.svg)
.svg)
.svg)
.svg)
.svg)
“The driver will call you about half an hour before we arrive.” It sits in booking confirmations and driver inductions across last-mile and contracted delivery alike, and it is good service when it happens. Thirty minutes is enough time to come in from the garden, end a call, move a car, or clear a dock bay.
It is also free to promise. No system has to exist for it, no integration has to be built, nobody has to configure anything. A planner can commit to it on Monday and the operation can start delivering against it on Tuesday.
But it is a promise made at a desk and kept in traffic, and the gap between those two places is where it comes apart.
The person who knows when you will arrive is the person who cannot tell you
A driver thirty minutes from a stop is doing one of a few things. Driving. Parking. Carrying something heavy up a stairwell. Standing at a dock waiting for a signature. None of those leaves a comfortable window to look up the next stop, work out a realistic arrival time, place a call, and negotiate whether that time still works for whoever is on the other end.
So, the call gets made when a gap appears. Gaps appear after a stop is finished, which is usually well past the promised window, or at a red light, which is not where anyone wants to manage an appointment.
The driver is the only person in the operation who really knows where the vehicle is, and the same fact that makes them best informed makes them least available to act on it.
This is not a discipline problem. Treating it as one is how operations end up covering the thirty-minute call in every driver induction and seeing no change in how many calls land on time.
A phone call can only ever reach one person
Even a perfectly timed driver call solves for exactly one relationship: the driver and whoever is standing at the door. That was fine when the obligation was that simple.
It rarely is, once a stop sits inside a contract. The shipper wants to know their freight is tracking to plan. The consignee wants to know before the truck is in the yard, not after. Dispatch needs to know the moment a delay changes the five stops scheduled behind it. Whoever owns the account on your side wants to know before the client does.
Example
A regional carrier running dedicated routes for three retail accounts described the pattern this way: one delay, one call, and the consignee, dispatch, and the account manager each found out at a different time, from a different person, telling a slightly different story about why.
A single call was never going to close that loop, because it was never built to. It was built for a driver and a doorstep, and most of your obligations now involve neither.
Your contract has two clocks. Most systems are only watching one.
A delivery SLA sets a tolerance for the truck: how late is still acceptable before it counts against the account. A separate clause, increasingly its own line item in carrier and 3PL agreements, sets a tolerance for the notification itself: how much warning the counterpart is owed before a deviation counts as undisclosed rather than managed.
Two clocks, one system
- Delivery SLA: when the delivery must arrive
- Notification SLA: how much notice time the counterpart is owed before the delay counts as undisclosed
An operation can breach the first and still make the second.
Those two clocks do not run together. On paper, a missed notification looks like a documentation gap. In a dispute, it is the finding, because the freight arrived roughly on time and the account was still charged for the miss.
Notifying someone about a delay means knowing about it before it happens
This is where most communication setups fall short of what they appear to offer. A notification that fires when a shipment is dispatched, a route starts, or a delivery completes is reporting something that already happened. Easy, because the event is a fact by the time the message goes out.
Telling a receiving dock at 09:00 that a 10 to 12 window has become 12:15 is a different kind of statement. Nothing has happened yet. The truck is not late. The claim is about the future, and it can only be made by an operation recalculating arrival times continuously against live traffic and current vehicle position, rather than against the plan optimized the night before.
An operation that knows where its vehicles are can report status after the fact. An operation that knows when they will arrive, and keeps knowing as conditions change, can notify every party while there is still time for them to act on it.
Wherever a contract sets a tolerance, the same dependency shows up
Where this bites hardest
- Subcontracted carrier networks, where a shipper's SLA sits on top of whatever the carrier's own systems can see
- Retail replenishment windows, where a dock staffed for on time absorbs a late arrival as shelf time, not customer patience
- Cross-dock and consolidation points, where one missed window reshuffles every outbound route scheduled behind it
- Direct-customer and direct-store delivery, same mechanism at smaller scale and shorter windows
The question is not whether somebody should call ahead. Sometimes it is exactly right, and no automated message replaces a call on a delivery that has already gone wrong.
The bottom line
The question is whether your notification obligation can survive depending on one person, one channel, and one phone call at a time, across every party in your SLAs. If it can't, the exposure is not reputational. It is contractual, and it is sitting in your next QBR whether anyone has flagged it yet or not.
