Inside a lawyers professional liability submission: what InsightXtract extracts — and why it matters

Lawyers professional liability (LPL) is a line where the risk lives in the nuances. Two firms with identical headcounts and identical requested limits can be worlds apart in exposure — one does routine residential real estate closings, the other does securities and M&A work with eight-figure deals behind every engagement. The submission that tells you which is which is short: a broker’s email with the ask, and an LPL application. But inside that application sit the firm profile, the exact shape of the claims-made cover, a roster of every attorney, and the firm’s prior claims — and those are what decide the price.

Most intake tools read one document at a time and hand back a flat list of fields. That is not how an LPL account is underwritten. InsightXtract runs an agentic, multi-document pipeline that classifies each file, extracts to a per-document schema, consolidates everything into one unified account record, links it into an entity graph, and validates it against your rules — so the underwriter opens a single, coded, cited record instead of a stack of PDFs.

One submission, two documents, one connected record — classify, extract, consolidate, link, validate.

The idea that governs everything: the practice mix is the risk

Before the categories, the single most important concept. In lawyers professional liability, headcount and limits tell you the size of the account — the practice mix tells you the risk. A ten-attorney firm doing wills, trusts, and residential closings is a fundamentally different exposure from a ten-attorney firm doing plaintiff’s personal injury, securities, or intellectual property — and the LPL rate should reflect that gap far more than it reflects raw size. The same is true of coverage: because LPL is written claims-made, the retroactive date and prior-acts terms silently define how much of the firm’s past work is actually on the hook.

So InsightXtract captures not just the headline fields but the detail that carries the exposure — the areas of practice, the per-attorney seniority, and the full claims history. In the tables below, the Captured column flags how each parameter is pulled:

• Point-in-time — a single static fact about the firm or the ask.
• Current — a term-level value for this submission.
• Per-row — a table: one attorney, or one claim, per row.

A · Firm identity

Who the insured firm is — legal entity, tax identity, location, and operations classification. This is the account key everything else hangs off. Source: LPL application.

Parameter	Captured	Why it matters
Insured name · DBA · entity type	Point-in-time	The contract party and unique account key — drives clearance, conflicts, and de-dupe.
FEIN	Point-in-time	Tax identity for matching, aggregation, and downstream systems.
Address · risk state	Point-in-time	Venue and jurisdiction — state bar rules and malpractice climate vary widely.
NAICS code · SIC code · business description	Point-in-time	Class-based rating and appetite/knockout — the operations behind the code.

B · Broker & submission

Who is placing the account and the exact shape of the ask — the distribution chain plus the product, term, and requested limit. Source: broker submission email.

Parameter	Captured	Why it matters
Broker name · broker email	Point-in-time	Distribution routing, binding authority, and correspondence.
Product	Current	Coverage form and program the ask maps to.
Effective date	Current	Term start and the binding deadline the desk works against.
Requested limit	Current	The capacity the broker is asking for — the headline of the submission.

C · Firm profile

The size, shape, and specialization of the practice — and this is where the risk really lives. The areas of practice matter more than any single count. Source: LPL application.

Parameter	Captured	Why it matters
Number of attorneys · professional count	Point-in-time	The primary exposure unit — LPL is rated per attorney.
Number of support staff · employee count	Point-in-time	Leverage and supervision load — a signal of practice structure.
Annual billings · billings · estimated revenue	Point-in-time	Firm scale and the size of the matters behind each engagement.
Total payroll	Point-in-time	Secondary exposure base and a cross-check on firm size.
Areas of practice · professional services	Point-in-time	The single biggest driver of LPL risk — securities and plaintiff work carry far more severity than trusts or closings.
Years established · years in business	Point-in-time	Stability credit and the depth of the firm’s track record.

D · Coverage & limits

The exact terms of the professional liability cover being requested. Because LPL is claims-made, the trigger, retroactive date, and prior-acts terms define how much past work is actually covered. Source: LPL application.

Parameter	Captured	Why it matters
Each-claim limit	Current	The most the policy pays on any single malpractice claim — the core of the ask.
Aggregate limit · limit aggregate	Current	Total annual capacity across all claims — caps the firm’s worst year.
Deductible	Current	The firm’s retained loss — skin in the game and net premium math.
Claims-made	Current	The coverage trigger — it fixes what makes a claim payable and when.
Retroactive date	Current	How far back covered work reaches — the true measure of tail exposure taken on.
Prior acts coverage	Current	Whether work before the retro date is picked up — a major swing in exposure.
Prior carrier	Current	Continuity and the source of any needed tail or nose coverage.
Premium	Current	Target / expiring price and the rate-change baseline for a renewal.

E · The attorney schedule — one row per attorney

An LPL rate is built attorney by attorney, not firm by firm. InsightXtract reads the roster as a table — one row per attorney — so the desk can see seniority, licensure, and specialization across the whole firm. Source: LPL application.

Parameter	Captured	Why it matters
Attorney name	Per-row	The named professional — the unit that is actually rated and vetted.
Bar state	Per-row	Where each attorney is licensed — venue spread and multi-jurisdiction exposure.
Years admitted	Per-row	Experience — a direct signal of malpractice frequency; new admittees carry more risk.
Area of practice	Per-row	Per-attorney specialization — reveals hazardous-work concentration the firm-level summary hides.

F · Claims history — one row per prior claim

The single biggest pricing input. InsightXtract reads each prior malpractice claim as a row and captures both its narrative and its dollars — paid, reserved, and total incurred. Source: LPL application.

Parameter	Captured	Why it matters
Claim number	Per-row	The unique identifier tying each loss back to its file.
Date of loss	Per-row	When the claim arose — frequency timing and its relation to the retro date.
Description	Per-row	The nature of the alleged error — the qualitative read on how the firm gets into trouble.
Status (open / closed)	Per-row	Whether exposure is still developing or resolved.
Paid · reserve · incurred	Per-row	Cost to date, open exposure on the books, and total severity — the loss-cost picture that drives the rate.

From two documents to one connected record

Pulling these parameters out of two documents is only half the job. The value is in consolidation: the insured named on the application, the broker on the email, the coverage terms, the attorney roster, and the claims all describe one account. InsightXtract merges them into a single record by a declared source-of-truth priority (the application wins over the email on firm details; the email carries the ask), applies reference-data lookups — NAICS and state codes standardized and validated — and links the result into an entity graph: the firm at the centre, connected to broker, submission, coverage, its attorneys, and its claims history.

Why it matters to the business

Comprehensive, structured, detail-aware extraction isn’t a data-entry nicety — it changes the economics and quality of the book:

• Areas of practice = risk. Capturing the practice mix at both firm and per-attorney level — not just headcount — lets you rate to the hazard of the work, which is where LPL money is won or lost.
• Claims-made done right. The retroactive date, prior-acts terms, and prior carrier are captured by default, so the desk sees exactly how much past work the policy takes on — and where tail or nose coverage is needed — instead of discovering it after a claim.
• Attorney experience, seen clearly. Years admitted and bar state per attorney turn a roster into a frequency signal — a firm loaded with new admittees in a high-severity practice is a very different risk from a seasoned bench.
• Prior acts and loss history, front and centre. The firm’s own claims — incurred, reserved, open or closed — are the strongest predictor of the next one, and they reach the underwriter instead of staying buried in an application appendix.
• Faster quotes, consistency, and auditability. An underwriter opening a consolidated, cited record triages and prices in minutes — the same categories, coded the same way every time, with provenance to the source document.

The Lawyers Professional Liability agent extracts all of this today — the firm identity and profile; the broker and submission; the full claims-made coverage terms; the per-attorney schedule; and the firm’s claims history — consolidated into one coded record of 35+ fields and two linked tables, every value cited to its source document and standardized against NAICS and state reference data. The point this post makes is why it matters: the practice mix, the attorney seniority, and the claims-made nuances are exactly what move an LPL price, so they’re captured by default. And because it’s all configuration — fields and tables in the agent’s output contract, not code — the schema keeps pace with what underwriters ask for.