Rate Con AI Parser

★ Core Capabilities

What It Does

The Rate Con AI Parser reads broker rate confirmations in any format and converts them into structured load data. Here are the six key data categories it extracts from every document.

📍

Pickup Location & Date

Extracts the full pickup address including facility name, street address, city, state, and ZIP code. Parses the scheduled pickup date and time window, including early/late pickup times when specified. Handles addresses formatted as single lines, multi-line blocks, or embedded within paragraph text. Facility contact names and dock numbers are captured when present on the document.

🏁

Delivery Location & Date

Identifies the consignee address and delivery appointment with the same precision as pickup data. For multi-stop loads, the parser detects and extracts all intermediate stops in the correct sequence. Delivery window constraints (must deliver by, appointment required, FCFS) are parsed and flagged so dispatchers know which loads have tight deadlines and which offer scheduling flexibility.

💰

Rate & Accessorials

Captures the primary line-haul rate whether expressed as a flat rate, per-mile rate, or percentage. Identifies accessorial charges including detention pay, lumper fees, TONU (truck ordered not used), fuel surcharges, and stop-off fees. The parser distinguishes between the total rate and individual line items, and calculates the rate per mile when the total and mileage are both available.

🏢

Broker Information

Reads the broker company name, their internal load or reference number, and the broker agent’s contact details. This data is critical for matching the rate con to the correct broker account in your TMS and for invoicing after delivery. The parser recognizes common broker letterhead formats from major freight brokerages and adapts its extraction logic accordingly.

📋

Load Number & Reference IDs

Extracts the broker’s load number, PO numbers, BOL numbers, and any other reference identifiers printed on the rate confirmation. These are mapped to the appropriate fields in your TMS load record. The parser handles cases where multiple reference numbers appear on the same document and distinguishes between broker load numbers, shipper PO numbers, and internal reference codes based on label context.

⚠

Special Instructions

Captures free-text notes that brokers include on rate confirmations: temperature requirements for reefer loads, hazmat placarding requirements, driver-assist instructions, appointment scheduling procedures, facility entry requirements, and commodity handling notes. These instructions are placed in the load notes field in your TMS so the dispatcher and driver both see them before the load moves.

PDF rate con being parsed by AI in real time

🤖 Under the Hood

How AI Parsing Works

This is not a simple text search or regex pattern matcher. The Rate Con AI Parser uses machine learning to understand document structure the way a human dispatcher would.

Rate Con Arrives

The rate confirmation enters the system through one of three channels. The most common path is email: you forward the rate con (or set up automatic forwarding rules) to a dedicated parsing address. The second option is manual upload through the TMS dashboard, where you drag and drop a PDF file. The third option is fax-to-email, where inbound faxes are converted to PDF and fed into the same pipeline. All three channels converge into a single processing queue within seconds.

OCR Extracts Text from PDF or Image

The document enters the OCR (Optical Character Recognition) stage. For native PDFs (digitally generated by broker software), the text layer is extracted directly with perfect accuracy. For scanned documents or faxed images, the OCR engine converts the image to text. The system uses multiple OCR passes at different resolutions and compares results to maximize accuracy on low-quality scans. The OCR stage also detects the document orientation and auto-rotates if needed.

AI Model Identifies Fields

This is where the intelligence happens. Unlike traditional parsers that rely on fixed templates or regex patterns, the VAU0 AI model understands the semantic meaning of text on the page. It knows that “Ship From,” “Origin,” “Pickup Location,” and “PU” all refer to the same concept. It recognizes that a dollar amount near a “Rate” or “Total Compensation” label is the line-haul rate, while a dollar amount near “Detention” is an accessorial charge. The model has been trained on thousands of rate confirmations from hundreds of different brokers.

Structured Data Validated

After the AI extracts field values, a validation layer checks the data for consistency. Are the pickup and delivery addresses real locations? Is the pickup date before the delivery date? Is the rate within a reasonable range for the lane? Does the ZIP code match the city and state? Validation catches extraction errors before they reach your TMS. Fields that fail validation are flagged for human review rather than silently passed through with incorrect data.

Load Created in TMS

The validated data is used to create a new load record in the VAU0 TMS. All fields are pre-populated: origin and destination addresses, dates and times, broker name and contact, load reference numbers, rate and accessorial charges, and special instructions. The original rate con PDF is attached to the load record for reference. The load appears on your dispatch board with a “Parsed — Review Required” status indicator.

Carrier Reviews and Confirms

The dispatcher opens the parsed load and reviews all extracted fields against the original rate con, which is displayed side by side. Any field can be edited if the parser made an error. Once the dispatcher is satisfied that the data is correct, they click “Confirm” and the load status changes to “Ready for Dispatch.” The entire review typically takes 15 to 30 seconds because the dispatcher is checking pre-filled fields rather than typing everything from scratch.

Rate con to load creation in VAU0 Portal

⏲ Time Savings

Before vs After

The difference between manual data entry and AI-powered parsing is dramatic. Here is what your workflow looks like in both scenarios.

✗

Manual Process

1. Dispatcher receives rate con via email or fax
2. Opens the PDF and reads through the entire document
3. Types the pickup address into the TMS manually
4. Types the delivery address, dates, and times
5. Enters rate, broker info, reference numbers, and instructions
6. Double-checks entries against the PDF for typos
7. Saves the load — total time: 5 to 10 minutes

Common errors: Transposed ZIP codes, wrong delivery dates, missed accessorial charges, misspelled facility names, incorrect broker load numbers. Each error can lead to missed appointments, billing disputes, or loads dispatched to the wrong location.

✓

With AI Parser

Rate con arrives via email, upload, or fax
AI extracts all fields in under 30 seconds
Load created in TMS with every field pre-populated
Dispatcher reviews parsed data against original PDF
One click to confirm — load is ready for dispatch
Total time: 30 seconds parsing + 15 seconds review

Result: A dispatcher processing 30 rate cons per day saves approximately 3 hours of manual data entry. Over a month, that is 60+ hours freed up for actual dispatching, driver communication, and load optimization.

5–10 min

Manual Entry Time

45 sec

AI Parse + Review

95%+

Field Accuracy

3 hrs/day

Saved (30 loads)

📄 Format Support

Supported Formats

The parser handles rate confirmations from virtually any broker, regardless of how the document is formatted or delivered.

🗎

PDF Rate Confirmations

The most common format. Native PDFs generated by broker TMS software are parsed with near-perfect accuracy because the text layer is clean and structured. The parser handles single-page and multi-page PDFs, landscape and portrait orientations, and documents with embedded tables, logos, and watermarks. Password-protected PDFs are supported if you provide the password during upload.

📧

Emailed Rate Confirmations

Many brokers send rate confirmations as email attachments. You can forward these emails to your dedicated parsing address, and the system automatically extracts the PDF attachment and processes it. Some brokers include rate con details in the email body rather than an attachment — the parser handles this too by reading the email content directly. Inline images and HTML-formatted emails are both supported.

📠

Scanned & Faxed Documents

Some brokers still send rate confirmations by fax. Fax-to-email services convert these to image-based PDFs, which the parser processes using OCR. The system handles typical fax artifacts: slightly skewed pages, reduced resolution, fax headers, and noise around the margins. Scanned paper documents uploaded as JPEG or PNG images are also supported through the same OCR pipeline.

🏢

Broker-Specific Formats

The AI model recognizes rate confirmation formats from major freight brokerages including CH Robinson, TQL, XPO Logistics, Coyote Logistics, Echo Global, Landstar, JB Hunt Brokerage, Schneider, and dozens more. Each broker uses a different layout, terminology, and field arrangement. The parser adapts to each format automatically without requiring manual template configuration. New broker formats are learned from the first document submitted.

Multi-stop loads: The parser correctly identifies and sequences multiple pickup and delivery stops on a single rate confirmation. Each stop’s address, date, and instructions are extracted as separate records and linked in the correct order within the TMS load record.

🔍 Transparency

Accuracy & Edge Cases

We believe in being honest about what AI can and cannot do. Here is a straightforward assessment of the parser’s capabilities and limitations.

✓ What It Catches Reliably

Standard fields: addresses, dates, rates, and reference numbers from any common broker format
Multi-stop loads with up to 10 pickup and delivery locations on a single rate con
Accessorial charges including detention, layover, lumper, TONU, and fuel surcharges
Broker contact information: company name, agent name, email, and extension
Temperature requirements for reefer loads (both Fahrenheit and Celsius)
Weight and piece count when specified on the rate confirmation
Appointment types: FCFS, appointment required, and specific time windows
Commodity descriptions and hazmat indicators
Equipment type requirements: dry van, flatbed, reefer, step deck, etc.

⚠ What It Might Miss

• Handwritten notes added to printed rate cons after they were generated (OCR struggles with cursive handwriting)
• Heavily formatted documents with overlapping text boxes, colored backgrounds, or unusual table structures
• Non-standard layouts from very small brokers who create rate cons in Word or Excel with no consistent structure
• Ambiguous accessorials where the charge description does not clearly indicate what the fee covers
• Extremely low-resolution faxes where characters are indistinguishable (below 150 DPI)

Human review is always available. Every parsed load goes through a confirmation step before it becomes active on your dispatch board. The original rate con PDF is displayed alongside the extracted data so you can verify accuracy. No load is ever dispatched based solely on AI parsing without a human confirming the details. This design ensures the parser augments your workflow rather than replacing human judgment.

📈 ROI Analysis

Rate Con Parser vs Manual Entry

A side-by-side comparison of key metrics for carriers processing rate confirmations manually versus using the AI parser.

Metric	AI Parser	Manual Entry
Time per rate con	✓ 30–45 seconds	5–10 minutes
Data entry errors per 100 loads	✓ 3–5 (flagged for review)	12–18 (often undetected)
Cost per month (30 loads/day)	✓ $0 (free through 2026)	$2,400+ (dispatcher labor)
Scalability	✓ Unlimited volume	Limited by headcount
Processing availability	✓ 24/7 automated	Business hours only
Multi-stop handling	✓ Auto-sequenced	Error-prone manual entry
Accessorial capture	✓ All line items	Often missed or simplified
Document archival	✓ Auto-attached to load	Filed separately (if at all)
Consistency	✓ Same process every time	Varies by dispatcher
New broker adaptation	✓ Automatic learning	No improvement over time

Bottom line: For a carrier handling 30 loads per day, the AI parser saves an estimated 60 to 75 hours of dispatcher time per month. That time can be redirected to higher-value work like negotiating rates, optimizing routes, and communicating with drivers.

🔗 Connected Workflow

Integration with TMS

The Rate Con AI Parser is not a standalone tool — it is deeply integrated into the VAU0 Carrier TMS. Here is how parsed data flows through the rest of your operations.

📄

Parsed Data → TMS Load Board

Once the parser extracts data from a rate confirmation, a new load record appears on your TMS dispatch board. The load card shows the origin and destination cities, pickup and delivery dates, rate, and broker name. A “Parsed” badge distinguishes AI-created loads from manually entered ones. The load is fully searchable, filterable, and sortable alongside all your other loads. Dispatchers can immediately see the new load in their queue without refreshing the page — real-time updates push the load card to all active dashboard sessions.

👤

Driver Assignment → Tracking

After confirming the parsed load, the dispatcher assigns a driver. The driver receives the load details on their VAU0 Driver App with all the extracted information: addresses, dates, special instructions, and navigation-ready coordinates. From that point, the load enters the standard tracking workflow. GPS location updates flow from the driver app to the TMS, and the dispatcher can monitor the load’s progress on the live tracking map.

💰

Delivery → Invoicing

When the driver marks delivery complete and uploads the POD, the load transitions to the billing stage. Because the rate and accessorial charges were already parsed from the rate confirmation, the invoice is pre-populated with the correct amounts. The back office simply reviews and submits. The broker’s load number and reference IDs (also parsed from the rate con) ensure the invoice matches the broker’s records, reducing payment disputes and accelerating collections.

🚀 Quick Start

Getting Started

Setting up the Rate Con AI Parser takes less than 5 minutes. There are three steps to go from signup to automated parsing.

Connect Your Email or Upload Portal

In the VAU0 TMS settings, navigate to the Rate Con Parser section. You have two options for getting rate confirmations into the parser. Option A: set up email forwarding by adding a rule in your email client that automatically forwards all rate confirmations to your dedicated parsing address (provided during setup). Option B: use the manual upload interface in the TMS dashboard to drag and drop PDF files directly. Most carriers use email forwarding for automation and manual upload as a backup for rate cons received through other channels like fax or hand-delivery.

Rate Cons Parsed Automatically

Once email forwarding is active, every rate confirmation that arrives is processed automatically. The parser runs 24 hours a day, 7 days a week. A rate con emailed at midnight on a Saturday will be parsed and ready for review by the time your dispatcher opens the TMS on Monday morning. The average processing time is under 30 seconds per document. You will receive a notification in the TMS each time a new load is created from a parsed rate con, including a confidence score indicating how certain the parser is about its extraction accuracy.

Review and Confirm in Your TMS Dashboard

Open the TMS dispatch board and look for loads with the “Parsed — Review Required” status. Click into any load to see the extracted data displayed alongside the original rate con PDF. Verify that the addresses, dates, rate, and special instructions are correct. Edit any fields that need adjustment. Click “Confirm” to move the load to “Ready for Dispatch” status. The confirmed load is now ready for driver assignment. Over time, as the parser processes more documents from the same brokers, its accuracy improves and the review step becomes even faster.

< 5 min

Setup Time

24/7

Processing

30 sec

Per Document

Through 2026