What came back
Six that came back as something else
The cheapest returns on the record were substitutions, at a median of thirty-five per cent.
The daily stand-upBack twice a week instead of five, standing and timed
week 0 · crept back after 3 weeks
Chasing unpaid invoices every weekBack in full
week 3 · crept back after 2 weeks
The spare key logReplaced by something else doing the same job
week 10 · crept back after 14 weeks
Two people on every delivery
week 4 · crept back after 9 weeks
Crept back 4
Four removals were written down with the week each of them stopped. Every one of them produced an event inside the 78 weeks watched. Of these, four crept back. Four of the eleven that crept back without anyone deciding.
Six of the twenty-four returns were not the thing coming back. They were a different thing doing the same job, at a median of thirty-five per cent of the old load — the cheapest group on the record.
The six
| what stopped | what appeared instead | share |
|---|---|---|
| the spare key log | a labelled board by the door | 20% |
| keeping the stock list on paper as well | a weekly export, printed once | 25% |
| the induction folder, kept current | a short list of who to ask | 30% |
| a written note after every phone order | the order system's own field | 40% |
| shadowing new starters for a week | a named person to ask for a month | 50% |
| logging why a job was turned down | a shared note of refused requests | 55% |
What a substitution is evidence of
In every one of the six, the removal revealed something the organisation had not distinguished before: the need and the mechanism were separate, and the mechanism had been standing in for the need for so long that nobody could see past it.
The spare key log is the clearest. The need was to know where the keys were. The mechanism was a book in a drawer, which was how this had been solved in about 1990 and had never been revisited. When the book stopped, the keys went missing, and somebody put a labelled board on the wall.
The board was always available. It was never proposed, because the book existed and the problem was therefore solved. A procedure occupies the space where its own alternatives would be considered.
You cannot look for a better way of doing something that is already being done. The removal is what creates the gap in which an alternative becomes visible.
The uncomfortable corollary
If removals produce substitutions, and substitutions are cheaper and sometimes better, then there is an argument that a shorter week is valuable as a forcing function regardless of the hours.
The record supports that in six cases out of thirty-eight. It does not support the stronger version, for two reasons.
First, a substitution only appears if the need survives the removal visibly. All six of these were noticed. The procedures whose needs were invisible did not produce substitutes; they produced problems, and two of the five problems were eventually fixed with substitutes only after costing something.
Second, the substitutions arrived at a median of fourteen weeks. They are not a design process. They are what somebody improvises when they find a gap, and improvisation under mild pressure produces the labelled board. It would not have produced anything at all if the gap had not been felt by a person who could act on it.
Two of the six were repairs
The shadowing and the refusal log are in this table and they are also two of the five problems. That is not a contradiction; it is the sequence.
The shadowing stopped in week six. Nothing visible happened for forty-seven weeks. Three people left. The organisation went looking, found the change, and put in a substitute: a named person for each new starter to ask, for a month.
That substitute costs half the original and may well be better — a named person for a month is more available than a week of sitting alongside. But it arrived eleven months and three departures after the need did, and the cheapness of the fix is not much consolation given when it arrived.
The same shape applies to the refusal log, at sixty-four weeks.
What separates a cheap substitution from a late one
Nothing about the substitution. In both groups what appeared was sensible, cheap and arguably better than what it replaced.
The difference is entirely upstream: whether anybody was standing in a place where the gap was visible, and how soon. The four fast substitutions were all for procedures whose absence inconvenienced somebody within a fortnight. The two slow ones were for procedures whose absence inconvenienced nobody at all until a third party arrived.
This is the same division that runs through everything on this site, and the substitutions make it unusually clear, because here the remedy is identical in both cases and only the timing differs. The gap was findable. The organisation found four of them in a fortnight and two of them in a year, and the difference cost three people.
Substitutes are not audited either
The six substitutions are the cheapest and probably the best outcomes on this record, and they share a weakness with everything else that happened quietly: nobody checked whether they actually cover the same ground.
The labelled board tracks where the keys are. The book had also tracked who had taken them and when, which nobody has needed in the window but which is the part that matters when a key does not come back.
The named person a new starter can ask covers questions the new starter knows to ask. A week sitting alongside somebody covers the things they do not know to ask about, which is most of what that week was for.
Both substitutions are cheaper. Both are narrower. Neither narrowing was noticed at the time, because a substitution arrives as a relief — the gap is filled, the matter closes.
The check is one question, asked a month after a substitute appears: what did the old one do that this one does not? On this record it was never asked, and in two of six cases there is a clear answer.