The quiet ones
An objection is an appointment
The value of an objection is not that it is right. On this record objections were wrong two thirds of the time and were still the most useful thing in the room.
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.
Six of the nine removals that somebody argued against went ahead and turned out fine. Three of those nine are among the fourteen things nobody has asked for in seventy-eight weeks.
By the only measure that most people apply to an objection — was it correct? — the objectors on this record did badly. They were right about some of the detail, wrong about the substance two thirds of the time, and the transition would have been smaller and worse if they had been deferred to.
And every one of the nine removals they argued about reported back inside twelve weeks, which is better than any other group on the record.
The two things an objection does
It makes a claim. This thing matters, and removing it will cost something. The claim may be right or wrong, and on this record it was usually wrong.
It creates a watcher. Somebody is now attending to what happens next, has reason to notice, and has standing to say so.
The second is almost entirely independent of the first. A person who objects for bad reasons watches just as carefully as one who objects for good ones. Their attention is not calibrated by the quality of their argument; it is created by the act of having argued.
This is why objections are useful even when they are wrong, and it is the opposite of how they are usually treated. The question in the room is always "is this person right?" — and when the answer is no, the objection is set aside and so, invisibly, is the watcher it would have created.
What this looks like on the record
| argued removals | n | what happened |
|---|---|---|
| asked back | 4 | median 2 weeks |
| crept back | 2 | 6 and 9 weeks |
| never missed | 3 | nothing, 78 weeks |
| surfaced as a problem | 0 | — |
Four of the nine came back by request, fast. In two of the four the reinstated version was substantially smaller than the original — which is to say the objector was half right, the removal was half right, and the organisation found the midpoint inside a month because somebody was paying attention.
That is the best outcome available anywhere on this record. It is better than a removal that works, because it is cheaper than the original and better fitted than either position argued in the room.
The failure mode this suggests
Not "listen to objections more." On this record, deferring to objections would have prevented three good removals and produced a worse transition.
The failure mode is quieter: an objection that is overruled and then forgotten.
When somebody argues and loses, two things are created and only one is used. The claim is dealt with — considered, answered, set aside. The watcher is created and then nobody tells them they are a watcher, nobody asks them anything afterwards, and in most organisations the social cost of returning to a settled argument means they do not volunteer.
On this record they did volunteer, four times out of nine, inside a fortnight. That is a well-functioning organisation and it should not be assumed. In a place where losing an argument is remembered, those four requests become four silences, and nine fast reports become nine slow ones.
Two sentences that cost nothing
At the moment an objection is overruled:
"You may well be right. If you are, what will we see, and when?"
And then, at that time, asking them.
This takes the thing that objections are actually good for and makes it explicit rather than relying on it being volunteered. It also has a side effect worth having: it converts an argument about whether somebody is right now into an agreement about what would show it later, which is a considerably easier conversation to be on the losing side of.
Where it does not reach
Seventeen removals on this record had no objector at all, and no amount of good practice around objections helps there. There is nobody to ask.
Those seventeen need a different mechanism entirely, which is the subject of the rest of this part of the site. The point here is narrower: of the thirty-eight removals, nine arrived with a free monitoring system attached, and the organisation used it without noticing that it had it.
The objections that were right
Three of the nine objections were vindicated, in the sense that the thing came back at or near full weight after a short interval.
All three had the same structure, and it is worth naming because it is the only thing on this record that distinguishes a good objection from a bad one at the time it is made.
Each of the three was specific about a mechanism: not this matters but this is what it does, and here is the case it catches. The printed rotas were defended by somebody describing exactly who picked one up and when. The shift handover was defended with an account of a particular kind of job that spans a shift boundary.
The six that were not vindicated were general. They asserted importance without describing a case.
This is not a rule for deciding which objections to accept — two of the six general ones were about things that later needed fixing anyway. It is a rule for what to ask when an objection is made: what is the case this catches? An objector who can answer has told you where to look later, which is more valuable than the objection.