Automate Carrier Selection and Rate Negotiation

Small logistics firms often run on a Transportation Management System (TMS) like MercuryGate or McLeod, supplemented by Outlook and Excel. The TMS is a great system of record for loads and financials, but it fails as a system of intelligence. Its built-in rating tools only work for large carriers with expensive EDI or API connections, leaving out the dozens of smaller, regional carriers that offer competitive rates.

Consider a 15-person 3PL that needs to quote a new FTL lane from Chicago to Atlanta. A logistics coordinator starts by searching their email for past loads on that lane. Then they check a shared spreadsheet of carrier contacts. They log into three separate carrier portals to pull spot rates. After 45 minutes of searching, they email five carriers, get three responses over the next two hours, and manually key the rates and fuel surcharges into another spreadsheet to find the cheapest option. This entire process takes half a day for a single quote, delaying customer response and introducing data entry errors.

The structural problem is that a TMS is designed to manage structured data, but the most valuable rate information lives in unstructured documents. Emails, PDFs, and Word documents from smaller carriers are invisible to the TMS. There is no off-the-shelf tool that can read an email thread, extract the final negotiated rate for a specific lane, and make it searchable alongside API-based rates from major carriers.

This gap forces firms to rely on institutional memory and manual effort. The consequences are rate leakage from not always finding the best option, slow quote times that lose business, and high operational costs from coordinators spending hours on low-value data entry instead of building carrier relationships.

An engagement with Syntora begins with a data audit. We would collect a representative sample of rate confirmations, quotes, and email exchanges from your top 15-20 carriers. The goal is to map every data source and document format to understand what information is available and where it lives. This audit produces a clear data ingestion plan and identifies the highest-value data to target first.

The technical approach involves building a data processing pipeline on AWS Lambda using Python. The Claude API is ideal for this task; it can parse the unstructured text and tables from emails and PDFs, extracting key fields like origin, destination, equipment type, rate, and accessorial charges. This structured data is then stored in a Supabase PostgreSQL database. We've used this exact pattern to process complex financial documents, and the same principles apply directly to logistics paperwork.

The delivered system would be a simple FastAPI service that exposes an API endpoint your team can query. When a coordinator needs a quote for a new load, they input the lane details. The API queries the database and returns a ranked list of carriers based on historical performance, cost, and availability, all in under 10 seconds. This system would integrate with your existing TMS to enrich load data, not replace your core workflow.