Floor notes · warehouse office / sales group / finance month-end
We have walked many floors three months after launch. The icon is on the PC. The desk still has printouts and an open sheet. Sales says the system is slow, the customer is chasing in the group. The warehouse ships first and backfills later. The owner asks why nobody uses what they paid for. That is not “should we build a system.” It is whether the formal path was closed. If the group still takes orders, the system stays an elective.
Nobody-uses usually shows in three places on the floor
First, the entry. Real orders still live in the group, the phone, and a salesperson’s private sheet. The system opens when a manager visits. Pulling intake out of the group is why you build a system. See The risk of taking orders and shipping from WeChat groups. If the group stays open, more training is a performance.
Second, speed. If the system is slower than copying a sheet — extra fields, long approval, phone will not open — the floor votes with its feet. Using it as an electronic ledger you backfill later is not the same as walking the document at the work entry. See ledger vs. operations system. Third, ownership. Nobody can sign “this order must enter the system,” and nobody owns exceptions. Exceptions stay in the group. Rules live only in a manual.
Observation: nobody-uses is rarely “low staff quality.” More often: the old road is still open, the new road is slower, and nobody closed it.
Worth it is whether you will change the work — not login counts
If the company decides quote, outbound, and reconcile must follow the system, and WeChat is notice only, the system is worth building even if phase one is thin. Logins rise after the path is forced. If you “launch and see if people like it,” you go back to the sheet. When people leave, documents that never entered the system break with them. See Why operations do not fall apart when people leave.
The premise is specific: phase one covers only the main path; each role can finish in ten minutes; name an admin and an exception owner; close formal group intake within two weeks of launch. Talk cost and cycle on that scope, not a module pile. See What a custom business system costs and how long it takes. A fat unused feature set costs more than skipping a module.
How people will use it goes on the launch list, not into a slogan
Pilot with real documents, not a demo account. Warehouse outbound in the system that day. Sales changes price in the system that day. Finance exports a statement that week. Fix fields that will not walk — do not wait for “later optimization.” If the floor is still messy, diagnose in order. Do not start by blaming people. See The system is live and the floor is still messy: how to diagnose it. Visible efficiency is one fewer file match and one fewer mis-ship — not a video wall. See The efficiency a business system can make visible.
When DaXi builds a business system, which old road closes goes into acceptance — we do not only hand over accounts. To close it on your floor, go to Business systems service page and say how you take orders today. The full path is in From brief to launch: how we build a business system. Why sheets and WeChat do not last is in why a company cannot run operations on spreadsheets and WeChat alone.
Two stocks in a parallel run are messier than no system
The first week is the danger: some people start, some stay in the group, two stocks exist. That is messier than no system. Set parallel-run days. After that, only system documents count. The owner looks up documents in the system and stops accepting “I have a screenshot.” If scorecards still count shipped pieces and ignore whether a document entered the system, the floor will keep backfilling. Write “no system number, no ship, no payment request” as one rule. It beats three pep talks.