Extracting data from fleet lease quotes
Fleet management companies price every vehicle on its own open-end lease quote, and each company lays its rate page out differently. Fleet managers and procurement offices receive them as emailed PDFs, bundled into packets that cover a whole replacement cycle. Public agencies attach the packet to a board agenda for approval.
Buyers need the capitalized cost, depreciation reserve, monthly lease charge, maintenance fee and reduced book value at term. A fleet management information system compares those figures across vehicles, then projects each unit's monthly cost and its gain or loss against book value at resale.
What makes fleet lease quotes hard
One quote per vehicle
A fleet order arrives as a packet of near-identical quotes, each repeating the same labels. Every figure has to stay with its own quote.
Two monthly totals, similarly labeled
The rate page sums the lease lines into a monthly rental excluding services, then adds the maintenance program into a second total.
Every lessor has its own layout
Rate pages differ from lessor to lessor. One prints every amount to the left of its label, beside a separate panel of order details.
Parse, extract, verify
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.
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.
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.
One commercial fleet lease quote, end to end
About this document: Seven open-end (equity) lease rate quotes from Enterprise Fleet Management to Hidalgo County, Texas, 32 pages in all, published in the county's public meeting record. Each quote prices one pickup or van over four to six pages: a rate page followed by vehicle pricing and feature lists. Page 5 is the rate page of the second quote, a Ford F-250. Personal and account identifiers were replaced with stand-ins before publication.
Processing objective: The objective is to follow one quote from the vehicle to its value at the end of the term, with each dollar figure tied to its own label. On the rate page every amount sits to the left of its label, and a boxed Order Information panel runs alongside the dollar column.
Demonstration fields: Vehicle model, Lease term (months), Total capitalized amount, Total monthly rental, Reduced book value at term
The document
4 of 32 pages
Page 1
Page 2
Page 3
Page 29
Where each value came from
quotes[1].vehicle.model
{
"value": "F-250",
"page": 5,
"range": {
"end": 13513,
"start": 13478
},
"box": {
"xmin": 0.10899,
"ymin": 0.12226,
"xmax": 0.36821,
"ymax": 0.13596
}
}quotes[1].term_months
{
"value": 60,
"page": 5,
"range": {
"end": 13635,
"start": 13570
},
"box": {
"xmin": 0.01921,
"ymin": 0.1529,
"xmax": 0.48863,
"ymax": 0.16709
}
}quotes[1].total_capitalized_amount
{
"value": 38000,
"page": 5,
"range": {
"end": 14742,
"start": 14731
},
"box": {
"xmin": 0.04541,
"ymin": 0.34447,
"xmax": 0.2095,
"ymax": 0.36002
}
}quotes[1].total_monthly_rental_including_services
{
"value": 679.71,
"page": 5,
"range": {
"end": 15642,
"start": 15634
},
"box": {
"xmin": 0.04593,
"ymin": 0.60617,
"xmax": 0.2094,
"ymax": 0.629
}
}quotes[1].reduced_book_value_at_term
{
"value": 8360,
"page": 5,
"range": {
"end": 15730,
"start": 15720
},
"box": {
"xmin": 0.04597,
"ymin": 0.62874,
"xmax": 0.20939,
"ymax": 0.6467
}
}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
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. VPC and on-prem deployment require an Enterprise plan.
