Four regions, eighteen cycles.
A public rebuild of the KPI dashboard I ran for a multi-region supply chain in 2022–23. Same four measures, same regions, same monthly cadence. The vendors, orders and figures are generated, and tuned so the outcomes land where the record says they did.
One panel per region, one scale for all four.
Nobody has to ask which line is Brazil. Each region sits on its own panel at the same scale, so a lagging region is visible from across the room.
On-time delivery by region
Faster, and staying faster.
Turnaround is the measure that moved the most and the one the root-cause work was aimed at. Backlog age is the measure that tells you whether the gain is real or borrowed.
Turnaround, order to delivery
Average days, weighted by orders. Target 10 days.
Open backlog by age
Why shipments were late.
Eighteen cycles of late-shipment reasons, coded and counted. The point of the chart is the distance between the two dots, cause by cause. The split is modelled, not measured.
Late shipments per 100 orders, by cause
Accountability, one row per vendor.
Sort any column. Select a row to filter the whole exhibit to that vendor. Status is on-time delivery against the 95% target: on target, watch below 95, breached below 90.
Vendor scorecard
| Status |
|---|
Six work streams, one variance column.
Budget tracking ran alongside the shipment work. The column that mattered was variance against plan to date, because it surfaced a cost problem before it became a timeline problem. These six streams and their budgets are illustrative.
Capital-import work streams
Illustrative work streams with synthetic budgets, USD thousands. Variance is spent minus planned to date; over $25K either way is flagged.
| Work stream | Budget | Spent | Variance | Complete | Status |
|---|