Skip to content
Document extraction

Rate Confirmation Extraction Software

Every rate con becomes structured data before anyone opens the PDF

Rate confirmations arrive as PDFs from hundreds of brokers, each with its own layout. Pysar.AI reads them the way a human does — finding the linehaul rate, the fuel surcharge, the lane, and the load number wherever they sit on the page — and returns clean fields your team and your systems can act on.

Book a DemoSee it run
pysar engine · live

Overview

The rate con is where the money is agreed — and where the typing starts

Every load begins with a rate confirmation, and every rate confirmation begins a chain of manual work. Someone opens the PDF, reads the linehaul rate, notes the fuel surcharge basis, checks the free time before detention accrues, and types it all into a TMS or a spreadsheet. Multiply that by a few hundred loads a week and the cost is measured in headcount.

The typing is not the real problem. The real problem is that the typed version becomes the version of record. If the rate was keyed as $2,850 when the document said $2,580, nothing downstream can catch it — the invoice will be checked against the wrong number, and the difference is simply lost.

Extraction removes the retyping step entirely. The values that enter your systems are the values printed on the document, with a confidence score attached to each one so anything ambiguous gets a human look before it is trusted.

Why every broker's rate con looks different — and why that stops mattering

There is no standard rate confirmation. Some brokers issue a dense single page with rates in a right-hand column; others spread the same information across two pages with the accessorial schedule in a footnote. Load numbers appear as 'Load #', 'Reference', 'Order', or 'Pro'. Fuel surcharge might be a percentage, a per-mile figure, or a flat amount.

Template-based tools handle this by asking you to configure each layout, which works until the broker changes their form or a new customer sends their first load. Pysar.AI is layout-agnostic: the model was trained on transportation documents rather than on one company's forms, so it recognises a linehaul rate by what it is, not by where it sits.

In practice that means a new broker's rate confirmation extracts correctly on the first upload, with no setup task standing between you and the load.

Extraction is only useful if it feeds verification

Pulling fields off a rate confirmation has limited value on its own. The value appears when those fields are compared against the rest of the load package — the bill of lading that records what actually shipped, the proof of delivery that records that it arrived, and the carrier invoice that asks for payment.

Because extraction and cross-verification run in the same platform, the agreed rate is checked against the billed rate automatically, and accessorials that never appeared on the rate confirmation are flagged as soon as they show up on an invoice.

That is the difference between saving data-entry time and preventing overpayment. Both are worth having; only the second one shows up in margin.

Fields extracted from a rate con

  • Broker name, MC number, and contact
  • Carrier name, MC/DOT number, driver
  • Load, order, and reference numbers
  • Origin and destination with ZIP
  • Pickup and delivery dates and appointment windows
  • Linehaul rate and fuel surcharge basis
  • Accessorial schedule and detention terms
  • Equipment type, weight, and commodity
  • Payment terms and quick-pay options

Caught · Rate confirmation allows two hours of free time before detention accrues. The carrier invoice billed detention from the first hour — a $150 difference flagged before payment.

What the extractor handles

Any broker layout, first upload

No per-broker templates, no mapping rules, and no sample documents to train on. New formats work immediately.

Scans, photos, and faxes

Rate cons that arrive as phone photos or third-generation faxes are deskewed and read, with lower-confidence fields marked for review.

Confidence scores per field

Every value carries a score, so high-confidence fields flow straight through and only genuinely uncertain reads reach a person.

Straight into cross-verification

Extracted rate terms feed the comparison against BOL, POD, and carrier invoice without a second upload.

How it works

  1. 1

    Rate con arrives

    Forward it to your Pysar inbox, drop it in the browser, or push it through the API — however the broker sent it.

  2. 2

    Document is classified

    The model identifies it as a rate confirmation rather than an invoice or BOL, which determines the fields it looks for.

  3. 3

    Fields are extracted

    Rates, lanes, dates, parties, and accessorial terms are pulled with a confidence score on each value.

  4. 4

    Data lands where you work

    Structured JSON to your TMS or factoring platform via API or webhook, ready to check the rest of the load package against.

Related

BOL data extractionFreight invoice auditGuide: how to verify a rate confirmation

Questions

Rate confirmation extraction, answered

What is rate confirmation extraction software?

Software that reads rate confirmation PDFs and pulls out structured data — broker, carrier, linehaul rate, fuel surcharge, accessorials, lane, load number, payment terms — automatically, replacing manual data entry.

What fields does Pysar.AI extract from a rate con?

Broker and carrier details (names, MC/DOT numbers, contacts), load and reference numbers, origin/destination, pickup/delivery dates, linehaul rate, FSC, accessorial charges, and payment terms.

Can it handle rate cons from any broker?

Yes. Every broker formats rate cons differently — Pysar.AI is layout-agnostic and reads new formats on the first upload without template setup.

Can extracted rate con data be compared against the BOL and invoice?

Yes — that's the point. Extraction feeds directly into cross-verification: rate con vs BOL vs carrier invoice, with every mismatch flagged before funding or payment.

Ready to Stop Manual Document Processing?

See Pysar.AI in action with your own documents. Book a 30-minute personalized demo.

Try the Interactive DemoBook a Demo

Join companies already processing 500K+ documents with Pysar.AI