Reading on-time delivery without misleading averages
A single on-time percentage can hide chronic late shippers. Here is how Hong Kong procurement teams we work with split the measure before a supplier review.
Many supplier scorecards open with a single on-time delivery percentage. That figure is easy to present and just as easy to misread. A vendor who misses every shipment on one lane can still look acceptable if the rest of the lanes arrive early.
In recent Panel Nestcore engagements for Hong Kong traders, we split on-time delivery by destination lane and by order size before the quarterly review. The company-wide average stayed on the cover sheet; the lane table sat underneath so category owners could see chronic exceptions.
A working definition
Agree the clock before you calculate. Receipt date against the confirmed ship window is common for sea freight into Kwai Tsing; for air parcels, teams often use the promised delivery date on the booking confirmation. Write the definition into the scorecard legend so the next quarter does not reopen the debate.
Why averages hide late shippers
Suppose forty of fifty orders arrive on time. The 80% headline looks manageable. If the ten late orders all belong to the same forwarder on the Jakarta route, the problem is concentrated. Lane-level counts make that concentration visible without inventing new software — a pivot table and a short note often suffice.
What we put in the exception register
We list suppliers with three or more late receipts in the period, plus any single delay that held a production line or a customer shipment. Mild late arrivals that the buyer accepted in writing stay out of the register so the meeting stays on decisions, not noise.
If you are preparing a supplier performance reporting pack, start with the measure definition and the lane split. The rest of the scorecard becomes clearer once those two choices are settled.