What came back
Twelve that came back smaller
The largest group of returns came back at a median of forty per cent of the old load. Each one is a small negotiation that the original procedure never had.
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.
Twelve of the twenty-four returns came back as reduced versions of themselves: fewer people, less often, shorter, or with a step removed. Their median weight was forty per cent of what the same thing had cost before.
This is the largest group on the record and the most ordinary, which is why it is usually missing from accounts of shorter weeks. Nothing dramatic happens in it. It is simply where most of the saving lives.
What reduction looked like
| removal | came back as | share |
|---|---|---|
| the daily stand-up | twice a week, standing, timed | 30% |
| site visits to the two furthest | four a year, not monthly | 30% |
| the standing slot for ownerless work | fortnightly, half an hour | 35% |
| the quarterly reconciliation | annual, with a named owner | 40% |
| month-end figures to everyone | to four people who use them | 40% |
| counting stock | quarterly, sampled | 40% |
| the Monday planning hour | twenty-five minutes | 45% |
| a named person on the phone to six | to five, then voicemail | 45% |
| the meeting before the meeting | fifteen minutes, when needed | 50% |
| the Friday clean-down | fortnightly | 60% |
| a second reviewer on routine changes | on changes above a threshold | 60% |
| two people on every delivery | on deliveries above a size | 70% |
Three shapes of reduction
Reading down that column, the reductions are of three kinds and they are not equally good.
Less often. Seven of the twelve. The thing is unchanged; it happens at a lower frequency. This is the easiest reduction to make and the easiest to get wrong, because frequency is usually set by the rate of the thing it guards against, and halving it halves the coverage in a way that is invisible until something slips through.
Smaller each time. Three. The stand-up timed at ten minutes, the planning hour at twenty-five minutes, the pre-meeting cut to what it was actually for. These are the cleanest: the same coverage, less ceremony.
Conditional. Two. The second reviewer above a threshold, two-person deliveries above a size. These are the most interesting, because they are not reductions of the procedure at all. They are the procedure with a rule attached that nobody had thought to write down while it was running unconditionally.
The conditional ones are the real find
Two entries out of thirty-eight, and they are the only ones where the organisation ended up with something better rather than something cheaper.
A second reviewer on every routine change is a rule that treats all changes alike. Nobody designed it that way; it simply started that way and was never revisited, because there was no occasion on which anybody had to justify it. The occasion was its removal.
When it came back it came back with a threshold, and the threshold is a piece of knowledge the organisation now has written down and did not have before. The same is true of the delivery rule.
The useful thing a removal produces is not the saving. It is the moment at which somebody has to say out loud what the procedure was for, which on this record had not happened for several of them in a decade.
What this group does to the arithmetic
Twelve entries at a median of forty per cent is, in hours, the largest single component of what the shorter week recovered — larger than the fourteen that never came back, because the twelve were on average bigger procedures to begin with.
It is also the component least likely to appear in a report. A procedure that came back does not read as a saving. It reads as a reversal, and the report says "twelve of the removals had to be reinstated", which is true and tells the reader almost nothing.
The sentence that carries the information is longer and has a number in it: twelve were reinstated at a median of forty per cent of their previous cost.
The caution
A reduction is a bet that the thing it guarded against happens less often than the old frequency implied, and on this record that bet has not yet been settled for most of the twelve.
Of the seven less-often reductions, the record shows no failure through week seventy-eight. That is reassuring and it is also a seventy-eight-week window on procedures whose events may be annual or rarer. A quarterly reconciliation moved to annual has been tested once.
The same observation applies here as everywhere else on this site: the absence of a problem in the window is not the absence of a problem, and the twelve reduced entries are the group with the longest remaining uncertainty attached to them.
Who decided the new size
In eleven of the twelve, the new size was set by the person who restarted the thing, in the course of restarting it, without anybody approving anything.
That is worth stating because the reductions are the largest saving on the record and nobody authorised any of them. The stand-up became twice weekly because the person who called it decided that was enough. The month-end figures went to four people because whoever sent them got tired of a long list.
The twelfth is the exception: the second reviewer threshold was discussed and written down, and it is the only one of the twelve that exists as a rule rather than as a habit.
The difference matters for durability rather than for the hours. Eleven of the twelve reductions are held in one person's practice and will revert, or drift further, or disappear when that person does. One is a rule.
An organisation that wanted to keep the saving would spend an hour converting the other eleven into rules, which nobody has done and nobody has any particular reason to do until the first one quietly goes back to being monthly.