An automotive parts importer we worked with last year had a replenishment problem that looked like a forecasting problem. Their German supplier's quoted lead time was 21 days, but actual receipt averaged 34 days over the previous six months. Their domestic supplier quoted 5 days and delivered in 4.8. The planning team had adjusted safety stock for the German lane manually, but nobody had visualised the drift over time.
Why averages hide the pattern
A single average lead time per supplier is useful for purchase order scheduling but useless for spotting drift. Lead times creep upward in small increments — a day here, three days there — often masked by occasional fast shipments that pull the average back down. By the time a planner notices, safety stock has been wrong for two quarters.
The comparison panel we built
We added a lead-time comparison panel to their inventory visibility dashboard with three layers:
- Quoted lead time: The value stored in the supplier master record
- Rolling 90-day actual: Average days from PO creation to goods receipt, recalculated weekly
- Variance band: Highlighted when actual exceeds quoted by more than 20%
Domestic and overseas lanes are shown side by side, grouped by supplier country. The German lane's variance band turned red in September, two months before the planner would have noticed from the exception queue alone.
Connecting drift to replenishment
The panel links to the inbound PO tracker: clicking a flagged supplier lane shows all open POs for that lane with expected vs. actual receipt dates. This closed the loop between "lead times are drifting" and "these specific POs are late."
The procurement director used the panel in a renegotiation call with the German supplier in January. Whether the supplier improves remains to be seen, but the conversation started with shared data instead of anecdotal complaints.
Data requirements
This panel needs PO creation dates and goods receipt dates from your ERP. Most SAP and local Korean ERP extracts include both fields, though field names vary. During a mapping workshop we confirm the exact column names and test against six months of historical data before building the visual layer.
Takeaway
If your replenishment team adjusts safety stock by gut feel for overseas lanes, you probably need a drift visualisation before you need a better forecast. The data is usually already in your ERP extracts — it just is not on anyone's screen.