SOLUTIONS · MANUFACTURERS AND DISTRIBUTORS
Manufacturing and distribution
Freight rate contract intelligence is AI that reads your carrier contracts and rate cards (base rates, discounts, fuel surcharge tables, accessorials and minimums), checks every freight invoice against the rate you actually agreed to, and traces what that freight did to the margin on the customer quote. OutcomeCatalyst builds and runs it for manufacturers and distributors as an AI agent on top of the ERP, TMS exports and spreadsheets they already use. Freight audit and payment providers are good at paying and auditing carrier bills; this sits on the shipper side and answers the question they usually do not: which customers, lanes and quotes are losing money on freight, and what to put on the table at the next carrier renewal.
The freight question nobody can answer by Friday
The CFO asks a simple thing. How much did we spend on freight last quarter that we did not recover from customers?
Nobody knows. The carrier contract is a PDF in someone's inbox, with two amendments and a fuel table pasted in as an image. The rate cards are spreadsheets, one per plant or branch, and they drifted apart sometime last year. The freight charged to the customer is on the quote and the sales order in the ERP. What the carrier billed is on an invoice that AP paid, or that your freight audit provider paid for you.
The gap sits between systems, and it sounds like this: we quoted freight at one number and paid another, and we go into renewals with the carrier's data instead of ours.
What is actually in a carrier contract
Most freight leakage hides in the terms people read once at signing. A typical LTL or truckload agreement for a shipper includes:
Base rates or a discount off a base tariff, by lane, weight break and freight class, often with an absolute minimum charge per shipment.
A fuel surcharge table, usually stepped against the weekly U.S. on-highway diesel price the U.S. Energy Information Administration publishes, with a base price, a step size and a percentage or per-mile charge per step.
Accessorial schedules: liftgate, residential delivery, inside delivery, limited access, appointment, detention, reweigh and reclass fees, each with its own rule for when it applies.
Effective dates and amendments that override earlier pages.
An invoice can be wrong against any one of those. The fuel step applied a week late. A minimum billed on a lane where the contract waived it. Small on one shipment, real across thousands, and tedious to catch by hand.
What the agent does, step by step
Reads every contract and rate card into one model. The agent extracts lanes, weight breaks, classes, discounts, minimums, fuel tables and accessorial rules from PDFs and spreadsheets, and keeps track of which amendment is in force on which date.
Compares rate cards across carriers and plants. Same lane, same weight, different plant: the agent lines up what each carrier would charge under each contract so you can see where one site is paying more than another for the same freight.
Matches each freight invoice to the contracted rate. For every shipment it rebuilds the expected charge from the contract in force on the ship date, including the fuel step for that week, and compares it line by line with what the carrier billed. Variances get a reason: wrong fuel step, accessorial not supported by the contract, minimum misapplied, class change, duplicate.
Ties freight back to the customer quote. It joins the shipment to the sales order and the quote that priced it, and computes quoted vs actual freight cost per order, so you can see whether a customer, lane or freight term (prepaid and add, free freight over a threshold) is losing money.
Drafts the dispute or the fix. Carrier errors get a dispute with the clause and math attached. Underquoted freight gets traced to the quoting rule or rate table that caused it.
Builds the renewal file. Before a carrier negotiation it assembles your own history: volume by lane, accessorial frequency, billed vs contracted variance and how each carrier compares on the lanes that matter.
Quoted vs actual freight, a worked example
Here is the math with illustrative numbers, not a customer result. A distributor ships 12,000 LTL orders a year. Customer quotes use a freight table built two years ago. Since then fuel moved and the main carrier started billing limited access fees on job site lanes.
Line | Quoted to customer | Billed by carrier | Gap |
|---|---|---|---|
Linehaul after discount | $142.00 | $142.00 | $0.00 |
Fuel surcharge | $28.40 | $30.20 | $1.80 |
Accessorials | $0.00 | $1.70 (average across orders) | $1.70 |
Total per shipment | $170.40 | $173.90 | $3.50 |
A $3.50 gap is invisible on one order. On 12,000 shipments it is $42,000 a year that came straight out of gross margin, and nobody signed off on it. The carrier invoice may be entirely correct, which is why a pure invoice audit would never flag it. The leak is in the quote. It is why we argue for measuring true margin per SKU and cost to serve rather than trusting the gross margin on the order.
The systems it reads
The agent works from a connected view of the systems your teams already use. OC builds it as a governed context layer we call the brain, without copying your data into a new warehouse:
Carrier contracts and amendments, as PDFs, scans and email attachments.
Rate cards and fuel tables, usually spreadsheets kept by plant, branch or logistics manager.
ERP: customers, quotes, sales orders, freight terms, shipments, AP and the GL accounts freight posts to.
TMS or shipping system exports, where you have one.
Freight invoices or audit provider files: the carrier bills themselves, or the data file your freight audit and payment provider already produces.
Public index data: the EIA weekly diesel price that most fuel surcharge tables reference.
Freight is one slice of a bigger problem. Our direct spend agent applies the same contract-to-invoice logic to materials and supplier pricing, and the email to quote agent for distributors uses this freight history to check freight before a quote goes out, which is where the leak is cheapest to stop.
Where freight audit and payment providers fit
If you already use a freight audit and payment provider, keep it. These firms do work OC does not do, at a scale we would not claim:
Cass Information Systems says it processes and pays $37 billion in freight spend and 35 million freight invoices a year, with more than 15,000 carriers in its payment network and payment in 114 currencies, backed by its own bank subsidiary. For a large shipper that wants carrier payment, accruals and audit handled as one outsourced service, that is a strong fit.
nVision Global describes audit and payment across all modes, validation of carrier invoices against contractual terms, duplicate payment prevention and AI with expert oversight across 200+ audit points. Its freight audit campaign page claims savings of 6% to 10% of freight spend, and it positions itself for Fortune 1000 companies.
Trax Technologies describes AI-powered audit of 100% of invoices across modes, carriers and regions, $25 billion in transportation spend managed, 125+ global enterprise customers in 120 countries, a 5% to 7% average spend reduction, and GL assignment down to SKU and charge level.
Those are each vendor's own claims, read on their sites on October 6, 2026. All three center on the carrier invoice and answer "was this bill correct and paid on time?" very well.
Freight rate contract intelligence starts from a different question. Did the freight we charged customers cover the freight we paid, and if not, which quotes, terms and lanes caused it? That needs the customer quote, the sales order and the rate tables your sales team uses, which sit in your ERP and spreadsheets rather than in the audit provider's system. The two work together: the audit provider's data file becomes one of the sources the agent reads.
What stays human
Disputes. The agent drafts them with evidence. Your logistics or AP lead decides which to send.
Contract interpretation. When a clause is ambiguous or two documents conflict, a person settles it once and the agent applies that ruling from then on.
Price and term changes. Sales and finance decide whether to reprice an under-recovering customer or accept it for a strategic account.
What to measure
Name the unit before anything gets built. For this work the unit is the shipment, and the core number is freight gap per shipment: freight recovered from the customer minus freight paid to the carrier, rolled up by customer, lane, carrier, plant and freight term. Alongside it:
Share of carrier invoices matched to a contracted rate without manual work.
Billed vs contracted variance per shipment, by reason code.
Dollars disputed and dollars recovered per month.
Number of quotes flagged for underquoted freight before they went out.
Baseline it on the first pass through history, then ask at every review whether the gap is shrinking.
How a project starts
It starts with a working session, not a software trial. We ask for a sample: your main carrier contracts and amendments, the rate cards your quoting team uses, and a few months of freight invoices or your audit provider's file, plus read access to the matching orders in the ERP. From that sample we show you where invoices and quotes diverge from the contracts and roughly how big the gap is. If it is not worth chasing, you will know that too. If it is, OC builds the connections, trains the agent on how your team reads contracts and handles exceptions, and runs it with your team approving the work.
Frequently asked questions
Is there AI that reads our carrier rate contracts and flags invoices that do not match?
Yes. The agent extracts rates, fuel tables, accessorials and minimums from your contracts and amendments, rebuilds the expected charge for each shipment from the terms in force on the ship date, and flags lines that do not match, with the clause attached.
We already pay a freight audit company. Why would we need this?
You may not, if your only goal is accurate carrier payment. Audit providers check the carrier bill well. This work adds the customer side: quoted vs actual freight per order, under-recovery by customer and freight term, rate card drift between plants, and a renewal file built from your own history. It can read your audit provider's data rather than replace it.
How do we compare freight rate cards across carriers and plants?
The agent puts every rate card into the same lanes, weight breaks and classes, applies each contract's discounts, minimums and current fuel step, and shows what the same shipment costs under each carrier and at each plant.
Do we need a data team to run this?
No. OC builds the connections, trains the agent and runs it. Your leads approve its work and correct it when it is wrong.
Do you copy our data into a new warehouse?
No. The agent reads from the systems you already run through a governed context layer. Your ERP, TMS and contract files stay where they are and remain the system of record.
Our contracts are scanned PDFs and the fuel table is an image. Does that work?
Usually, yes. Where a scan is unreadable or a term is unclear, the agent asks a person instead of guessing and keeps that answer for later invoices.
Why not build this ourselves?
You can. Parsing one contract is easy. Keeping amendments, fuel steps, rate cards, quotes and invoices joined after go-live is what stalls internal builds. Our buy vs build guide lays out the trade-offs honestly.
Sources
Cass Information Systems, "Freight Audit and Payment: Why Cass," accessed October 6, 2026. cassinfo.com
nVision Global, Freight Audit, accessed October 6, 2026. corporate.nvisionglobal.com
nVision Global, "AI-Driven Freight Audit and Payment," accessed October 6, 2026. corporate.nvisionglobal.com
Trax Technologies, homepage, accessed October 6, 2026. traxtech.com
U.S. Energy Information Administration, "Gasoline and Diesel Fuel Update," weekly on-highway diesel prices by region, accessed October 6, 2026. eia.gov
Book a strategy call to see this on your own operation.
Your systems, your documents, and what your people have been carrying around in their heads. Thirty minutes to see what an AI brain could look like in your company.
Book a strategy call
© 2026 OutcomeCatalyst