Freight Invoice Audit: Where the 5-10% Billing Errors Hide (2026)

What is a freight invoice audit, and what does it catch?

A freight invoice audit is the line-by-line check of each carrier bill against the contracted rate, the actual shipment, and the purchase order, to find charges that are duplicated, miscalculated, or never earned. It is not the same as AP confirming the total looks about right before cutting a check. AP verifies the invoice is arithmetically correct and has a PO attached; a freight audit verifies every line matches what your contract says you owe.

A freight invoice audit reconciles each carrier invoice against your negotiated rates, the shipment that actually happened, and the PO behind it, to recover charges you never owed. The most common findings are duplicate invoices, phantom accessorials, fuel surcharges run off the wrong index, reweigh and dimensional-weight upcharges, and demurrage past free time. For importers there is a second reason to care: freight charges feed your customs value, so a freight overcharge can quietly inflate the duty you pay CBP too, not just the bill you pay the carrier.

Primary source: Tompkins Ventures, "Why Freight and Parcel Shipping Require Post-Auditing", which establishes the working benchmark that 5 to 10 percent of freight and parcel invoices carry some kind of error.

One disambiguation first, because two very different "audits" get confused. A freight invoice audit is about the money you pay the carrier or forwarder. A duty audit is about the money you pay CBP. This guide is about the first. If you are reconciling what your broker filed against your own records for a CF-28 or reasonable care, that is the separate exercise of reconciling ACE entry data against your own records. The two lanes touch at one point, which we will get to: freight feeds customs value.

What percentage of freight invoices actually have errors?

Freight-audit advisory firm Tompkins Ventures estimates that 5 to 10 percent of freight and parcel invoices carry some kind of error, with cases running as high as 40 percent (Jim Tompkins, Tompkins Ventures, July 2026). That is the most credible named, dated, freight-specific figure available, and it is worth being precise about what it is: an experience-based estimate from a firm that audits freight for a living, not a measured study.

Here is a frame worth internalizing before you cite anything in a CFO deck: the freight-audit stat everyone repeats does not exist. You have seen "up to 20 percent of freight invoices contain errors," or "10 percent per NASSTRAC," or "15 percent per the Journal of Business Logistics." Chase any of them to a primary source and the trail goes cold. NASSTRAC's site does not publish the 10 percent figure; no dated original backs the 15 percent or the "80 percent of carrier invoices" numbers. They are orphan attributions that vendor blogs quote from each other. Budget against the defensible 5 to 10 percent. For a measured adjacent benchmark, Ardent Partners' State of ePayables 2025 reports an average invoice exception rate of about 14 percent across all accounts-payable invoices (Ardent Partners, 2025) - broader than freight, but a real number.

The percentage is not even the useful part. The directionality is. Because the carrier builds the invoice, the errors do not distribute randomly around zero; they lean toward the carrier. You will occasionally find a charge in your favor, but most of what an audit surfaces is money owed back to you. That is why "self-correcting" almost never happens.

Why do carriers not just correct their own billing errors?

Carriers do not self-correct because the incentive runs the other way and the complexity hides the error. This is structural, not a conspiracy claim. The invoice is generated by the carrier's rating engine, priced against a contract with hundreds of lanes, dozens of accessorials, a fuel formula, and free-time terms few people on either side have memorized. When that engine gets a fuel index wrong or applies an accessorial that was not performed, nothing flags it, because from the carrier's side the invoice is "correct." The error only surfaces when someone on your side matches it back to what actually happened and what you agreed to pay.

And most shippers never do that match, because it is hard at volume. A mid-market importer running $6M across LTL, parcel, and ocean can see 800 to 1,500 invoices a month, arriving as unstructured PDFs and EDI in a dozen formats, priced against rate sheets scattered across inboxes and shared drives. AP cannot reconcile a 40-line ocean invoice against a contract nobody can find, so it rubber-stamps it. That gap is where the recoverable spend hides, commonly a few percent of freight spend and, at your volume, real money.

What are the most common freight invoice errors, and who do they favor?

The common errors fall into a taxonomy, and the useful move is to tag each by who it favors and how catchable it is, so you know where to spend your limited hours. Here is the overcharge taxonomy with directionality, the frame I use to triage a stack of invoices.

Error type	Who it favors	How you catch it	Catchability
Duplicate invoice	Carrier	Dedupe on carrier + pro/BOL + amount vs. paid records	Auto-catchable
Wrong rate or class	Carrier	Invoice vs. contracted rate for that lane and freight class	Needs the rate sheet
Accessorial never performed	Carrier	Accessorial vs. a matching shipment event	Needs shipment proof
Fuel surcharge miscalculation	Carrier	Recompute off the contract's named index, peg, escalator	Needs the fuel formula
Reweigh / dimensional-weight upcharge	Carrier	Billed weight/dims vs. your original declaration	Needs original declaration
Demurrage / detention past free time	Carrier	Charged days vs. free-time allowance + gate dates	Needs free-time terms
Currency / conversion error	Either	Billed currency and FX vs. contract currency	Needs contract terms

Read down the "catchability" column and a strategy falls out. Duplicates are free wins you can automate today. Everything below them needs a reference you may not have handy - the rate, the shipment record, the fuel formula, the free-time allowance - which is why manual auditing stalls: not because the errors are subtle, but because the evidence to prove them is scattered. Accessorials deserve special attention: they are both the most common invented charge and, unchecked, one of the largest, because they stack line by line on top of the base rate.

How do you actually audit a freight invoice, step by step?

You audit a freight invoice by running it up a ladder of four matches, each one harder than the last and each catching a different class of error. I call it the four-way match ladder, the spine of the whole method. Most guides tell you to "check the rate" and stop; the rate is only the first rung.

• Rung 1 - Invoice vs. rate. Match every line to your contracted rate for that lane, mode, and freight class. Catches rate misapplication, lost volume discounts, and wrong-class charges. This is the rung AP thinks is the job.
• Rung 2 - Invoice vs. shipment. Match the invoice to what physically happened: the weight and dimensions you declared, the accessorials the shipment actually required, the transit that occurred. Catches reweigh and dim-weight upcharges and phantom accessorials. Requires a shipment record, which is why teams without a TMS stall here.
• Rung 3 - Invoice vs. contract terms. Match the time-and-formula charges to the fine print: the fuel index and escalator, the free-time allowance, the disputability of a demurrage bill. Catches fuel-surcharge errors and demurrage past free time. This is where the money hides on ocean.
• Rung 4 - Invoice vs. PO. Match the paid invoice back to the purchase order and the GL: right amount, right cost center, not already paid. Catches duplicate payments and miscoding. This is the rung Finance cares about most, where freight audit meets PO-to-invoice reconciliation.

Work the rungs in order. Most single-invoice audits die around rung 2 because the shipment evidence is not at hand, and almost nobody reaches rung 4 by hand across thousands of invoices. That is a data problem, not a discipline problem: contracts, shipment records, and POs live in different systems that do not talk to each other - the same trade-data fragmentation that makes duty numbers refuse to reconcile makes freight invoices impossible to audit at scale.

How do you audit an ocean freight invoice for imports?

You audit an ocean invoice in three passes, because it is really three invoices stapled together. Pass one is the base ocean freight against your service-contract rate. Pass two is the surcharge stack - BAF or bunker adjustment, terminal handling (THC), the bill-of-lading fee, delivery order, and chassis - each checked against your contract's terms rather than accepted as "just how ocean works." Pass three is the time-based charges: demurrage and detention, matched against your free-time allowance and the actual gate dates.

The ocean invoice is where rung 3 earns its keep, because the surcharge stack and the demurrage lines are exactly the charges no rate sheet in AP's hands can verify. It is also where the import angle becomes concrete, and this is the frame most freight-audit content ignores: freight overcharges compound into duty errors. For U.S. imports, freight and related charges feed the transaction value that determines your landed cost and, in many cases, the customs value you declare to CBP. An inflated ocean-freight line or a phantom surcharge does not just overpay the carrier - it can inflate the duty you owe, making an ocean-invoice error a compliance issue, not only a cost issue. For how freight rolls into duty, MPF, and HMF, see how to estimate landed cost for U.S. imports; for how those freight and assist figures should be governed inside declared value, see customs valuation governance for manufacturers.

Is a demurrage or detention bill disputable, and how do you contest it?

A demurrage or detention bill may be disputable on procedure alone, and the test is not just whether the charge feels unfair. This is the disputable-demurrage test, the least-commodity lever in freight auditing because most content skips it. Under the Federal Maritime Commission's demurrage and detention billing rule, an invoice must be issued to the correct party within a defined window and must contain specific required information; where a bill fails those requirements, the billed party has grounds to challenge the obligation to pay.

So before you argue the merits, check the paperwork:

• Right party. Were you the correct party to be billed under the rule?
• Timely issuance. Was the invoice issued within the required window?
• Sufficient information. Does it contain the detail the rule requires to be enforceable?

Verify the exact current requirements against the live FMC rule text before you dispute, because the specifics have moved through rulemaking and litigation. The frame is durable even as the details shift: a demurrage bill can be defeated on how it was billed, not only on why.

Should you audit freight in-house, buy software, or outsource it?

Audit in-house until volume and dispute tracking break you, then buy software or outsource - and be honest about which pain you actually have, including when you should not buy anything.

• In-house, by hand. Free, and genuinely the right call below roughly $500K in annual freight spend, or when your errors concentrate in a handful of lanes you can watch closely. If a disciplined checklist on your top 20 lanes captures most of the recoverable dollars, do not pay anyone a percentage to do it. For many smaller importers this is the correct path.
• Software. You license a rate engine, load your contracts once, and it matches every invoice and routes exceptions to your team. Payback typically lands in 6-9 months above about $2M in spend. Best when you want to keep the data and pay function in-house, especially when freight ties into customs value.
• Outsourced freight audit and payment (FAP). A provider audits and pays carriers for you, often priced as a percentage of recovered savings. The incentive aligns, but the honest catch: a savings-share provider optimizes for the recoveries it can capture, not necessarily all of yours, and some hold your payment file. If your real need is outsourcing the entire pay function - or you are parcel-heavy and a pure-play parcel auditor would win on late-delivery and GSR refunds - use them.

There is a fourth honest case: if a capable 3PL or broker already audits your freight as part of their service, you may not need a separate solution at all. Check what you already pay for before buying again.

For most mid-market importers the deciding factor is not cost recovery in isolation - it is whether the tool also reconciles to POs and to customs value, because freight, duty, and PO data are the same fragmented problem wearing three hats. Look for the matching-and-reconciliation layer that turns a manual checklist into a repeatable at-volume process. If your pain is pure outsourced payment, an FAP fits better.

How do you actually recover a freight overcharge once you have found it?

You recover a freight overcharge by moving it through a tracked pipeline from flag to confirmed credit, because finding the error is the easy part and recovering it is where most in-house audits quietly fail. This is the dispute-to-credit gap, the frame that separates an audit that saves money from one that just generates findings nobody collects. Before you open a single dispute, size the prize: our free freight invoice overcharge estimator turns your annual freight spend and error assumptions into a recoverable-dollar range, so you chase the claims worth chasing.

The pipeline has five states, and every disputed charge should sit in exactly one with a named owner:

1. Flagged - the audit surfaced the discrepancy, with evidence attached.
2. Disputed - a formal claim is filed with the carrier, not a "this looks off" email.
3. Credit promised - the carrier agrees; capture the amount and the date.
4. Credit applied - the credit actually appears, on an invoice or a check.
5. Validated - you confirmed the credit landed and the "corrected" invoice did not reintroduce a different error.

Be this formal because disputes die in email. A claim raised in a thread with no status gets lost, and a carrier's re-issued invoice often carries a fresh error in a different line, so a credit you never validate is a credit you never got. Percentage-of-savings providers are disciplined about states 1 through 4 for the recoveries they can capture; whether they chase all of yours is worth asking. However you audit, the recovery pipeline is where the money is made. For recovery on the duty side, the trade-offs between scanning past entries and drawback are in tariff audit software vs duty drawback, and building freight audit into a standing program follows the same logic as a standing CBP audit-readiness program.

The takeaway: freight accuracy is a margin lever and a compliance lever

Treat freight invoice auditing as recurring hygiene, not a one-time hunt, because the errors regenerate every billing cycle and the carrier's incentive never changes. Run the four-way match ladder on every invoice you can, automate the duplicates and rate checks, reserve human hours for the ocean surcharge stack and the demurrage disputes, and never let a found overcharge stall in email. The few percent an audit recovers is real money, but the quieter win is that clean freight numbers give you clean landed-cost and customs-value numbers, so the discipline that protects your freight budget also protects your duty.

If your pain is thousands of unstructured invoices, contracts nobody can find, and disputes that die in inboxes, that is the at-volume matching and reconciliation problem GingerControl works on. If a dedicated FAP or a parcel-only auditor fits your need better, that is the honest answer instead.

Freight overcharges are one of the seven leakage places importers lose money without noticing; The Duty Leakage Recovery Program maps all seven to the documents where each is visible.