A carrier's on-time dashboard can make a perfectly good carrier look like a disaster. The number says a big share of packages are running late, which is the kind of number that gets a carrier shut off the same afternoon.
Often it is a measurement gap, not a delivery problem.
Two clocks, two very different stories
Every "on-time" number is really a stopwatch, and the result depends entirely on when you press start.
- Label creation. The moment you print the label. This is the earliest possible timestamp, and it is the one a lot of systems default to because it is the first thing they see.
- Induction. The moment the carrier actually takes physical possession of the package, scans it, and the transit clock genuinely begins.
The gap between those two events is dwell time sitting in your own building: orders printed in a batch, staged, and picked up hours or sometimes a day later. None of that is the carrier's transit. But if your clock starts at label creation, all of that dwell gets charged to the carrier's SLA.
Start the clock at induction instead and the same carrier can look completely healthy. A meaningful share of the "late" cohort has often already delivered by the time anyone flags it. The packages were never the problem; the start line was.
Why this matters before you cut a carrier
The cost of the wrong clock is not just a bad chart. It is a decision. It is easy to be one report away from cutting a carrier whose pricing is strong and whose delivery is, on the honest measurement, excellent.
A few things worth doing before you act on an on-time number:
- Ask what timestamp the SLA starts on. If the answer is "label creation" or nobody knows, the number is suspect until you confirm it.
- Separate your dwell from their transit. Label-to-induction time is yours to fix. Induction-to-delivery is the carrier's. Don't blame one party for the other's clock.
- Check what's already delivered. A late-looking cohort that has mostly arrived is a reporting artifact, not a service failure.
This is the same trap as building rate-shopping rules on estimated delivery dates that drift: the data looks authoritative, and the gap only shows up when it changes a decision. Before you fire a carrier, make sure you're timing the right race.