Buy a carrier, inherit an ELD vendor.
Do that three times over five years and you have four portals, four logins, four definitions of a duty status, and four answers to a question that should only have one: how long can this driver still legally work today.
Nobody chose this. It arrived one acquisition at a time. But the effect on the operation is real and it compounds. Dispatch plans against whichever portal is already open. Safety reconciles by exporting to a spreadsheet on Monday. Compliance finds out about a malfunction when the repair clock is most of the way expired. And every one of those people believes they are looking at the truth, because each portal is internally consistent.
Four systems that are each correct and collectively useless
The trap is that no single vendor is wrong. Each one records what its devices saw, applies its own rules, and shows you a clean screen.
The disagreements live in the seams:
- One vendor counts yard moves as on-duty driving. Another does not.
- One auto-assigns short segments to the last known driver. Another leaves them unassigned and waits.
- Personal conveyance is configured differently in each, usually by whoever set it up, usually years ago.
- Two of them round to the nearest minute in opposite directions.
Individually these are footnotes. Together they mean the remaining hours you plan against depend on which fleet the truck came from. That is not a compliance risk in the abstract. That is a dispatcher committing a driver to a delivery the clock will not support, and finding out at the shipper.
Unassigned driving is two problems, not one
Every mixed fleet has an unassigned driving pile, and almost everyone treats it as a single chore to work down.
It is really two very different things wearing the same label.
Most of it is yard movement. A mechanic repositioned a tractor. A hostler moved a trailer forty feet. Somebody idled while a door opened. This is noise. It has no bearing on anybody's hours, and the only correct handling is to recognize it and stop showing it to a human.
The rest is road movement, and it matters a great deal. Somebody drove, somebody has to claim it, and until they do your hours-of-service picture has a hole in it that an auditor will find faster than you will.
Treating those two as one pile is why the pile never goes down. The yard noise buries the real segments, people stop working the list because it is mostly junk, and the segments that needed a driver's attention age out in the middle of it.
Coro, the ELYON ELD aggregator, separates them rather than treating them as one problem, routes the genuine road segments to the driver to claim, and writes the trail as it goes. Splitting the pile is most of the work. Once it is split, the human list is short enough that people actually work it.
The malfunction nobody read
Every ELD vendor reports malfunctions. Almost every operation finds out about them late.
The reason is mundane. The malfunction appears in a portal that somebody checks weekly, if that, and it looks like one more row in a list of rows. Meanwhile the regulatory clock for repairing it started the day the device reported it, not the day a human noticed.
So the compliance question is never "did the device fail." Devices fail. The question is whether the clock started when the device said so or when a person happened to look. Reading vendor malfunction reports on the day they appear, and starting the clock then, is the difference between a documented repair and a finding.
One clock, published once
The fix is not another portal. It is refusing to let there be four answers.
Normalize every vendor's duty statuses into one model with one meaning. Compute remaining hours once. Publish that single answer to everything that needs it: dispatch, safety, planning, the customer promise. When a new vendor arrives with the next acquisition, absorb it into the same model rather than adding a fifth login to somebody's morning.
That last point is the one operators underrate. The cost of a mixed fleet is not the four systems you have. It is that the fifth one is coming, and today your integration plan for it is "a person will check it too."
Reconciling against the underlying engine events is what makes this more than averaging. When two vendors disagree, you do not want a compromise. You want the answer the engine supports, and a record of why the other one was rejected.
What an audit actually asks for
The compliance argument for one clock is usually made in the abstract. It is more persuasive when you look at what a review actually requests.
It is rarely "show me your hours." It is narrower and more awkward than that. Show the unassigned driving for this period and how each segment was resolved. Show when this malfunction was reported and what you did about it, with dates. Show why this driver's record was edited, who edited it, and whether the driver agreed to the change.
Every one of those questions is about the trail, not the number. And in a mixed fleet the trail is the weakest part, because it is spread across systems that were never designed to be read together. You end up assembling the answer by hand, from four exports, under time pressure, about events from eight months ago.
The version of this that goes well is boring. The segments were split and resolved as they occurred. The malfunction clock started the day the device reported it. Edits carry the reason and the acknowledgement. You are producing a record rather than reconstructing one, and reconstruction is where both the cost and the exposure actually live.
The clock is only half the day
Hours are the constraint everybody tracks, because the penalty is obvious. The rest of what grounds a truck is just as real and gets far less attention.
A credential expired. A fault code has been open for six weeks. A safety flag came back from a roadside inspection nobody logged. A trailer is somewhere it should not be, and the only person who knows is on vacation.
Flotta puts everything that needs a decision today into one list, in one order, worst first: hours, credentials, faults, safety, trailers and fuel. The value is not the data, which you already have scattered across systems. The value is the single ordering. A maintenance manager with one ranked list makes better calls than the same person with six tabs and a sense that something is probably overdue.
The equipment itself has the same blind spot. Falco tracks every owned thing that carries a tag: where it is, whether the tag is alive, and who has it, including the ones it cannot see. That last clause is the one that earns its keep. A tag that has gone quiet is not the same as an asset that is fine, and a system that only reports what it can currently see will quietly lose you trailers.
What good looks like
You can test this without buying anything. Ask three people in three roles the same question about the same driver, right now: how many hours does this driver have left today.
If you get three answers, or one answer and two shrugs, the problem is not discipline and it is not headcount. It is that you are running four sources of truth and asking humans to reconcile them in their heads, several times a day, under time pressure.
One clock, computed once, published everywhere. Everything downstream, from a realistic delivery promise to a clean audit, depends on it.
If the promises made on top of that clock are the recurring argument with customers, that is the same problem seen from the freight side: supply chain visibility that acts before the phone rings. And if your trucks are burning their remaining hours waiting at a door, the appointment is the constraint, not the clock: dock scheduling that rebooks itself when the ETA moves.