Skip to content
US MGA — ACORD

ACORD Form Automation — Data Extraction for US MGAs and Brokers

ACORD form automation for US MGAs, wholesale brokers, and retail agencies. AI extraction from ACORD 25, 125, 126, 127, 130, 131, 140 with 99%+ accuracy. Extracted data flows directly into underwriting workbenches and rating engines — eliminating 15-30 minutes of manual entry per submission.

ACORD Form Automation — Why It Matters for US MGAs

ACORD forms are the connective tissue of US commercial insurance. Every commercial submission generates ACORD forms — the 125 for general commercial submission, the 126 for casualty, the 127 for commercial general liability, the 130 for workers compensation, the 131 for personal umbrella, the 140 for property. Certificates of insurance flow on ACORD 25. Across the US commercial market, ACORD forms are the lingua franca between retail agents, wholesale brokers, MGAs, and carriers.

The problem is not the standard — ACORD is well-designed for its purpose. The problem is the manual labour wrapped around it. ACORD forms arrive as PDFs. Underwriters and underwriting assistants read them, type the data into the underwriting workbench, validate the extraction, and re-key into the rating engine. For an MGA handling hundreds of commercial submissions monthly, this manual processing consumes 15-30 minutes per submission and introduces transcription errors that downstream into rating, binding, and bordereaux reporting.

ACORD form automation eliminates the manual step. Regure's document AI extracts every field from an ACORD PDF — insured name and address, FEIN, business description and NAICS, prior carrier information, premium history, claims history, named additional insureds, requested coverages, limits, deductibles — with field-level accuracy exceeding 99% on standard ACORD versions. The extracted data flows into your underwriting workbench or rating engine via API, ready for underwriter evaluation without manual re-entry.

For the broader US MGA context, see US MGA solutions. For the FNOL side, see FNOL automation.

7 Standard ACORD FormsACORD 25, 125, 126, 127, 130, 131, 140 — full field-level extraction
99%+ Field AccuracyTrained on millions of submissions across all commercial lines
15-30 Min Saved per SubmissionAcross hundreds of submissions monthly — compound time recovery
Direct Rating-Engine FlowExtracted data lands in your rating engine — no manual re-entry

The standard ACORD forms — what each one covers and what Regure extracts

Each ACORD form serves a specific purpose in the commercial insurance distribution chain. Regure extracts the complete field set from each, with line-of-business-aware validation rules.

ACORD 125 — Commercial Insurance Application

The master commercial submission form. Captures applicant identity (insured name, DBA, address, FEIN, entity type), business description and NAICS classification, requested coverages, prior carrier history, loss history summary, and applicant signature. Regure extracts all fields including subjective narrative descriptions of operations.

ACORD 126 — Commercial General Liability Section

The casualty supplement to the 125. Captures liability-specific information: products and operations descriptions, gross sales, payroll, number of employees, subcontractor relationships, contractual liability assumed, and named additional insureds with their relationships. Regure extracts the often-overlooked subcontractor and additional-insured details that drive underwriting decisions.

ACORD 127 — Business Auto Section

Commercial auto supplement. Vehicle schedules, driver schedules, garaging addresses, radius of operations, classes of vehicle use, hazmat indicators, and prior commercial auto loss history. Regure handles the table extraction for vehicle and driver schedules and validates VIN format and driver license formats.

ACORD 130 — Workers Compensation Section

WC supplement with payroll by classification code, owner/officer information, experience modification factor, and prior WC loss history. Regure extracts the classification code and payroll combinations that drive WC rating and validates against the latest NCCI class code table.

ACORD 131 — Commercial Umbrella / Excess

Umbrella and excess submission. Underlying coverage details, requested limits, hazardous operations, prior umbrella history, and underlying carrier information. Regure cross-references the underlying coverage details against the underlying ACORD submissions for the same insured to flag inconsistencies.

ACORD 140 — Property Section

Commercial property supplement. Location schedules with construction details, protection class, occupancy, square footage, year built, building values, business personal property values, business income, and protection features (sprinklers, alarms). Regure extracts the location schedule as structured data ready for property-level rating.

From ACORD PDF to underwriting workbench — the actual flow

Field-level extraction is only useful if the extracted data lands in the underwriting workflow. Regure's ACORD automation includes the integration layer that gets data from PDF to rating engine without manual re-entry.

PDF Ingestion

ACORD PDFs arrive by email, broker portal, or API. Regure ingests, classifies the ACORD type (125, 126, etc.), and queues for extraction. Email-based intake to a designated submission inbox is supported alongside structured broker portal submission.

Field-Level Extraction with Confidence Scores

Every extracted field carries a confidence score. High-confidence extractions flow to the underwriting workbench directly. Low-confidence extractions route to human review with the original PDF and the extracted candidate side-by-side. This human-in-the-loop pattern keeps accuracy above 99% on first-pass.

Rating Engine API Integration

Extracted data flows into the rating engine via API — ISO ERC rating, AAIS rating, custom proprietary rating models. Submission data populates the rating worksheet so underwriters generate quotes without re-keying the application data they just processed.

What US MGAs and brokers ask about ACORD automation

Which ACORD versions does Regure support?

The current ACORD versions (typically the most recent release plus the previous two) are pre-configured. Older versions still in circulation are supported with slightly reduced field-level accuracy but full extraction coverage. State-specific variants (where they exist) are configured as needed.

What happens if the ACORD form is partially completed or unsigned?

The extraction proceeds for the populated fields and flags missing fields against the form's required-field set. Submissions with critical fields missing route to the underwriter or underwriting assistant for completion before the underwriting workflow proceeds.

Can Regure handle non-ACORD submission forms?

Yes. Many MGAs use proprietary supplemental questionnaires alongside ACORD forms — for specialty risks, program business, or specific carriers. Regure trains custom extraction on these proprietary forms as a configuration step.

How does ACORD extraction integrate with our underwriting system?

Regure exposes the extracted data via API for ingestion into the underwriting workbench. For common underwriting platforms (Applied Epic, Vertafore AMS360, custom platforms), pre-built integration templates accelerate the connection. The extracted data populates the underwriting record with the same fields a manual entry would have produced. See integrations.

What about ACORD 25 certificates of insurance?

ACORD 25 certificates are handled differently from submission forms — they document existing coverage rather than requesting new coverage. Regure handles 25s by extracting the certificate details and validating against the insured's policy records. Compliance checks against contractually required limits and additional insured language are automated where the contract terms are available. See ACORD forms glossary entry.

See ACORD extraction on your actual submissions

Book a 20-minute demo. Bring sample ACORD forms from your book — we'll extract the data live and show you how it flows into your underwriting workflow.