Documentation Index
Fetch the complete documentation index at: https://docs.tread.ai/llms.txt
Use this file to discover all available pages before exploring further.
What this solves
A 50-truck day generates 400 tickets. Most billing teams spend the next morning typing them into a spreadsheet — and then a person spends another four hours reconciling weights against dispatch. That’s where the headcount problem starts: every ticket is a person-hour. This workflow puts two AI agents on it. Ticket Chaser reads the ticket image and pulls the fields. AI Auto Approvals scores each load against tolerance rules and queues the clean ones. You only touch the exceptions.Walkthrough
Open the Approvals queue
Go to Approvals → Loads. Every captured load lands here, grouped by job and date. The AI confidence score sits next to each row.
Review what Ticket Chaser pulled
Each load shows the parsed ticket fields — number, gross, tare, net, material, timestamps — beside the ticket image. Click a row to see the side-by-side comparison and any flags.
Let AI Auto Approvals score the load
The agent compares the ticket weight, time onsite, cycle time, and material against the order’s tolerances. Loads inside tolerance get a green check. Loads outside tolerance get an amber flag with the reason.
Bulk-approve the clean loads
Filter to Auto-Approved and click Approve All. These move straight to the next stage — billing, settlements, or driver pay. Most operators approve 70-90% of a day’s loads in one click.
Resolve the flagged loads
Open Needs Review. Each flagged load shows the rule it broke (e.g., “weight 24% over capacity”) and a recommended action. Edit, accept, or reject. The agent learns from each decision.
Send approved loads downstream
Approved loads flow into Settlements & Driver Pay and into customer invoices. No re-export, no re-keying.
Watch it
Common pitfalls
Related
- Track Drivers & Collect Tickets — how loads get into the queue.
- Timesheet & Daily Hour Approvals — the hours side of approvals.
- Tickets & Timesheets — the data model.
- Agent Configurations — set tolerances and rules.