signal weighting is explicit and logged
The weighting applied to each signal is recorded with the assessment, so a reviewer can see not just what was found but how heavily it counted.
Human decision recorded with the analysis
The underwriter’s decision, and any departure from the assessment, is captured against the report. That record is what a performance review actually asks to see.
Model-flexible by design
The assessment framework is not bound to one model provider. Model choice is a configuration, so the system survives the next generation of models without a rebuild.
The breakthrough
What changes once it is running
The time figure is a delivered result. The book-level figures are ranges, because they depend on submission volume and mix.
5 minper company assessmentDown from roughly four hours of analyst research
8vectors on every submissionNo vector skipped under time pressure
100%of findings source-citedEvery signal bound to a primary source document
2–3×more submissions assessedModelled at constant underwriter headcount
