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

Extracting data from commercial leases

Landlords and their counsel draft each commercial lease, usually opening with a summary of basic lease terms ahead of dozens of pages of provisions. Lease administrators, lenders and acquisition teams abstract every lease in a building, from PDFs in a data room or signed copies from the landlord.

Abstraction needs the parties, the premises and rentable area, the term, the base rent schedule and the security deposit. Lease accounting software builds the payment schedule from those, and a lender's model projects the building's rental income.

Start for FreeTalk to Us
Page 1 of a sample Commercial Lease, with extracted values outlined
2,447Markdown characters
38parse blocks
43fields extracted
The problem

What makes commercial leases hard

Numbers written in words

Leases state key terms in words with numerals in brackets, such as "forty-eight (48) full calendar months" for the term.

Rent that steps up

Base rent runs as a schedule of periods, each with its own monthly amount and rate per square foot, often with abated first months.

A grid of defined terms

The basic lease information is a two-column grid of defined terms and their values, mixing short entries with full paragraphs of prose.

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 commercial lease, end to end

About this document: The Basic Lease Information page of an office lease for an entire building in California, filed publicly on SEC EDGAR by the tenant as a material contract. One portrait page: the parties' notice blocks, the premises, a 48-month term, an escalating base rent table and the deposit. Page 1 is featured. Both parties are companies.

Processing objective: The objective is to abstract the lease's core economics: who the tenant is, how much space it takes, for how long, at what rent and against what deposit. The rentable area and the term are read out of sentences that spell each number in words.

Demonstration fields: Tenant, Rentable square feet, Term (months), Monthly base rent (months 2-12), Security deposit

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

1 page
  • Page 1 of the sample Commercial LeasePage 1shown below

Where each value came from

Page 1 of the sample Commercial Lease, with every featured field outlined
Tenanttenant.name
Serve Robotics Inc.
The region of the sample Commercial Lease containing Tenant, with the extracted value outlined
extraction_metadata["tenant.name"]
{
  "value": "Serve Robotics Inc.",
  "page": 1,
  "range": {
    "end": 326,
    "start": 298
  },
  "box": {
    "xmin": 0.05252,
    "ymin": 0.28595,
    "xmax": 0.36419,
    "ymax": 0.31529
  }
}
Rentable square feetpremises.rentable_square_feet
4,200
The region of the sample Commercial Lease containing Rentable square feet, with the extracted value outlined
extraction_metadata["premises.rentable_square_feet"]
{
  "value": 4200,
  "page": 1,
  "range": {
    "end": 659,
    "start": 385
  },
  "box": {
    "xmin": 0.23118,
    "ymin": 0.36143,
    "xmax": 0.94832,
    "ymax": 0.40859
  }
}
Term (months)term_months
48
The region of the sample Commercial Lease containing Term (months), with the extracted value outlined
extraction_metadata["term_months"]
{
  "value": 48,
  "page": 1,
  "range": {
    "end": 778,
    "start": 661
  },
  "box": {
    "xmin": 0.05738,
    "ymin": 0.42339,
    "xmax": 0.94623,
    "ymax": 0.45434
  }
}
Monthly base rent (months 2-12)base_rent_schedule[1].monthly_base_rent
16,800
The region of the sample Commercial Lease containing Monthly base rent (months 2-12), with the extracted value outlined
extraction_metadata["base_rent_schedule[1].monthly_base_rent"]
{
  "value": 16800,
  "page": 1,
  "range": {
    "end": 1047,
    "start": 1034
  },
  "box": {
    "xmin": 0.72662,
    "ymin": 0.56303,
    "xmax": 0.835,
    "ymax": 0.58032
  }
}
Security depositsecurity_deposit
73,431.24
The region of the sample Commercial Lease containing Security deposit, with the extracted value outlined
extraction_metadata["security_deposit"]
{
  "value": 73431.24,
  "page": 1,
  "range": {
    "end": 2303,
    "start": 2274
  },
  "box": {
    "xmin": 0.05207,
    "ymin": 0.87321,
    "xmax": 0.31059,
    "ymax": 0.88924
  }
}

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

Commercial Lease processing highlights

Landlord and tenant parties
Premises and rentable area
Term and commencement
Base rent schedules
Rent per square foot
Rent waivers and abatements
Security deposit
Permitted use
Brokers and notice addresses
Numbers written in words
No per-landlord templates
Bounding box for every value
FAQs

Questions, answered

Yes. Type the field as a number in your schema. A rentable area or lease term written in words with numerals in brackets comes back as a single number. It grounds to the sentence that states it.

Yes. Define the rent schedule as an array of periods in your schema. Each period comes back as its own entry, with its months and monthly base rent, grounded to its row of the rent table.

No. A schema names the fields you want and describes them in plain language. Extraction reads the parsed lease for meaning, so one schema covers each landlord's own form and defined terms.

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 commercial lease data

Reliable, structured outputs with full traceability in minutes