The measurement question usually shows up inside the budget approval, not at the end of the process
When a company puts budget behind an executive's coaching, the question arrives sooner or later: did this work, and how do we know. By the time it arrives, there is usually something in hand — but it is not a measurement. It is a satisfaction score. The executive liked the process, got on well with the coach, left each session a little lighter. None of that is wrong. None of it answers the question either.
Satisfaction is a byproduct, not a result A satisfaction survey is easy to collect, because it asks one person one question, and the answer always comes back, since nobody wants to admit they disliked how they spent their own time. Satisfaction is a record of how the process felt, not of what it changed, and the two records split more often than either side expects: an executive can leave every session pleased and return to the same table making the same calls.
That split usually reaches HR about a year later, when the budget comes up for renewal. What's on file is a year of high satisfaction scores, and the person sitting across the table is still making the same call the same way. The question in the room by then is no longer whether they were satisfied. It's what changed — and there is usually no record that answers it, because none was ever kept. Some companies notice the gap at the second budget cycle and simply cut it; not unfairly, since they never saw what they were paying for.
What coaching gives the company already shows which channel it moves through — not the room, but the decisions that leave it. This piece asks a different question: how that movement gets tracked, at what point, and what has to be seen before the work counts as real.
The process already carries its own measurement points; they are not added at the close
The process already carries its own measurement points; they are not a report tacked on at the end. They are three stops, planned from the start.
Three stops, three different questions - At intake: which decision, which habit, which reaction is meant to change — something concrete, something observable. - At the midpoint: where has that change already shown up, and where has it not yet. - At close: did anyone else notice the change, or did it stay inside the executive's own account of it.
Skip one of the three stops and the measurement goes with it. With only intake and close, the months between are a black box — the change is either there at the end or it isn't, with no record of when it started. With only intake, there is nothing yet but a goal; measurement hasn't begun at all. None of the three stops takes long — each is a half-hour conversation; the cost is not time, it is discipline.
How many sessions, over how many months, is a separate question, and another piece already carries it. What matters here is the stops themselves: each one asks for a different kind of evidence, and no single survey collects all three.
What gets measured is a decision, not an impression
What gets measured has to be concrete, not general The more concrete the goal named at intake, the easier the measurement at close. 'Communicate better' cannot be measured; 'listen before deciding in the team meeting' can, because the second one is a behavior — observable, and answerable with a plain yes or no on whether it happened.
The difference carries through the whole process. When a goal is reduced to a behavior, the midpoint question sharpens with it — what happened in that meeting last month, who noticed, what changed. When the goal stays general, the midpoint question stays general too, and a general question gets a general answer back.
Two executives can start at the same company in the same month; one names the goal as 'become a stronger leader,' the other as 'share promotion decisions with the team openly, with reasons attached.' Six months later, the second one's file holds three concrete examples — which promotion, what was said, how the team responded. The first one's file holds a single note: 'going well,' because nothing was ever defined to measure.
The pattern from the room usually breaks in the same place: a process that opens with a goal like "become a better leader" still cannot say, three months in, what it is measuring, because the goal was never converted into a behavior.
Satisfaction is a record of how the process felt. Measurement is a record of what it changed.
The real evidence sits outside the room, in behavior the room never sees
The real site of measurement is not the coaching room. What gets said there shows an intention; whether the intention held only shows up outside the room, with the team, in a meeting, at the moment a decision actually gets made.
The evidence comes from outside the room An executive can spend three months discussing listening with a coach and never once practice it, and the only person who notices is someone on the team — not the coach, not the executive either. That is why the closing review is not asked of the executive alone; where possible, it also goes, briefly and without naming names, to one or two people closest to the behavior in question.
How often that question gets asked matters too. Asking every month creates pressure and skews the answers; asking only at close can arrive too late, since six months spent drifting the wrong way leaves no time to correct course. The midpoint stop exists for exactly this — an early, gentle chance to adjust before the record is final. Who gets asked matters as well: the person closest to the executive carries the sharpest observation but also the most cautious answer, which is why the question stays short and unnamed.
This does not compromise the privacy of the room. It protects it, because what gets asked outside is not content but observed change. Not what was said, but what is different.
From the room
At the start of a process, naming the goal concretely is usually the longest part of the first session. An executive says "become more strategic." The question becomes "which meeting last month made you notice you weren't being strategic," and the goal doesn't sharpen until an answer comes back.
A sharpened goal brings its own measurement with it — because now there is something specific to look for, and the question to ask three months later is already written.
Resistance is its own kind of data Some executives resist this sharpening, because a concrete goal also carries a concrete chance of failing at it — 'become more strategic' can never quite be called a failure, but 'listen first in that meeting' either happened or it didn't. Some of these conversations take forty minutes; others take the whole session. A goal that sharpens quickly is usually one that was already thought through before the room.
A process that starts without that sharpening tends to close the same way: with general satisfaction. It may have gone well. Asked what actually changed, the answer still comes back as a feeling.