Freight Bill Audit and Payment Companies

What a freight bill audit and payment company does

Freight is one of the few large cost lines in a company's accounts that arrives already itemized, already priced by the vendor, and almost never checked. A mid-sized manufacturer can receive tens of thousands of carrier invoices a year, each with a base charge, a fuel surcharge and a variable tail of accessorial charges, against contracts running to dozens of pages of rate logic. Accounts payable is not equipped to evaluate any of it, so the function gets outsourced.

Scale makes the point. Cass Information Systems, the Nasdaq-listed provider that anchors the category, processed 35 million freight invoices representing roughly $37 billion of freight spend in 2025 across more than 15,000 carriers, and disbursed approximately $94 billion across all its invoice lines. Note that payment here means disbursement, not acceptance: these companies pay a shipper's vendors on its behalf and are not merchant acquirers.

Why carrier invoices are wrong often enough to justify the function

Most invoice errors are mechanical rather than dishonest. Freight rating is genuinely hard, carriers rate at scale under time pressure, and the inputs move. The errors nonetheless recur in predictable families.

• Accessorial charges. Detention, liftgate, residential delivery, reconsignment and limited access are billed as exceptions to the base rate, and whether each was legitimately incurred depends on facts recorded at a dock rather than in a rate table.
• Fuel surcharges. These reset on published indices at defined intervals, and applying the wrong week's index is both easy and invisible without recalculation.
• Classification and reweighs. In less-than-truckload freight the class assigned to a commodity drives the rate, and carriers reclassify and reweigh in their own terminals; in parcel, dimensional weight adjustments do the same job.
• Duplicate billing. A carrier issues a corrected invoice without cancelling the original, or the same invoice arrives through two channels. Cass runs explicit duplicate-payment detection for exactly this reason.
• Complex movement types. Cass names the cases its audit logic handles: pool shipments, stop-offs, milk runs, intermodal, Rule 11, white-glove service and spot quotes. Rule 11 is worth defining because it produces the most double payments — it is a rail arrangement in which each carrier bills separately for its own segment rather than one carrier issuing a single through rate, so one physical move generates multiple invoices and multiple chances to pay twice.

The mistake most companies make here is assuming accounts payable can do this with a checklist. An AP clerk can confirm an invoice matches a purchase order. No AP clerk can determine whether detention was correctly assessed on the second stop of a stop-off move that travelled intermodally under a Rule 11 arrangement. The function is not clerical; it is a rating engine plus people who know freight.

The six stages of the workflow

Every serious provider runs a version of the same six stages, and knowing where each sits determines what you can hold a provider accountable for.

1. Receipt and ingestion. Invoices arrive as EDI transactions, flat files or paper to be imaged and captured. The provider also ingests the shipper's own shipment data, usually from a transportation management system, because the shipment record rather than the invoice is the reference truth. A provider working only from invoices can check arithmetic but not reality.
2. Pre-audit against contracted rates. A rating engine recalculates what each shipment should have cost from the carrier contract; Cass does this through its Ratemaker application. The check happens before the money leaves, which is the point — prevention is a credit nobody has to chase.
3. Exception handling. Anything the rules cannot resolve goes to trained freight audit staff who work it against the contract and dispute it with the carrier. This stage separates providers and is the one buyers under-examine. The meaningful metrics are first-pass match rate, exception volume as a share of invoices, and median days to close an exception — not a claim about auditing 100% of invoices, which everyone makes.
4. Payment. The approved amount is disbursed on an agreed calendar over ACH, check, card or cross-border rails. Cass disburses through Cass Commercial Bank, its own wholly owned FDIC-insured bank; most competitors instruct payment through a partner bank.
5. Post-audit. A second review, often by a different firm and often priced on contingency, hunts what the pre-audit ruleset missed — duplicate payments, missed contract terms, expired accessorial waivers.
6. Reporting. GL-coded spend data, dashboards and benchmarking flow back into the ERP and into procurement's next carrier negotiation.

Accrual and GL coding: the part finance is actually buying

Ask a transportation team why they use a freight audit provider and they will talk about recovered overcharges. Ask the controller and you get a different answer, and the controller's answer usually justifies the contract.

Freight invoices arrive weeks after the freight moves. Without a rating engine the month closes on an estimate that is fundamentally a plug. With one, the rated shipment record generates an accrual derived from the actual contract at the moment the shipment happens, and that accrual is trued up as real charges land. That is a materially better close, and it is auditable in a way a plug is not.

The second half is allocation. Approved charges are coded to the general ledger at whatever level the shipper specifies: cost center, plant, business unit, mode, customer, sometimes SKU. Freight is frequently the largest cost in a manufacturer's P&L that nobody can attribute, and without shipment-level coding a company knows what it spent but not which products, customers or lanes consumed it.

The mistake here is treating the accounting output as a free extra. At a mature shipper with clean contracts, recovery alone often does not pay for the contract; accrual quality, coded spend data and negotiating leverage generally do. Build the case on all three, or the renewal conversation will be awkward when recovery rates fall — which they should, because a functioning audit program trains carriers to bill correctly.

The economics nobody explains: float

This is the section that reframes the category, and it is almost entirely absent from vendor materials and from the buying guides that circulate in procurement.

Cass Information Systems is publicly traded and has to disclose the split, which makes it the best available window into how this category earns. In its FY2025 results Cass reported fee revenues of $108.0 million against net interest income of $81.2 million, out of total net revenue of $190.75 million — roughly 43% of total net revenue coming from interest on money in transit rather than from any stated fee. Its disclosed net interest margin rose from 2.74% in 2022 to 3.83% in 2025, and its filings warn that a decline in market interest rates would hurt results.

Read that back as a buyer and the implication is uncomfortable: a shipper who negotiates only the per-invoice processing fee has negotiated on less than half of the provider's economics. The funding calendar — how many days sit between your money leaving and the carrier's money arriving — is a price term. It is simply never presented as one.

The direction of the float is not fixed either. Cass's FY2025 risk factors state that it remits payment of invoices prior to receiving funds from its customers. In that configuration the provider is not sitting on the shipper's cash at all; it is extending the shipper credit, and will underwrite and price that credit accordingly — Cass recorded $7.847 million of bad debt expense in 2024 on exactly this exposure. So the same provider may hold your money on some arrangements and lend you money on others, depending entirely on what the contract says about funding timing, and a shipper with weaker credit may be offered terms that remove the working-capital benefit it thought it was buying. Note too that where a provider retains interest on funds in transit, paying carriers faster is not costless to it.

What to ask a provider about float, funding and interest

Almost nobody asks these questions, which is why they work. Ask them in writing, early, and treat evasion as information.

1. On what date do we fund, and on what date is the carrier paid? As a calendar, per payment method — ACH, check and card legs frequently differ.
2. Who holds the money between those dates? A bank you own, a partner bank, a trust account, or your own balance sheet?
3. Are our funds segregated or custodial? An insured deposit, a trust asset and an on-balance-sheet payable behave very differently in an insolvency.
4. Who earns the interest, and is any of it credited back to us? Ask for the number, not the principle.
5. If you pay carriers before we fund, are you extending us credit? If so, is it priced into the fee, and what is the implied rate?
6. If market interest rates fall, does our fee go up? Ask whether a repricing clause is tied to rates.
7. Who keeps early-payment discounts negotiated with carriers? If the provider captures a discount for paying early using float it earns interest on, the shipper funds both sides of that trade.
8. Who keeps duplicate-payment recoveries and carrier credits, and when are they returned? Money recovered on your behalf that sits with the provider for ninety days is float too.
9. What happens to uncashed checks and unclaimed balances? Escheatment and residual balances belong in the contract.
10. On termination, how long do you hold funds and how is the final float settled? Exit terms here are the least examined and often the most expensive clause in the agreement.

Bank-owned providers, non-bank providers, and AP platforms

Bank-owned or bank-affiliated. Cass owns Cass Commercial Bank, founded in 1906, FDIC-insured and regulated by the Missouri Division of Finance, the Federal Reserve Board and the FDIC; Cass Information Systems itself is a financial holding company under the Bank Holding Company Act. U.S. Bank Freight Payment is the other major bank-backed offering. Choosing between those two is choosing between two regulated bank balance sheets, and the advantages are genuine: supervised custody of funds in transit, capital requirements, and in Cass's case published financials a buyer can read.

Specialist non-bank processors. nVision Global, A3 Freight Payment, Trax Technologies, Intelligent Audit, enVista and Engine (formerly CTSI-Global) compete on audit depth, analytics and international coverage, and instruct payment rather than hold a banking licence. Several are deeper in specific areas such as parcel spend, but the counterparty analysis differs, and with private companies the balance sheet is not visible to the buyer.

AP automation platforms are a different product. BILL, Tipalti, Corpay and Modern Treasury can all pay a carrier competently, run approval workflow and sync to the ledger. None of them rates freight. They cannot tell you the fuel surcharge used the wrong index week or that a Rule 11 move was billed twice, because rating against carrier contracts is not what they do. If the requirement is contract-based freight rating, an AP automation platform is the wrong choice; where freight spend is small and contracts are simple, the reverse holds and a freight specialist is overkill.

One caution on bank ownership: a provider that owns a bank has a direct institutional interest in deposit balances, which is precisely the float above. The supervision is valuable; the alignment of interest on payment timing is not.

When a freight audit provider is the wrong choice

This category sells itself on recovered overcharges, which makes it sound as though it always pays for itself. It does not, and the failure modes are predictable.

• Small shippers with simple, flat-rate agreements. With three carriers on straightforward rates, implementation and coordination cost more than the recoveries.
• Shippers whose rate data is not clean. The audit is only as good as the contracts loaded into the rate engine. If agreements are not current, not machine-readable, or not the ones actually being applied, the provider validates against the wrong truth and issues a confident rubber stamp.
• Buyers who want audit without payment. It can be bought that way, but the float economics that make this business work sit on the payment leg, so audit-only pricing is worse and the provider is structurally less invested.
• Buyers who will not interrogate funding terms. Given that a listed provider here disclosed roughly 43% of total net revenue coming from net interest income, negotiating only the per-invoice fee is a decision to be priced on terms you never examined.
• Buyers who need real-time payment rails. Providers generally market scheduled, controlled disbursement; Cass publishes no RTP or FedNow support as of mid-2026.

For everyone else the evaluation is straightforward to run and rarely run properly. Take ninety days of real invoices, run a parallel test with two shortlisted providers, and measure four things: first-pass match rate, median days to close an exception, accrual accuracy against final actual cost, and the working-capital effect of each provider's funding calendar. Then score total cost as processing fee plus the value of the float plus the cost or benefit of the credit terms. That number, not the per-invoice fee, is what the contract actually costs.

Frequently asked questions