*** September 2026 Major Release ***Read and watch
Document type

Extracting data from business licenses

Cities, counties and states each issue business licenses in their own format, and most renew them every year or two. Banks, lenders and supplier diversity programs collect one from every business they onboard, as a scan or PDF uploaded with the application.

Onboarding needs the license number, the issuing jurisdiction, the business name and the validity dates. Business verification records them in the vendor or customer file, and flags the license for renewal before it expires each year.

Page 1 of a sample Business License, with extracted values outlined
1,778Markdown characters
35parse blocks
28fields extracted
The problem

What makes business licenses hard

Every jurisdiction issues its own

Local governments each design their own license, and a Florida business shows a local business tax receipt when asked for one.

Typed values beside printed labels

Labels print in one typeface and the typed values in another, sometimes on a different baseline from the label they belong to.

Valid for a fixed period

Each license covers a set period, often one fiscal year, so its validity dates decide whether the copy on file is still current.

How it works

Parse, extract, verify

1

Parse

Parse turns the document into reading-order Markdown and a hierarchy of blocks (text, tables, figures), and grounds each block to its exact location on the page.

2

Extract

You define the fields in a JSON schema, and Extract returns their values in a consistent, structured format, each linked back to where it appears in the Markdown.

3

Verify

Each value links back to where it appears, down to the character range in the Markdown, and the page and bounding box in the source document, so every result can be verified.

Worked example

One business license, end to end

About this document: A specimen Broward County, Florida, local business tax receipt, published as a sample document by a supplier diversity council. One landscape page, a single image with the publisher's sample banner across it, covering the business, its owner, its counts and the fees paid. Page 1 is featured. The business and owner are placeholders.

Processing objective: The objective is to read the receipt as a record. It shows who issued it, which receipt it is and how long the business has operated, with the count it is assessed on and what was paid. Each value grounds to its own typed entry on the image.

Demonstration fields: Receipt number, Issuing county, Business opened, Employees, Total paid

The document

1 page
  • Page 1 of the sample Business LicensePage 1shown below

Where each value came from

Page 1 of the sample Business License, with every featured field outlined
Receipt numberreceipt_number
327-123456
The region of the sample Business License containing Receipt number, with the extracted value outlined
extraction_metadata["receipt_number"]
{
  "value": "327-123456",
  "page": 1,
  "range": {
    "end": 260,
    "start": 237
  },
  "box": {
    "xmin": 0.58788,
    "ymin": 0.15777,
    "xmax": 0.96908,
    "ymax": 0.21109
  }
}
Issuing countyissuing_authority.name
BROWARD COUNTY
The region of the sample Business License containing Issuing county, with the extracted value outlined
extraction_metadata["issuing_authority.name"]
{
  "value": "BROWARD COUNTY",
  "page": 1,
  "range": {
    "end": 41,
    "start": 0
  },
  "box": {
    "xmin": 0.1832,
    "ymin": 0.04782,
    "xmax": 0.78813,
    "ymax": 0.08014
  }
}
Business openedbusiness.date_opened
2020-03-27
The region of the sample Business License containing Business opened, with the extracted value outlined

Printed on the page as Business Opened: 03/27/2020, normalized on extraction.

extraction_metadata["business.date_opened"]
{
  "value": "2020-03-27",
  "page": 1,
  "range": {
    "end": 438,
    "start": 410
  },
  "box": {
    "xmin": 0.56429,
    "ymin": 0.22956,
    "xmax": 0.80907,
    "ymax": 0.25088
  }
}
Employeescapacity.employees
2
The region of the sample Business License containing Employees, with the extracted value outlined
extraction_metadata["capacity.employees"]
{
  "value": 2,
  "page": 1,
  "range": {
    "end": 539,
    "start": 525
  },
  "box": {
    "xmin": 0.36783,
    "ymin": 0.34409,
    "xmax": 0.53489,
    "ymax": 0.39474
  }
}
Total paidfees.total_paid
16.5
The region of the sample Business License containing Total paid, with the extracted value outlined
extraction_metadata["fees.total_paid"]
{
  "value": 16.5,
  "page": 1,
  "range": {
    "end": 957,
    "start": 952
  },
  "box": {
    "xmin": 0.80567,
    "ymin": 0.48268,
    "xmax": 0.90777,
    "ymax": 0.51286
  }
}

How to try it yourself

Everything on this page is reproducible. The sample, its schema, and the Parse and Extract API outputs are published. You can repeat the same two calls from whichever surface you already build on.

  • Claude Skilldescribe the job; the agent writes it
  • CLIno code at all
  • MCPexpose ADE docs as an agent tool
  • REST APIany language
  • Pythonlandingai-ade
  • TypeScriptlandingai-ade
Why LandingAI?

Documents into data

Accurate

The best accuracy on the difficult documents other tools get wrong, such as messy scans, complicated tables, checkboxes and handwriting.

Auditable

Every answer cites its sources. Values come back with an audit trail (page, bounding box, character range) so a human can review them.

Agentic

Applies multi-step reasoning, and plans before it acts. Handles documents it has never seen before and adapts to variations.

API-first

Composable APIs that drop into any tech stack and can be called from anywhere. Easy to build on and embed in your pipelines and products.

Affordable

Cost-competitive rates with complete price transparency. Spend controls before you run, and every job reports its exact cost.

Enterprise-grade

Built for production and scale from day one. Fully observable and monitored, running securely in our cloud or yours.

Built for high-volume, high-throughput production

  • Zero data retention
  • Priority rate limits
  • Enhanced support
  • VPC and on-prem deployment

Available on Team and Enterprise plans.

Compare plans
Capabilities

Business License processing highlights

License or receipt number
Issuing jurisdiction
Business name and address
Owner name
Validity period
Business opened date
Employee and unit counts
Fees and total paid
Image-only scans
Dates normalized to ISO
No per-jurisdiction templates
Bounding box for every value
FAQs

Questions, answered

Yes. Parsing works from the rendered page, so a license scanned or saved as a single image parses the same way as a native PDF. Every value is grounded to its position on the image.

Yes. Describe the start and end of the validity period as date fields. Each comes back in ISO form, and grounding still points at the date as the license printed it.

No. A schema names the fields you want and describes them in plain language. Extraction reads the parsed license for meaning, so one schema covers each city's, county's and state's own format.

Every extracted value carries the character range it was quoted from in the parsed Markdown. The Ground API maps that range to a page and a bounding box, so a reviewer sees each value on the line that printed it.

Start extracting business license data

Reliable, structured outputs with full traceability in minutes