The spreadsheet after the ERP
Walk most packaging yards and you will still find the spreadsheet after the ERP.
The ERP holds the orders. The sheet holds the truth the dock actually uses. Who is on which trailer. Which drops share a run. What got bumped when production slipped an hour. That sheet is where the day gets decided.
Why the sheet survives
It is fast. It is flexible. Anyone can edit it. When a customer moves a window or a line finishes late, someone opens the file and shuffles rows. The ERP does not feel that change until much later, if at all.
I do not blame planners for this. They are keeping trucks moving with the tools they have. The problem is that the sheet becomes a second system of record. Then Fabric, WMS, and the load plan are all chasing a moving target.
Version control is a polite word for what happens next. One planner has the morning file. Another has a copy on a desktop. The shift handover is a conversation about which columns still matter.
What the sheet is really doing
It is clustering stops by gut feel. It is guessing fill. It is hoping axle weight works out at the weighbridge. It is carrying multi-drop sequence in a column that only one person understands.
That works until volume rises or SKUs get denser. Then the same sheet that saved the morning creates the afternoon scramble.
I have seen colour codes do the work of a proper constraint engine. Green means maybe. Red means move it. None of that survives holiday cover cleanly.
Keep the ERP. Replace the shadow plan.
I am not arguing for ripping out finance or stock systems. Keep the ERP. Clockwork sits after it. Move order and ship data once via Fabric. Build the load plan in a place that understands cube, weight, axles, and drop order.
The goal is simple. One plan the yard can trust, instead of a spreadsheet that only the planner on shift can read. When that shadow plan goes, the dock stops living in a private language.
You will still need judgement on the day. Judgement works better when the baseline plan is already axle-legal and sequenced, not when the baseline is a grid of cells.