A customer emails on the 22nd about a delivery on the 3rd. Six cartons short. Your pick ticket says the pallet went out complete, your WMS agrees, and by the time you read the email the legal position has already moved against you.
That last part surprises most people. Start there.
Two clocks run on a shortage, and most shippers watch the wrong one.
The first is the filing deadline. In the United States it is not fixed by statute. The Carmack Amendment sets a floor instead: a carrier may not provide for a period shorter than nine months for filing a claim, or shorter than two years for bringing a civil action after the claim is disallowed (49 U.S.C. § 14706(e), read 9 September 2026). The real deadline sits in the bill of lading, and nine months is what most of them say. What a claim has to contain is set out in 49 CFR § 370.3: facts sufficient to identify the shipment, an assertion of liability, and a demand for a specified or determinable amount of money.
The second clock decides the outcome, and it stopped on the day you loaded.
To recover from a carrier under Carmack you have to make out a prima facie case, and the Supreme Court set out the three elements of it in *Missouri Pacific Railroad Co. v. Elmore & Stahl*, 377 U.S. 134 (1964): delivery to the carrier in good condition, arrival in damaged or deficient condition, and the amount of damages. Read the first element again. Good condition at origin is your burden. The carrier does not have to prove the pallet left complete. You have to prove it did.
International road carriage in Europe allocates it the other way, which matters if you ship into the EU. Under Article 30 of the CMR Convention the consignee has until the moment of delivery to object to apparent loss or damage, seven days in writing for damage that is not apparent, and twenty-one days for delay. Acceptance without reservation is prima facie evidence that the goods were received as described in the consignment note (CMR Convention, Article 30, UN Treaty Series, read 9 September 2026). There the presumption hardens against the receiver who stays silent. Under Carmack the burden starts with whoever is claiming. Opposite mechanics, identical instruction to you.
Now read the email on the 22nd again. Nineteen days have passed and neither system has done you a favour. If you are claiming against the carrier, you are the party who has to show the goods left in good order. If your customer is deducting from your invoice, Carmack does not enter into it at all, because that is a contract and UCC question governed by whatever your supply agreement says. All three routes end at the same place, which is your own dock on the day of loading.
You can still win these. Presumptions are rebuttable, burdens are met with ordinary commercial records, and commercial reality is not a courtroom. Most of these are settled by two people comparing what they can show each other, long before anyone reads a statute. But it explains something that otherwise looks like bad luck: the shipper who has nothing to show tends to pay, and the shipper who can produce the pallet tends not to.
The practical consequence is uncomfortable. The evidence has to exist before you know you need it. There is no version of this where you respond to the claim by going to look.
Damage arrives with its own evidence. Something is broken, and a photograph of the broken thing at least establishes that the damage is real. The remaining argument is narrower than it looks: which side of the trailer door.
A shortage has no artefact. There is nothing to photograph at the receiving end, because the whole complaint is an absence. The consignee counts, finds a gap, and reports it. Everything after that is one company's system record against another company's count.
That asymmetry decides more of these than anyone likes to admit. A count is a person who was present. A pick confirmation is a database entry generated by the party now being accused of getting it wrong. Both are honest. Only one of them looks like evidence to a third party.
Here is the part worth taking away, because it changes what you photograph.
A reported shortage is one of four things, and they are not equally common. In descending order of how often they turn out to be the explanation:
One. The count was taken before the unload finished. Freight is still in the nose of the trailer, behind other product, or on a second pallet that has not come off. This resolves itself within a day and generates an apology, provided somebody can show what went on the truck.
Two. It was mislabelled or misdelivered. The cartons exist, at the right total, on the wrong pallet or at the wrong door. A multi-stop load makes this considerably more likely. The consignee is not wrong that their count is short. They are wrong about where the rest of it went.
Three. It genuinely left short. A pick error, a short pallet build, a carton left in staging. This is real, it is yours, and the useful thing is to find out in hours rather than argue for a month and then discover it.
Four. It went missing after it left you. Rarer than shippers assume and less rare than it used to be. Allianz Commercial's Safety and Shipping Review 2026 records nearly 160,000 cargo crimes globally over a two-year period and a five-fold increase in cargo theft losses dating back to late 2022 (Allianz Commercial, "Safety and Shipping Review 2026", p. 9, read 9 September 2026). The BSI Consulting and TT Club 2025 Cargo Theft Report, published April 2026, puts Germany at 27 per cent of recorded European cargo theft incidents, ahead of Italy at 13 per cent and the United Kingdom at 9 per cent (BSI Consulting and TT Club, "2025 Cargo Theft Report", p. 7, read 9 September 2026).
Four causes. Each is settled by a different photograph.
Cause one is settled by the loaded trailer, shot from the doors as the tiers go in. The image that ends this argument is not of a pallet, it is of a full trailer with a countable number of pallet faces. Sequence matters more than detail.
Cause two is settled by the label, legible, on the pallet and on a representative carton. This is the cheapest photograph in the whole set and it resolves the largest share of the disputes that turn out not to be shortages at all.
Cause three is settled by the completed pallet at the wrap station, taken so the top layer and the carton faces are readable against a known tie and high. If the pallet is short, this shows it, which is the point. Evidence that only ever exonerates you is not evidence, it is marketing.
Cause four is settled by the seal, with its number readable, photographed in place at the doors. An intact matching seal on arrival moves the argument off your dock entirely.
That is four images. Add a wide shot of the load before the doors close and you have five. Five covers the four causes without generating a library nobody can search.
The instinct to document everything is the enemy here. Photograph every step of every order and you produce a volume of images that cannot be retrieved under time pressure, which is functionally the same as having none. Two capture points, consistently, beats seven capture points, patchily. Partial coverage on the one shipment that gets claimed is worse than no policy at all, because it invites the question of why that load and not this one.
Capture is the easy half. It is also the half everyone optimises. Retrieval is where this fails.
The claim does not arrive on the day of shipment. It arrives weeks later, quoting an order number, and someone has to produce the evidence while it still changes the outcome. Photographs in a shared drive organised by date, or in a supervisor's camera roll, or in a WhatsApp group, technically exist. Under time pressure they do not. Teams in this position usually conclude that documentation did not work for them, when what failed was the index.
The requirement is narrow. Any image must be findable by the identifier the dispute will use, which is the order number, the delivery note number, the pallet ID or the seal number. If a person can paste the number from the claim email and see the load inside a minute, the record is doing its job.
This is the specific thing Blimp is built around. Capture happens on a shared device rather than a personal phone, with PIN access instead of individual accounts, because the alternative puts company evidence on hardware that leaves when the person does and creates a data protection question you do not want. Images are written offline and sync when the device finds signal, because dock doors and trailer interiors are where reception ends. Every photograph is attached to a shipment reference at the moment of capture rather than sorted afterwards, which is what makes the retrieval work and what lets customer service answer a claim without routing it through the warehouse. The output is a photo report against a shipment number, which is the format an insurer or a customer will accept.
Now the limit, stated plainly, because you should have it before anyone demos anything.
Blimp does not count for you. A photograph of a pallet is evidence that a person can verify. It is not an automated piece count, there is no computer vision doing carton recognition, and it will not tell you that a pallet is short before it ships. If your requirement is verified piece-level accuracy at the point of pack, that is a scanning and weighing problem, not a camera problem, and you should buy for that instead. What structured photo documentation does is make the state of the load at a specific moment reconstructable by a third party afterwards. Those are different jobs and only one of them is ours.
The honest reason to do it anyway is that the alternative in most operations is nothing. Not a worse system. Nothing, plus a shared drive.
Two things. Only one is the reason people buy.
The expected one is that disputes get shorter. You answer with an image instead of an assertion and the conversation ends.
The unexpected one is that your own error rate becomes visible. Some proportion of the claims you have been absorbing are real, and photographing outbound loads will prove it. Most manufacturers who start this find a mix, and the confirmed ones point at a specific pick or pack step that can be fixed. That is worth more over a year than the disputes you win, and it is the part no vendor puts on a landing page, because it sounds like an admission rather than a benefit.
Whatever your bill of lading says, subject to a statutory floor. Under 49 U.S.C. § 14706(e) a carrier may not allow less than nine months to file a claim or less than two years to sue after the claim is disallowed, and nine months is the usual figure. 49 CFR § 370.3 sets out what the claim must contain, and a communication that does not demand a specified or determinable amount is not a claim.
Under Carmack, the claimant. The prima facie case from *Elmore & Stahl* is good condition at origin, deficient condition at destination, and the amount of damages, and the first of those is the one that fails for want of evidence. Under CMR Article 30 the allocation runs the other way: a consignee who accepts without reservation faces prima facie evidence that the goods arrived as described. If your customer is deducting rather than claiming against a carrier, neither applies and your supply agreement governs.
Photographs carrying a reliable timestamp and an association to a specific shipment are routinely used in support of freight claims. What weakens them is ambiguity about when and where they were taken, which is precisely the weakness of images retrieved from a personal phone months afterwards. Capture that attaches the shipment reference at the moment the image is taken avoids that objection.
Five covers the four causes: the loaded trailer in sequence, the pallet label, a representative carton label, the completed pallet, and the seal with its number readable. Beyond that you are creating a retrieval problem rather than an evidence advantage.
No, and starting that way tends to fail. Begin with the customers who actually generate claims, or the lanes where they cluster, and hold that consistently for a quarter before widening. Consistency on a subset is defensible. Patchy coverage across everything is not.
Blimp lets you easily capture, organise and share photo documentation and professional photo reports. Available on mobile and web, Blimp simplifies reporting, reduces manual work and streamlines claims handling across industries.
Contact us
