A county wholesaler: month-end from a week to an afternoon
A sixty-year family wholesaler was closing the month in a week — three accounts clerks, two warehouse supervisors, and a reconciliation ritual that nobody enjoyed. The ledger now closes itself.
The reading
Three generations in the business. Two depots, four thousand lines of stock, a counter that takes orders by phone, in person, and by email, and a warehouse that fulfills them from a goods-in sheet printed each morning from a system that has not matched the counter’s view of the world since 2016.
The month closed in a week. Three accounts clerks, two warehouse supervisors, a shared frustration. The closing ritual had become so familiar that it was treated as a feature of the calendar rather than as a cost. The partners knew the figure was wrong but did not know by how much.
The headline we gave the partners was: the fix is almost always to stop writing it down twice.
Four truths, no truth
The counter had its view. The warehouse had its view. The accounts system had a third. The partners had a fourth, assembled from the first three with reconciliations held together by one clerk’s domain knowledge, which was not scalable and which had been the reason the firm had not promoted her out of the role for eleven years.
Where the four disagreed:
- Counter vs warehouse: sold stock the warehouse had already dispatched to another branch. Resolved by hand, usually the next day, sometimes a week later.
- Warehouse vs accounts: goods issued without a matching invoice line because the counter wrote the ticket on paper and typed it in later.
- Accounts vs partners: VAT and intra-firm transfers posted to different periods depending on which clerk entered them.
The month-end was a week of resolving these.
What we built
One ledger. The counter sale is the warehouse instruction is the accounts entry. There is no second system to update, no printed sheet to reconcile against, no end-of-day batch to reconcile across. The sale is posted to a single record, the warehouse pulls from that record, the accounts posting is a side-effect of the sale rather than a separate exercise.
The auditor arrives and finds the entry pre-placed.
Six weeks of build. Two depots migrated in a weekend each. The old accounts system was kept running read-only for three months as a reference, then retired.
The numbers at the first month-end
| What | Before | After |
|---|---|---|
| Month-end close | 1 week | 1 afternoon |
| Warehouse supervisor headcount | 2 | 0 |
| Entries requiring manual reconciliation | ~ 340 / month | 0 |
| VAT return preparation time | 3 days | 2 hours |
| Partners’ view of margin (lag) | 6–8 weeks | real-time |
The warehouse supervisors moved to depot operations. Neither role was redundancy; both had been doing clerical work the system had made necessary.
What this is not
It is not a new accounting package. It is not a “digital transformation” in the brochure sense. It is a decision to stop keeping four books and to keep one, implemented in software because that was the cheapest way to enforce the decision across every point in the firm where someone was writing something down.
The partners asked, at the end of the first month-end that took an afternoon, whether we had kept the old reconciliation reports in case they needed to be re-run. We said no, there is nothing to reconcile. They understood what we meant on the third month-end.