The quiet ones
Why loud objections point the wrong way
The removals that generated the most argument were, without exception, cheap to get wrong.
The quarterly reconciliation of the project ledger
week 3 · surfaced after 31 weeks
Shadowing new starters through their first week
week 6 · surfaced after 47 weeks
Reading supplier contracts before renewal
week 2 · surfaced after 22 weeks
Logging why a job was turned down
week 9 · surfaced after 64 weeks
Surfaced 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 surfaced as a problem. The four removals nobody said a word about, all of which later surfaced as problems.
The nine removals that produced an argument on this record are, read back, a strikingly unthreatening list. A planning hour. Printed rotas. A second reviewer on routine changes. Site visits. The pre-meeting. Two people on deliveries. A monthly deck. Copying managers on orders. The away day.
Not one of them surfaced as a problem. Four were asked back quickly, two drifted back, three were never missed.
Meanwhile the four removals that cost the most produced no argument at all.
The uncomfortable reading
Stated baldly: on this record, the amount of argument a removal generated was inversely related to how much it mattered.
That sounds like an indictment of the people arguing, and it is not. Every one of the nine objections was reasonable on its face. The people making them were not being territorial or foolish. They were doing the thing that objecting requires, which is noticing that something is about to change and caring about it — and that is the clue to the whole pattern.
What makes a removal arguable
To object to a removal in a meeting, you need three things, and they have nothing to do with whether the removal is wise.
- You have to know it is happening. Which means it was proposed, which means somebody wrote it on a list, which means it had a name.
- You have to be able to say what it is for in about one sentence, because that is the amount of airtime available.
- You have to be affected in a way you can describe — otherwise the objection sounds like an abstraction and does not survive the room.
Those three conditions select hard, and they select for exactly the wrong thing. A procedure with a name, an obvious purpose and a visible daily effect on somebody is a procedure whose absence that somebody will notice next week. Which is to say: a procedure with a short lag.
The selection, stated as a mechanism
The conditions that let a removal be argued about are the same conditions that let its absence be noticed quickly. So the arguable removals are drawn, almost by construction, from the fast-reporting population.
This is why the pattern is not a coincidence of nine observations. It would reproduce in any organisation with meetings, because the filter is the meeting itself.
A procedure that belongs to nobody fails the third condition. One that was never proposed fails the first. One whose purpose takes a paragraph to explain fails the second. All three of those are also descriptions of a long lag.
What this does to the value of the argument
It does not make the arguments worthless. The nine argued removals produced four fast, useful corrections and three cases where the objector was simply wrong and the removal was right — and both of those outcomes are information arriving cheaply.
What it does is tell you what the argument is covering. The discussion in the room is a thorough examination of the subset of removals that would have told you about themselves anyway.
That is not nothing — the four corrections arrived inside a fortnight rather than a quarter, and two of them landed at a better size than either original position. It is simply not coverage of the risk.
The temptation to invert it
A natural response is to flip the rule: ignore the loud ones, examine the quiet ones.
Half right. The quiet ones do need examining and nobody is doing it. But the loud ones do not need ignoring, because attending to them costs almost nothing — they arrive, they are handled, they are over by week fourteen. The cost of the discussion is the discussion.
The actual error is one of accounting. Having spent an afternoon arguing about nine removals, an organisation feels that it has reviewed the transition. It has reviewed nine thirty-eighths of it, selected by a filter that chose the nine least likely to matter, and the sense of thoroughness is the problem rather than the time spent.
What to say at the end of the meeting
One sentence, after the argued items are settled:
"That is the list we discussed. Which of these did nobody mention?"
Read the unmentioned ones aloud. Not to debate them — there is nothing to debate, which is why they were not mentioned — but to produce a written list of entries about which the meeting generated no information at all.
On this record that list would have had seventeen names on it, four of which later cost money. Producing it takes two minutes and it is the only artefact of the meeting that points at the right population.
The removals that could not be argued about at all
Ten of the thirty-eight were not available for argument, for structural reasons rather than social ones.
Seven were never proposed — they stopped because the hours they lived in stopped existing, so there was no meeting at which an objection could have been made.
Three were proposed in rooms containing nobody who performed them, so the only people who could have described what the procedure did were not present.
Those ten contain four of the five problems.
This is the sharpest form of the selection described above, and it is not about the quality of anybody's argument. More than a quarter of this transition was never discussable, and the undiscussable quarter carried most of the risk.
An organisation can fix part of this cheaply — the three absent-team cases are caught by one question — and cannot fix the seven at all, except by looking for them afterwards.