Part of our work on logistics
Logistics
Why Detention and Demurrage Disputes Are a Documentation Problem
Big Sky Consulting Group · September 21, 2026 · 6 min read

The dispute is not an argument any more
Somebody on your team spends a meaningful part of the week fighting container charges. They know the carriers' portals. They know which terminal was closed. They lose most of the time, and nobody can quite explain why, because the reasoning in the denial is usually one line long.
Here is what changed underneath them. The Federal Maritime Commission's demurrage and detention billing rule took effect on 28 May 2024, and it turned the dispute from a negotiation into a filing. There are now fixed windows on both sides. Vessel-operating carriers and marine terminal operators have to invoice within 30 calendar days of charges last being incurred. NVOCCs get 30 days from the invoice they received. You get at least 30 calendar days to request mitigation, refund or waiver, and the billing party has 30 days to attempt resolution.
Those numbers get quoted constantly, and they are the least interesting part of the rule. The interesting part is the standard your request has to meet once you file it, because that standard is what quietly relocated the whole problem out of your dispute desk and onto your yard.
What "credible evidence" actually means in practice
The rule's language is about written documentation sufficient for an impartial observer to conclude the justification is legitimate. That sounds like lawyer padding until you read how the carriers implemented it.
Take one concrete case. If you dispute a charge on the grounds that there was no empty return location available, CMA CGM requires a valid screenshot of appointment unavailability taken the working day prior to the waiver date. If the terminal runs day and night shifts, you need screenshots covering both. No screenshot, and the submission is rejected as incomplete. Not argued with. Rejected. And the dispute is denied with respect to any day for which credible evidence is not provided, which means a single container's dispute can be half granted and half denied on the basis of which days somebody happened to document.
Read that sequence backwards and the real deadline appears. The outcome was determined on a Tuesday afternoon, by whether a dispatcher or a yard clerk took a screenshot that nobody had asked them for, of a screen they had no particular reason to open, about a charge that did not exist yet. By the time an invoice reaches your AP inbox three weeks later, the case is already won or lost. Your dispute team is not fighting charges. It is discovering, one at a time, what was and was not captured.
This is the pattern we see in every operation with a charge-dispute function, and it is not confined to ocean freight. It is the same shape as procurement timelines that turn out to be a document problem and not a policy problem: a team staffed to argue about an outcome that was fixed upstream, in a step nobody owns, by information nobody was told to record.
The money is real enough to justify the attention
Between April 2020 and March 2025, nine major ocean carriers collected roughly 15.4 billion dollars in demurrage and detention, on the FMC's own quarterly collection from CMA CGM, COSCO, Evergreen, Hapag-Lloyd, HMM, Maersk, MSC, ONE and Yang Ming. The trend has come off its peak. Billings and collections in the first quarter of 2025 fell 24 percent and 19 percent respectively against the prior quarter, and waivers ticked up 7 percent.
That last number is the one worth staring at. Waivers going up while billings come down is the market telling you that the charges are more contestable than they were, which raises rather than lowers the return on being able to contest them properly.
On the regulatory side, the rule survived its first serious challenge. On 23 September 2025 the D.C. Circuit set aside only 46 CFR 541.4, the section specifying who an invoice may be sent to. Everything else stands, including the required invoice content and the 30-day issuance deadline, and a carrier that misses those loses the obligation to be paid the charge as billed. If your team reads only one thing about the court decision, it should be that the invoice requirements it left intact are a second, separate line of defence that most shippers never check.
This is the general shape of the problem. Which parts apply to your process depends on answers only your systems can give.
Put us on it, from $5,000Why building a bigger dispute desk is the wrong instinct
The usual response to all of this is to staff up. Hire an analyst, buy a tool that scrapes invoices, measure recovery against salary. The economics look fine on a slide, because recovery is easy to count.
The problem is what that structure optimises for. A dispute desk is paid to process charges that already exist. It has no lever on the charges it cannot evidence, which are the expensive ones, and it has every reason to keep its volume high. We have watched this settle into equilibrium more than once: the desk recovers a respectable percentage of a number that keeps growing, and nobody asks why the number keeps growing. A recovery function that never reduces its own input is a cost centre wearing a revenue costume.
The leverage is one step earlier and it is unglamorous. Decide what evidence each charge type requires, then make capturing it part of the move rather than part of the argument. Not a policy. A field, a prompt, a timestamp, attached to the thing the driver or the clerk is already doing. When the evidence arrives with the move, the dispute stops being research.
The same logic decides whether automation helps you here. If the capture is not happening, software that files disputes faster just files incomplete ones faster, which is the same conclusion we reach on 3PL billing automation and accessorial rules: the sequencing is a commercial decision, and automating the back half of a broken sequence makes the break harder to see. And as with manual load matching in a brokerage, the cost does not land evenly. It lands on the hard cases, which are also the ones with the most money in them.
What we would ask you first
Three questions, and they are diagnostic rather than a plan.
What percentage of your disputes are denied for insufficient evidence as opposed to denied on the merits? Most operations cannot answer this, because the denial reasons are not coded, and the two failures have completely different fixes.
Which charge types account for most of your denied days, and what evidence would each one have needed? This is where the work is, and it is specific to your lanes, your terminals and your carriers. There is no generic list.
Who, by name, sees the first-pass acceptance rate on disputes every month? If the answer is nobody, that is the finding.
Answering those properly means sitting inside your gate process, your TOS screenshots, your carrier reason codes and your invoice audit for a couple of weeks. It is not something an article can do for you, and anyone who tells you otherwise is selling a dashboard. Rather than freight it further: the charges are containerised, but the problem is not.
If you are staffing a dispute desk against a number that will not come down, talk to us about where the evidence is actually being lost. We will tell you which of your charge types is worth the capture and which ones you should simply budget for, including the case where the honest answer is that your current process is fine and the money is somewhere else.
