Inside a commercial property submission: what InsightXtract extracts — and why it matters

Property is deceptively simple to quote and genuinely hard to intake. The account looks like a single number — a total insured value — but that number is the sum of a schedule that can run to dozens or hundreds of buildings, each with its own construction, occupancy, protection, and value split. The information that actually decides the price — the concentration of TIV in a wind zone, the mix of frame versus fire-resistive construction, whether the roofs are aging, how the account has actually run for losses — is buried in a Statement of Values workbook, an application, and a broker email, and it is almost never on any one page.

Most intake tools read one document at a time and hand back a flat list of fields. That is not how a property 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 property record, and validates it against your rules and reference data — so the underwriter opens a single, coded, cited record instead of a spreadsheet and two PDFs.

One submission, three document types, one connected record — classify, extract, consolidate, validate.

The idea that governs everything: COPE lives in the schedule

Before the categories, the single most important concept. In commercial property, the account-level total is a summary — the risk lives in the schedule. Two accounts can carry an identical $40M TIV and be completely different exposures: one is a single fire-resistive distribution center in a low-hazard zone; the other is fifteen frame retail units, half of them along the Gulf coast, several with 25-year-old roofs. You cannot see that difference in the header fields. You can only see it by reading the Statement of Values (SOV) as a table — one row per location — and by capturing COPE (Construction, Occupancy, Protection, Exposure) at the building level.

So InsightXtract captures both layers: the account-level application fields and the per-row schedules. In the tables below, the Type column flags how each parameter is captured:

• Point-in-time — a single account-level fact from the application or email.
• Coverage term — a value tied to the requested policy structure.
• Per-row — a table: one location, one peril sublimit, or one claim per row.

A · Insured, account & identity

Who the account is — legal identity, address, industry classification, and the operating profile that anchors clearance and appetite. Sources: application form, broker email.

Parameter	Type	Why it matters
Named insured · DBA / trade name · FEIN	Point-in-time	The contract party and unique account key — drives clearance, conflicts, and de-dupe.
Mailing / physical address · city · ZIP · risk state	Point-in-time	Where the risk sits — the first cut at cat zone, rate territory, and admitted status.
NAICS code · SIC code · entity type	Point-in-time	Class-based rating and appetite/knockout; NAICS is validated against reference data.
Business description · years in business · website	Point-in-time	Occupancy hazard beyond the code, plus a stability and continuity signal.

B · Broker & submission

The ask itself — who’s placing it, what product, when it incepts, and the limit being requested. Source: broker email.

Parameter	Type	Why it matters
Broker name · broker email	Point-in-time	Distribution routing, correspondence, and the binding relationship.
Product	Coverage term	The line and form being requested — workflow and appetite routing.
Effective date	Coverage term	Inception and the binding deadline the quote has to beat.
Requested limit	Coverage term	The capacity the broker is asking for — the core of the ask.

C · Exposure & financials

The account’s size and the exposure base behind the property values — revenue, payroll, and headcount. Source: application form.

Parameter	Type	Why it matters
Estimated annual revenue / sales	Point-in-time	Scales business-income exposure and sanity-checks the values on the schedule.
Total payroll · employee count	Point-in-time	Operational scale and occupancy intensity behind the buildings.

D · Coverage structure & limits

The shape of the property program — the total values at risk, how the limit is applied, and the perils in scope. Source: application form.

Parameter	Type	Why it matters
Total insured value (all locations)	Coverage term	The single biggest number in the account — the base the program is built on.
Number of locations · number of buildings	Point-in-time	Spread of risk and the size of the schedule the SOV must reconcile to.
Blanket vs. scheduled limit basis	Coverage term	Whether the limit floats across locations or is pinned per building — a real coverage difference.
Coinsurance percentage	Coverage term	The insured-to-value requirement — drives penalty exposure at claim time.
Requested perils	Coverage term	Named-peril vs. special form, and whether wind, flood, or quake are in.

E · Valuation & deductibles

How losses are valued and how much risk the insured retains — including the separate wind/hail structure that so often carries the cat exposure. Source: application form.

Parameter	Type	Why it matters
Valuation basis (RC / ACV)	Coverage term	Replacement cost vs. actual cash value — a direct driver of loss severity and premium.
All-other-perils (AOP) deductible	Coverage term	The retained layer on ordinary losses — frequency buffer and rate offset.
Wind / hail deductible (flat & percentage)	Coverage term	The cat retention — a percentage wind deductible on a coastal TIV can be the whole risk-transfer story.

F · Business income & extra expense

The time-element cover — often the most underestimated exposure on a property account, and the one that turns a fire into a claim years long. Source: application form.

Parameter	Type	Why it matters
Business income limit	Coverage term	Lost-earnings exposure during restoration — frequently the largest single component of a total loss.
Extra expense limit	Coverage term	The cost of staying operational after a loss — scope and adequacy of the time-element cover.

G · The locations schedule (SOV) — one row per location

This is the heart of the account. InsightXtract reads the Statement of Values workbook as a table — one row per location, with the value split and COPE for each. Source: SOV workbook (Excel).

Parameter	Type	Why it matters
Address · state	Per-row	Geocodes the exposure — cat zone, territory, and TIV concentration; state is validated against reference data.
TIV at location	Per-row	Where value actually sits — the concentration that drives PML and cat load.
Building value · contents value · BI / rental value	Per-row	The value split — how a loss at this site would actually break down.
Construction type (ISO class)	Per-row	The single strongest predictor of fire damageability — frame to fire-resistive.
Occupancy	Per-row	What happens inside — the hazard the walls contain.
Protection class · sprinklered	Per-row	Fire-service capability and active protection — a direct rate modifier per building.

H · Peril sublimits — one row per peril

A property program is rarely one limit for everything. InsightXtract captures the sublimit schedule as a table — one row per peril, with its cap and its own deductible. Source: SOV workbook.

Parameter	Type	Why it matters
Peril	Per-row	Flood, quake, wind, and other capped perils — where the real cat exposure is fenced off.
Sublimit	Per-row	The most the program pays for that peril — the true limit for the exposure that matters.
Deductible	Per-row	The retention for that peril — often far larger than the AOP deductible.

I · Protection & COPE at the account

The account-level protection and roof profile — the fastest read on physical risk quality before the schedule is even opened. Source: application form.

Parameter	Type	Why it matters
Protection class (ISO PPC)	Point-in-time	Public fire-protection grade — a core property rate driver.
Sprinklered	Point-in-time	Active fire suppression — a major credit and, if absent, a red flag.
Alarm type	Point-in-time	Detection and central-station monitoring — response speed and theft risk.
Roof age	Point-in-time	The number-one driver of wind and hail losses — aging roofs are where cat claims start.

J · Loss history — per-claim table + rollups

The single biggest pricing input after the schedule itself. InsightXtract reads each claim as a row, plus the account-level prior-loss summary. Source: application form.

Parameter	Type	Why it matters
Per claim: claim number · date of loss · cause of loss · status	Per-row	Each loss’s identity, timing, nature (fire, water, wind), and whether it’s still open.
Per claim: paid · reserve · incurred	Per-row	Cost, development, and open exposure still on the books.
Prior losses count · prior losses amount	Point-in-time	Frequency and total severity at a glance — the fast read on rate adequacy.
Expiring / target premium	Coverage term	The rate-change baseline — price-to-beat on a renewal.

From categories to one connected record

Pulling these parameters out of three document types is only half the job. The value is in consolidation: the insured named on the application, the broker on the email, and the buildings on the SOV workbook 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 insured details, the SOV wins on per-location values — and validates the result against reference data (state codes and NAICS are checked; the insured name is required). Currency and date fields are normalized on the way out, so TIV, deductibles, and loss amounts land in a consistent, comparable shape.

Why it matters to the business

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

• COPE drives the price. Construction, occupancy, protection, and exposure are captured per building — the difference between a frame strip mall and a fire-resistive warehouse at the same TIV is exactly what the rate has to reflect.
• Business income is the biggest hidden exposure. The BI and extra-expense limits, and the per-location BI values, are pulled by default — because time-element loss is where a fire quietly becomes the largest claim.
• Peril sublimits and wind/hail deductibles are the cat story. Capturing the sublimit schedule and the separate wind/hail structure means the true limit and retention on flood, quake, and wind reach the underwriter — not just the headline limit.
• The loss trend is visible. Each claim as a row, plus the prior-loss rollup, lets you rate to how the account has actually run — frequency, cause mix, and open development — instead of a single summary number.
• Consistency and auditability. The same categories, coded the same way, validated against the same reference data, every time — with provenance to the source document. That is the difference between a repeatable book and one that depends on which underwriter opened the workbook.

The Commercial Property agent extracts all of this today — the insured and account, the broker submission, the coverage structure and limits, valuation and deductibles, business income and extra expense, the account-level COPE, and three linked tables: the per-location SOV, the peril sublimit schedule, and the loss history — consolidated into one coded, cited record. The point this post makes is why it matters: the schedule and the sublimits are exactly what move a property 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.