The handover is where accountability is won or lost, because it is the point of custody change — and the last moment anyone can honestly say what condition a vehicle was in. Everything after that is inference. A finished vehicle passes through at least three custody boundaries between the end of the line and the dealer forecourt, and if condition is captured at only two of them there is a segment where damage cannot be attributed to anyone. That segment is where claims go to be argued and eventually split, which satisfies nobody. What makes this solvable rather than merely frustrating is that the industry already agreed a common language for it: a standard set of five-digit codes exists precisely so that every party can record transportation damage in the same format whenever possession transfers. The gap is usually not the standard. It is that nobody captured. Book a 30-minute session and bring one recent contested claim — we'll rebuild its custody chain in Fleet Rabbit and show you exactly which handover has no evidence behind it.
2026 GUIDE · CONDITION AT HANDOVER
Vehicle Damage Capture at Every Handover
Condition imaging at each custody transfer, damage mapped in the standard code format, and attribution between compound, carrier and dealer that survives the claim it was built for.
Compound
Captured
Carrier
No capture
Dealer
Damage found at the dealer. Two parties, one ungoverned segment, and no evidence that separates them. This is the claim that gets split — and neither side thinks the split was fair.
Why the Handover Moment Decides It
Four properties make the custody boundary the only place condition can be established, and all four are unforgiving of a missed capture.
It is the liability lineResponsibility transfers at that instant. A record taken an hour later documents a period in which two parties both had access, which is no record at all.
It is the only moment both parties are presentA joint observation is what makes a condition report an agreement rather than an assertion. After departure, one party is describing something the other cannot see.
Damage is cumulative and undatedA vehicle arriving with four marks tells you nothing about when each appeared. Only a sequence of captures separates them into segments.
Memory is not evidenceWeeks later, everyone involved is honestly recalling a vehicle they saw for ninety seconds among a hundred others. Nobody is lying and nobody is reliable.
The rule that follows from all four
Capture at every boundary, not at the ones where disputes have historically arisen. A chain with one ungoverned segment does not have one weak link — it has an unattributable zone that absorbs every unexplained mark in the whole journey, because it is the only place damage can have occurred without contradiction.
The Standard Code Format
Damage should be recorded in the agreed industry format rather than in free text, because the format is what allows a claim to move between systems and parties without translation.
Area
Where on the vehicle — the panel, component or zone affected
Type
What kind of damage — scratch, dent, chip, crack, missing part
Severity
How significant, on a defined scale rather than an adjective
A set of five-digit codes exists so that all involved parties can use a standard format to note transportation damage whenever possession is transferred from one party to another — recorded on delivery receipts or on handheld terminals for transfer into manufacturers' claims systems. The current guideline sits in its sixth edition and has been jointly developed on both sides of the Atlantic, so the same code means the same thing to a compound in Michigan and a carrier in Bavaria.
At origin
At receipt
Attribution is a comparison, and you need both frames to make it.
Bring one contested claim and the condition records either side of it to a short call. We'll line up the captures in Fleet Rabbit, run the comparison, and show you whether the evidence attributes the damage or merely documents it. Most teams discover the problem is a missing capture rather than a disputed one.
What a Capture Must Contain
Seven elements. A capture missing any of them still produces a file; it does not produce evidence.
1The VIN, scanned from the vehicleTied to the record at the moment of capture rather than entered afterwards. A condition report on the wrong VIN is worse than none, because it carries authority it has not earned.
2Time and location, recorded automaticallyWhen and where the transfer happened, from the device rather than from a handwritten field. This is what places the record inside a specific custody segment.
3The parties involved, both namedReleasing party and receiving party, with the carrier identified. An unsigned transfer is a transfer nobody accepted.
4Full-vehicle imaging at consistent anglesEvery panel covered, the same way each time. Consistency is what makes two captures comparable — inconsistent angles produce two records that cannot be placed against each other.
5Damage coded and photographed togetherBoth, always. A code without an image cannot be verified and an image without a code cannot be processed — and systems that let either be submitted alone reliably produce both.
6Specific descriptions, not adjectivesLocation, length and depth. "Minor scratch" means a light surface mark to one reader and a deep gouge to another, and ambiguity favours whoever is defending.
7Operational status and completenessOperable or not, keys present, missing parts or options noted. Damage is not only cosmetic, and a missing key surfaces at the worst possible moment.
How Attribution Actually Works
The mechanism is simple and it depends entirely on discipline rather than sophistication.
Step 1
Capture at the close of a custody segmentThe releasing party's final record. This is the "as handed over" baseline for that segment.
Step 2
Capture at the open of the nextThe receiving party's first record, taken before the vehicle moves under their control. Same angles, same standard.
Step 3
Compare the two and flag the deltaAnything present in the second and absent from the first occurred inside that segment. Automated comparison can do this reliably — the same pattern is used to attribute new damage to a specific shift in fleet operations.
Step 4
Attribute to the segment, not the partyThe record says when the damage appeared. Who bears it follows from the custody terms for that leg — which is a contractual question, and a far easier one to settle once the timing is not in dispute.
Why the comparison beats the inspection
A single inspection tells you a vehicle has damage. A pair of captures tells you when it appeared. Every argument about whose fault it is comes down to the second question, and no amount of skill at the first will answer it. This is why capture frequency matters more than capture depth — two adequate records at the right moments beat one excellent record at the wrong one.
Where Capture Quality Fails
Six failure modes, all recoverable, and five of them are fixed by the capture process rather than by better equipment.
← Swipe to see all columns →
| Failure | What it produces | The fix |
|---|---|---|
| Blurred or badly lit images | A file that exists and proves nothing when examined | Real-time image quality checks that reject the capture before it is accepted |
| Inconsistent angles | Two records that cannot be compared to each other | Guided capture prompting the same sequence every time |
| Codes without photographs | An unverifiable claim of damage | Make imagery mandatory rather than configurable |
| Photographs without codes | Evidence that cannot enter a claims system | Require the coded entry alongside the image |
| Partial vehicle coverage | Undetectable damage on the panels nobody photographed | Full 360-degree capture as the standard, not on suspicion |
| Dark or textured paint | Micro-damage genuinely missed even by good systems | Sufficient resolution and adequate lighting — the one that needs equipment |
The two configurable failures deserve attention. Where a system allows damage codes to be submitted without photographs, or photographs without codes, both will happen under time pressure — and each produces a record that fails at a different stage. Making both mandatory removes a decision from a driver who is already working against a departure slot.
Capture Methods
Four approaches, and the right answer usually varies by boundary rather than being uniform across the chain.
Guided mobile capture
Prompted sequences on a handset, with real-time checks on image quality and completeness before the record is accepted. Works at any boundary including a kerbside dealer handover, and needs no fixed infrastructure.
Best atCarrier pickup and dealer delivery, where the boundary moves
Drive-through imaging
Fixed stations capturing large numbers of high-resolution images per vehicle at speed, with high-volume throughput. Consistency is perfect because the angles are structural rather than human.
Best atCompound gates and plant exits, where volume concentrates
Automated comparison
Machine comparison of successive captures, flagging new damage against the previous record. Reported accuracy is high where capture conditions are good, and ensemble approaches reduce false positives by requiring agreement between models.
Best atThe attribution step itself, across every boundary
Structured manual inspection
A person, a defined sequence and coded entry. Still entirely valid, and considerably better than an unstructured automated capture — the standard was designed for it and manufacturers' claims systems accept it.
Best atLow-volume boundaries and as the fallback everywhere
Twenty minutes on your custody chain and last quarter's claims
On a working session we'll map every handover in your outbound chain, mark which boundaries have capture and which do not, and run your recent claims against that map in Fleet Rabbit — showing how many were lost or split because of a missing record rather than a genuine disagreement. You keep the chain map either way, and it is usually the shortest route to a business case.
What to Measure
Six figures. The first is the one that determines whether any of the others mean anything.
Boundary capture coverageShare of custody transfers with a compliant capture on both sides. Anything below full coverage leaves an unattributable segment somewhere in the chain.
Capture completenessRecords holding all seven elements — VIN, timestamp, parties, full imaging, codes with photos, specific descriptions and status.
Image quality rejection rateCaptures rejected and retaken at the point of capture. A healthy number rather than a worrying one, since the alternative is accepting them.
Damage attributed by segmentNew damage assigned to compound, transit or delivery. The output the whole programme exists to produce.
Unattributable damage rateDamage that cannot be placed in a segment. Should fall toward zero as coverage rises, and it is the cleanest proof the programme is working.
Claims resolved on evidenceSettled by the record rather than split by negotiation. Converts the whole exercise into a number finance recognises.
The direction of travel
Digital vehicle handover is an active industry priority rather than an optional modernisation. The damage code guideline now assumes electronic transmission, consignment note digitisation is progressing across Europe, and VIN-level digital traceability is being required by adjacent regulation. Operators who already hold a digital evidence chain are positioned for each of these; those still working on paper will be building the same capability under a deadline rather than on their own schedule.
Frequently Asked Questions
Why capture at every handover rather than just at delivery?
Because a single inspection tells you a vehicle has damage, while a pair of captures tells you when it appeared — and every attribution argument turns on the second question. A chain with one ungoverned segment does not merely have a weak link; that segment becomes the default explanation for every unexplained mark in the entire journey, because it is the only place damage could have occurred without being contradicted by a record.
What format should damage be recorded in?
The standard five-digit code format, which identifies area, type and severity. It exists precisely so that every party can note transportation damage in a common format whenever possession transfers between parties, and it is designed to be recorded on delivery receipts or handheld terminals for transfer into manufacturers' claims systems. Because it was developed jointly across regions, the same code carries the same meaning to a compound, a carrier and an inspection company anywhere in the network.
Do we need photographs as well as codes?
Both, always, and it is worth removing the option to submit one without the other. A code without an image cannot be verified when challenged; an image without a code cannot be processed by a claims system. Where a platform allows either to be submitted alone, both situations will occur regularly under time pressure — so making imagery and coded entry jointly mandatory takes the decision away from someone already working against a departure slot.
How does automated comparison assign fault?
It does not assign fault — it assigns timing, which is the harder part. Capture at the close of one custody segment and at the open of the next, then compare: anything present in the second record and absent from the first occurred inside that segment. The same approach is used in fleet operations to attribute new damage to a specific shift. Who bears the cost then follows from the custody terms for that leg, which is a contractual question and much easier to settle once the timing is established.
Why do good photographs still lose claims?
Usually because of inconsistency rather than quality. Two captures taken from different angles cannot be placed against each other, so neither can establish what changed. Guided capture that prompts the same sequence every time solves this, alongside real-time checks that reject blurred or badly lit images before they enter the record. And describe damage specifically — location, length and depth — because vague adjectives are read differently by whoever is defending.
Can automated systems detect everything?
Not everything, and it is worth knowing the limits. Reported detection accuracy is high where capture conditions are good, with ensemble approaches reducing false positives by accepting only findings that multiple models agree on. But micro-damage requires full vehicle capture at sufficient resolution and adequate lighting, and dark or heavily textured paint surfaces remain genuinely harder. Automation raises consistency substantially; it does not remove the need for good capture conditions.
Where should we start?
By mapping your custody chain and marking which boundaries currently have a compliant capture on both sides. Almost every chain has one that does not, and that boundary is generating a disproportionate share of your unattributable damage. Fix coverage before capture sophistication — two adequate records at the right moments beat one excellent record at the wrong one. Book a working session with last quarter's claims and we'll build the chain map with you.
Two Records, One Segment, No Argument
Capture on both sides of every custody boundary, damage coded and photographed together, descriptions specific enough to compare, and attribution resolved to a segment before anyone discusses who pays for it.
Bring one contested claim to the call · Works with standard damage codes and existing handhelds
August 24, 2026By Sam Parker
All Blogs
