The correction here is not "stop doing the bad thing". Standardising the process turned a termination instrument into a retention instrument, and then produced a second effect nobody was aiming at.
A formal improvement plan landed on someone who did not see it coming. They did not quit on the spot. They started looking that week, which in that particular case was survivable, because the decision had effectively been made.
But it kept happening, and the pattern is the thing worth writing down.
It only ever appears once someone has already decided. So the person reads it correctly, and reads it as a countdown rather than a chance.
We were communicative about performance. One-to-ones, clear expectations, regular conversations about what was and was not working. By any normal standard, nobody should have been surprised.
They were surprised every time. Even with quota missed for four or five months running, the reaction was that things had been broadly fine.
“We have talked about this repeatedly. They cannot claim not to know.”
“They are at sixty percent. That is not a secret.”
“I have been more than clear in our one-to-ones.”
A concern that lives only in conversation has not been raised, however many times you have said it.
And the reason is structural rather than a listening failure. Nothing in a one-to-one distinguishes a serious warning from ordinary coaching. The same room, the same tone, the same person, the same half hour. The signal has no weight to separate it from every other signal in the channel.
Which kills the defence every manager reaches for. You did tell them. It did not register, and "but I told them" was never going to be enough.
Four or five months at sixty percent stops registering as underperformance to the person living it. It becomes their normal, and from inside it does not feel like a slope.
Which is the same mechanism as a rep who hits every other month, seen from the other side of the desk. There, a year at seventy-two percent reads to the manager as a coachable near-miss. Here, months at sixty reads to the seller as fine.
Which is precisely why neither party raises it, and why it needs a line that exists outside both of their judgements.
One. The meeting stopped being an ambush. People came in knowing why, often named the problem themselves before anyone else did, and arrived with an informal plan of their own.
That is not politeness. A published threshold means the person has been watching themselves approach it. By the time the meeting happens they have already had the argument internally, so the manager is not delivering news any more. They are joining a conversation that has been running for weeks.
Two. It became a genuine reset rather than a countdown. There were people we wanted to keep who had a bad month or two. Many dug in, knew what had to change, and came out the other side.
The moment it is issued only when a manager has decided, being issued one is the decision, and every rational person starts looking.
Some still gave up and started looking anyway, and that belongs in the record. The standard does not save everyone. It stops the instrument from being a sentence.
Three, and nobody was aiming at this: people started asking for help. Before, they never did. Afterwards they were noticeably more thoughtful about asking well before they were anywhere near the threshold, and coachability improved across the team rather than just for the people in trouble.
You cannot ask for help against a standard you cannot see, and asking for help against an invisible one is just confessing.
This is the strongest argument in the whole corpus for published performance criteria, and it has nothing to do with fairness or legal defensibility. It converts everyone above the line from people managing their exposure into people managing their performance. A hidden bar makes asking for help dangerous. A published bar makes it strategic.
A monthly score that a seller can see is only half of it.
They also have to know the line, in advance, and be able to watch their own distance from it. A score without a visible line reproduces exactly the condition that made the old process an ambush.
Not their own score. The line. If nobody on your team can state the threshold from memory, then whatever you have been saying in one-to-ones has not been received, and the next formal conversation will land as a surprise no matter how many times you have had the informal one.
Run the model, free No account. Your numbers stay in the link.