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

Extracting data from Form 413 statements

Owners and guarantors of a small business complete SBA Form 413, a personal financial statement, for most SBA loan and certification applications. Lenders collect one from every principal named on each application, as a fillable PDF or a scan uploaded to the loan portal.

Credit analysts need total assets, total liabilities, net worth, annual income and contingent liabilities. The lender's credit memo uses them to judge each guarantor's liquidity and the strength of the guarantee behind the SBA loan.

Start for FreeTalk to Us
Page 2 of a sample SBA Form 413, with extracted values outlined
22,209Markdown characters
265parse blocks
53fields extracted
The problem

What makes SBA Form 413 statements hard

Answers sit in compact fields

Applicant responses sit inside compact fields over the form's fixed labels, whether the completed form arrives as a fillable PDF or a scan.

A balance sheet in two columns

Assets run down the left and liabilities down the right, each line joined to its amount by a dotted leader across the column.

Schedules mostly blank

Supporting schedules for notes, securities and real estate run for pages, and a typical applicant fills only a few of their rows.

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 SBA Form 413, end to end

About this document: A completed SBA Form 413 personal financial statement for a sole proprietor, filled in through the form's own fillable fields. Six portrait pages: program checkboxes, the balance sheet and income on page 2, supporting schedules, the certification and notices. Page 2 is featured. Every name and identifier was replaced with stand-ins.

Processing objective: The objective is to read the balance sheet and confirm that it balances. Total liabilities plus net worth equals total assets, and the business type is a ticked box above them. Each value grounds to its own line of the balance sheet.

Demonstration fields: Business type, Total assets, Total liabilities, Net worth

The complete example is on GitHubThe sample document, the full extraction schema, the complete parse output and every extracted field.
Open the folder

The document

4 of 6 pages
  • Page 1 of the sample SBA Form 413Page 1
  • Page 2 of the sample SBA Form 413Page 2shown below
  • Page 3 of the sample SBA Form 413Page 3
  • Page 4 of the sample SBA Form 413Page 4

Where each value came from

Page 2 of the sample SBA Form 413, with every featured field outlined
Business typebusiness.business_type
Sole Proprietor
The region of the sample SBA Form 413 containing Business type, with the extracted value outlined
extraction_metadata["business.business_type"]
{
  "value": "Sole Proprietor",
  "page": 2,
  "range": {
    "end": 5948,
    "start": 5808
  },
  "box": {
    "xmin": 0.05194,
    "ymin": 0.23463,
    "xmax": 0.96877,
    "ymax": 0.26697
  }
}
Total assetsassets.total_assets
23,000
The region of the sample SBA Form 413 containing Total assets, with the extracted value outlined
extraction_metadata["assets.total_assets"]
{
  "value": 23000,
  "page": 2,
  "range": {
    "end": 8075,
    "start": 8067
  },
  "box": {
    "xmin": 0.37894,
    "ymin": 0.5952,
    "xmax": 0.53784,
    "ymax": 0.62231
  }
}
Total liabilitiesliabilities.total_liabilities
0
The region of the sample SBA Form 413 containing Total liabilities, with the extracted value outlined
extraction_metadata["liabilities.total_liabilities"]
{
  "value": 0,
  "page": 2,
  "range": {
    "end": 7814,
    "start": 7811
  },
  "box": {
    "xmin": 0.83314,
    "ymin": 0.54395,
    "xmax": 0.96888,
    "ymax": 0.55657
  }
}
Net worthliabilities.net_worth
23,000
The region of the sample SBA Form 413 containing Net worth, with the extracted value outlined
extraction_metadata["liabilities.net_worth"]
{
  "value": 23000,
  "page": 2,
  "range": {
    "end": 7909,
    "start": 7901
  },
  "box": {
    "xmin": 0.8332,
    "ymin": 0.55649,
    "xmax": 0.96887,
    "ymax": 0.56943
  }
}

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
Run this sample yourselfThis document on GitHub
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

SBA Form 413 processing highlights

Total assets and liabilities
Net worth
Annual income sources
Contingent liabilities
Business type checkboxes
Program checkboxes
Notes payable schedules
Real estate and securities schedules
Fillable form field values
Dotted-leader balance sheets
Statement as-of date
Bounding box for every value
FAQs

Questions, answered

Yes. Parsing works from the rendered page, so values typed into form fields read the same way as printed text. Each value is grounded to the field where it appears on the page.

Yes. Parsing writes every checkbox into the Markdown in one standard form, marked checked or unchecked beside its label. A business type field comes back as the option ticked on the form.

Yes. Total assets, total liabilities and net worth come back as numbers, so liabilities plus net worth can be compared with assets. A mismatch points straight to the lines to review.

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 amount on the line that printed it.

Start extracting SBA Form 413 data

Reliable, structured outputs with full traceability in minutes