Blog · Board and management reporting · Shipping and logistics
How a forwarder or carrier measures the reliability of its transit times per customer per lane from milestone data: quoted transit against actual door-to-door, the share of shipments within tolerance, the variability that matters more to a shipper than the mean, and the customer lanes whose reliability fell before the tender did.
A forwarder's operations team measures on-time performance by lane. Its sales team learns about a customer's frustration at the tender. The milestone data, cut by customer and lane, shows the customer's own experience of the lane, which is the only one that matters to them. This guide sets out reliability and variability per customer lane and the trend that precedes a re-tender.
Per customer, per lane, per quarter:
Reliability = shipments delivered within tolerance of quoted transit ÷ shipments Variability = 75th percentile actual transit − 25th percentile Delay split = share of late days at origin, in transit, at destination
Customer identifiers only.
every delivered shipment has a delivery milestone dated after its departure milestone
A shipment with delivery before departure, or no delivery milestone though invoiced, fails it and is listed; milestone data is the least reliable export in logistics and the check is what makes the measure usable.
One customer, four lanes, this quarter against the trailing four.
| Lane | Shipments | Reliability now | Trailing 4 qtrs | Variability now | Delay split |
|---|---|---|---|---|---|
| Asia to N. Europe | 84 | 74% | 92% | 11 days | 60% destination dwell |
| Asia to US West | 61 | 90% | 89% | 4 days | |
| Europe to US East | 22 | 95% | 94% | 3 days | |
| Intra-Europe road | 140 | 97% | 96% | 1 day |
One lane has dropped eighteen points and its variability has tripled, and the milestones say the delay is at destination. That is an operations conversation about one port and a customer conversation before the customer starts one.
Reliability across the customer's lanes, weighted by shipments, and the share of their volume on lanes that dropped this quarter. A customer with 40 percent of its volume on a deteriorating lane is on the retention list whatever the average says.
Lane-level only. The customer's experience of the lane is its own shipments, not the lane's average.
Mean transit reported. The shipper plans around the spread.
Milestone gaps ignored. Half the shipments have no delivery milestone and reliability is computed on the half that do. Fail the check; show the coverage.
Tolerance varies by shipment. One rule per lane, stated.
Mapped once, the shipment and milestone exports produce reliability, variability, the delay split and the trend per customer lane every quarter. Covirage builds this from the exports as they are. The shipping and logistics page describes the setup, and the trade lane share guide covers the share measure that reliability protects.
At minimum, the quoted transit at booking and the actual delivery date. Departure and arrival milestones add where the delay occurred, origin dwell, on-water or in-air, destination dwell, and that split is what the operations conversation needs.
A lane that averages 28 days with shipments between 24 and 32 is plannable. One that averages 28 with shipments between 20 and 41 is not, and the shipper carries safety stock for the 41. The interquartile spread per customer lane is the figure the shipper feels.
The shipper's own, where the contract states it; otherwise a stated number of days by mode, two for air and three to five for ocean. On the report, the same for every shipment on the lane.