{}

Insurance Document Automation: Dec Pages, ACORD Forms & Claims

Declaration (dec) pagesACORD formsClaims formsProof of insurance

The problem

Agencies and MGAs manually re-key dec pages and ACORD forms into their agency management system at every renewal, audit, or new-business submission. Claims teams re-type the same claim-form fields into claims systems by hand.

How Dokyumi solves it

Define a schema for the fields you track — named insured, policy number, carrier, effective/expiration dates, coverage limits, premium, deductible — and get a dedicated endpoint. Submit scanned dec pages or ACORD forms.

Because extraction is field-semantic rather than a rigid template, the same schema generally holds up across different carriers' dec-page layouts and common ACORD form variants.

Example schema and result

A real schema shape for insurance documents, and the JSON you get back from one extraction.

Schema — GET /api/v1/schemas

{
  "slug": "dec-page-parser",
  "name": "Declaration Page Parser",
  "description": "Extracts policy and coverage data from insurance declaration pages",
  "ocr_mode": "standard",
  "fields": [
    {
      "key": "named_insured",
      "type": "string",
      "required": true
    },
    {
      "key": "policy_number",
      "type": "string",
      "required": true
    },
    {
      "key": "carrier",
      "type": "string",
      "required": true
    },
    {
      "key": "effective_date",
      "type": "date",
      "required": true
    },
    {
      "key": "expiration_date",
      "type": "date",
      "required": true
    },
    {
      "key": "premium",
      "type": "currency",
      "required": false
    },
    {
      "key": "coverage_limits",
      "type": "array",
      "required": false
    }
  ]
}

Result — POST /api/v1/extract

{
  "id": "ext_in2y8h",
  "status": "completed",
  "schema": "dec-page-parser",
  "data": {
    "named_insured": "Harborview Retail Group LLC",
    "policy_number": "CPK-5502931",
    "carrier": "Meridian Mutual",
    "effective_date": "2026-09-01",
    "expiration_date": "2027-09-01",
    "premium": 4820,
    "coverage_limits": "[ General Liability $1M, Property $500K ]"
  },
  "confidence": {
    "policy_number": 0.99,
    "premium": 0.97,
    "coverage_limits": 0.92
  },
  "meta": {
    "processing_time_ms": 1260,
    "page_count": 3,
    "ocr_cached": false
  }
}

How it works

1

Define your policy/claim schema

Named insured, limits, dates — whatever your workflow tracks.

2

Get a dedicated endpoint

Scoped to that schema alone.

3

Submit dec pages, ACORD forms, or claims

PDF or scanned image, one API call.

4

Sync structured data to your AMS

Push results via webhook into your agency management or claims system.

Insurance FAQ

Does Dokyumi support every ACORD form (25, 27, 28, 125, 126...)?+
Because extraction is schema-first, it can generally handle any ACORD form's fields — but complex, multi-schedule forms may need a more detailed schema (or one schema per form number) to capture every line accurately. Start with the fields you actually need and expand from there.
Does it integrate with EZLynx, Applied, or AMS360?+
Not with a native pre-built connector today. Results come back as JSON via the API or a webhook for you to route into your AMS.
Can it automate claims intake, not just policy dec pages?+
Yes — define a claims-specific schema (claim number, loss date, loss description, coverage amount) the same way you would for a dec page.
What if a dec page is missing a field I asked for?+
It comes back empty and is flagged so you know to check the source document — the rest of the extraction still completes.
Is pricing per page, like some OCR vendors?+
No — Dokyumi is flat-rate per extraction (per document), not per page, so a 12-page commercial package costs the same as a 1-page dec page.

Related use cases

Healthcare Billing

Pull claim numbers, billed/allowed amounts, and CPT codes off EOBs and medical bills instead of re-typing them.

Real Estate & Leasing

Pull lease terms, applicant income, and insurance details out of leases, applications, and dec pages — no per-property template work.

Accounts Payable

Extract vendor, line-item, and total data from invoices and receipts — no template building, no AWS Textract setup.

Stop re-keying declaration (dec) pages by hand.

100 free extractions every month — no credit card, no AWS account, no setup hell.