Let me be blunt: the first week after a WMS go-live is not won by the system. It is won by people. Getting the system running on cutover day is the easy part. The hard part is the seven days after, when something new breaks every day and somebody has to take the call, make the judgment, and own the decision. This WMS go-live war room checklist for the first week is the sheet for those seven days: who sits in the room, which screens to watch, what counts as a fire alarm, and where the rollback line is drawn. Without it, the first week is a battle with no commander.
There are two sides to this. Watch too closely and you strangle the floor — pickers wait for confirmation on every step and throughput drops below what it was before go-live. Watch too loosely and missed orders surface only when customers start calling, and fixing that costs ten times what watching would have. I have seen too many go-lives die in the first week, and they rarely die on technology. They die because nobody made a decision.
In the fall of 2023, a client warehouse in the Midwest — auto parts distribution — cut over at midnight. At 2 a.m. the wave froze. Pickers stood at the racks waiting for tasks while every order in the system showed "allocated" and nobody could print a pick ticket. The two people on duty looked at each other, and neither would touch anything — touch the wrong thing and whose fault is it? It took until 6 a.m. to find the cause: one switch in the wave release rules had never been turned on. A five-minute fix cost four hours. That morning I told the client: what you are missing is not technology. It is one person willing to make a call at 2 a.m.
Another warehouse spent its first week chasing dozens of new inventory variances every day. On day three they found it: a batch of stock in the old system had never been cleared from its old locations, so the new system looked for it in the new locations while it sat in the old ones. The variances were small. The frightening part was that nobody knew where they came from, and each day's new variances piled onto the old ones until the list was unreadable.
Who Sits in the War Room: Four Chairs, No More
The war room is not a meeting room. It is the command post for the first week, and it holds exactly four chairs. The first is the commander — one person only, usually the warehouse manager or the project lead. He is the only one with authority to decide, and once he decides, nobody re-approves it. The second is the systems person, from the WMS implementer, who knows the configuration and the interfaces and can get into the system the moment something breaks. The third is operations, a floor supervisor who knows the people and the floor — which shift is short-handed today, which line jams first. The fourth is the customer voice, someone from customer service or the client interface, so that anything going out to customers comes from one mouth, not from whoever on the floor picks up the phone.
Beyond the four chairs, you need a duty roster — nights included. Systems do not wait for daylight to break. When a wave freezes at 2 a.m., a roster with no name on it means nobody owns it. The person on duty carries an escalation card: what to handle alone, what wakes the supervisor, what wakes the commander. Write it down. Do not leave it to judgment at 2 a.m.
The Daily Rhythm: What the War Room Checklist Watches in Week One
Five screens hang on the war room wall, and missing any one of them leaves a blind spot. Order release: how many orders came in today, how many released, where the rest are stuck. Wave execution: how many waves released, how far picking has gotten, whether any wave is running late. Inventory variance: how many new variances today, how many have a known cause, how many are still unexplained. Shipping and dock: how much went out, how many trucks left, whether any order missed its cutoff still sitting in the building. System and interfaces: how many interface errors, whether the task queue is backing up, whether any handheld terminals have dropped offline.
The rhythm for the first week runs on this schedule:
| When | What happens | Who is there |
|---|---|---|
| 15 minutes before start of shift | Morning huddle: last night's issues, today's priority orders, who is on duty | All four chairs |
| 10 a.m. | Check order release and wave progress; anything stuck gets a decision on the spot | Commander, systems, operations |
| 3 p.m. | Check inventory variances: how many new, what category each falls into, who is chasing them | Systems, operations |
| Before close | Evening review: did shipping complete, why not, what to watch tomorrow | All four chairs |
| Overnight | Duty person watches the system and interface screens, escalates the moment a trigger fires | One person on duty |
The schedule looks clumsy. It has exactly one purpose: every problem closes the same day it appears. A problem that sleeps overnight becomes two problems by morning.
What Counts as a Fire Alarm: Triggers on the Go-Live Checklist
A fire alarm cannot be a feeling. It has to be written on the wall so that anyone who sees it makes the same call. The lines we set were these: an order unreleased for more than 30 minutes; more than 20 new inventory variances in a day with no known cause; wave execution running more than an hour behind plan; the same interface error three times in a row; the same customer complaint twice. Hit any one of them and the person on duty escalates immediately — no waiting for the morning huddle, no "let's watch it for a while."
The escalation path is written down too: duty reports to the supervisor, the supervisor has 15 minutes to decide or it goes to the commander, and the commander decides whether to stop the line or roll back. Every level has a name on the roster, not "relevant personnel" — a phrase that helps nobody find anyone at 2 a.m.
Where the Rollback Line Is: The Bottom Line for Week One
A rollback is not a failure. It is a bottom line, and it has to be drawn before go-live, not while the building is on fire. The lines we drew were three: the core process cannot be restored for two straight hours; inventory variances are so large that the day cannot be reconciled; customer orders are delayed at scale, with freight still sitting in the building past cutoff. Hit any one of them and the commander calls the rollback. No committee meeting first.
The rollback itself has to be rehearsed once before go-live: can the old system take over, how does the data come back, who calls the customers — walk it through in the rehearsal. A rollback that has never been rehearsed is more chaotic than the failure it is meant to fix. One more thing: only the commander can call it. Anyone on the floor can suggest it, but the decision belongs to one person. Otherwise you are back to two people looking at each other at 2 a.m.
In the first week after go-live, the most valuable thing in the war room is not the big screens. It is the person willing to make the call. That is the whole point of this WMS go-live war room checklist for the first week: it exists so the person making the call never has to improvise. The systems, the checklists, and the rosters are just tools built for that person.
By Lao Mi, Supply Chain Veteran (20 yrs)







